OFFENBACH 2012 POSTPRINTS of the 10Th Interim Meeting

OFFENBACH OFFENBACH ICOM

OFFENBACH

CC LEATHER & RELATED & LEATHER CC

August 29 August 2012

- 31,

2012

ICOM-CC LEATHER

& RELATED MATERIALS

WORKING GROUP

POSTPRINTS GR WORKING MATERIALS

of the 10th Interim Meeting

OUP

EDITED BY CÉLINE BONNOT-DICONNE, CAROLE DIGNARD AND JUTTA GÖPFRICH

ICOM-CC Leather and Related Materials WG Offenbach 2012

POSTPRINTS of the 10th Interim Meeting of the ICOM-CC Leather & Related Materials Working Group

Offenbach, Germany 2012

ACTES de la 10ème Réunion Intermédiaire du Groupe de Travail Cuir et Matériaux Associés de l’ICOM-CC

Offenbach, Allemagne 2012

Edited by/ Sous la direction de Céline Bonnot-Diconne (2CRC), Carole Dignard (Canadian Conservation Institute / Institut Canadien de Conservation), Jutta Göpfrich (Deutsches Ledermuseum Schuhmuseum).

ICOM-CC Leather and Related Materials WG Offenbach 2012

Organizing committee / Comité d’organisation

Villa Médicis, Rome, Italy and 2CRC, Moirans, Céline Bonnot-Diconne France Carole Dignard ICC/CCI, Ottawa, Canada Nina Frankenhauser Offenbach, Germany Jutta Göpfrich Offenbach, Germany

Editorial Committee / Comité de lecture

Céline Bonnot-Diconne 2CRC, Moirans, France Carole Dignard ICC/CCI, Ottawa, Canada Theo Sturge Northampton, United-Kingdom

Acknowledgments / Remerciements - ICOM Germany, Dr. Klaus Weschenfelder (President) - DLM- Deutsches Ledermuseum/ Schuhmuseum Offenbach - Académie de France à Rome, Villa Médicis (Eric de Chassey, Annick Lemoine). - Canadian Conservation Institute/Institut canadien de conservation

These Postprints have been supported by Hermès. Cette publication a bénéficié du soutien de la Maison Hermès.

To order this book, please contact:

DLM- Deutsches Ledermuseum/ Schuhmuseum Offenbach Frankfurter Strasse 86 D-63067 Offenbach, Germany T. 0049 (69) 829798- 0 [email protected] www.ledermuseum.de

ICOM-CC Leather and Related Materials WG Offenbach 2012

Introduction Dear members, dear colleagues,

I would like to express to you all a warm welcome to Offenbach and to this conference! It is a pleasure to see so many of you here present today, to enjoy and participate in this anniversary conference – the 10th Interim Meeting of our Working Group.

Offenbach is indeed a welcome return for us: twenty- three years ago, in 1989, the ICOM-Committee for Conservation’s Working Group on Leather and Related Materials met in Offenbach for our Second Interim Conference. Already at that time, our Working Group had recognized Offenbach and its Leather and Shoe Museum as a centre of excellence in our field, as it remains so to this day. Dr Rathke, I cannot thank you enough for your institution’s kind and generous offer of hosting this conference, and having us once more benefit from your institution’s warmth, hospitality and wealth of knowledge. I know how much time Jutta and Nina devoted to organizing this conference and making sure every detail was in place for it to run smoothly, and I would like to thank both of them from the bottom of our hearts for their remarkable work. And as well, I would like to express my warmest thanks to the whole team of colleagues from the Museum who also worked diligently behind-the- scenes to make these next two days of conference, a most pleasurable and rewarding experience for us. And, last but not the least, thank you to all speakers and contributors on the program: we are looking forward having you share your knowledge and expertise with us, and we are confident that you will achieve this fruitful exchange while respecting the schedule and each speaker’s allocated time.

I would say that a hallmark of this conference is: its diversity. My Assistant- Coordinators, Jutta Göpfrich, Carole Dignard and myself, decided to forego having a unique or specific theme for this conference. Rather, we opted to open up the focus to all fields of current interest to our Working Group membership, thus favoring and at the same time, bringing to the forefront our members’ most recent, innovative work, research or achievements. And so, we look forward to this conference’s broad range of topics: from archaeological leather to parchment, and from ethnographic skins to upholstery leather and gilt leather wall hangings.

The program for these next two days is therefore rich and diverse. The common thread of course is the material, leather, for which we all share a passion and a fascination in its beauty, and its perennial yet perishable nature. Personally it saddens me that still many museums are slow in giving prominence to leather heritage objects as they deserve to be recognized, as artworks or masterpieces of craftsmanship, or valuable testimonials of history, peoples and cultures. How many countries have a dedicated Leather Museum? Too few, and this omission perpetuates the oversight of having still too few leather heritage objects displayed to, and enjoyed by, the public.

ICOM-CC Leather and Related Materials WG Offenbach 2012

Our Working Group contributes to making leather heritage objects known, recognized and valued, and I hope that here, with this Interim Conference in Offenbach and in its Leather and Shoe-Museum venue, we will find a wealth of information exchange opportunities which may not present themselves in each of our home countries. I thank all of you present in advance for your openness in sharing your insights and experience with others, and for your active participation in this conference’s discussions.

And so, it is a pleasure for me now to let you start these two days of talks, which I am sure will be rewarding for each one of us.

Céline Bonnot-Diconne Coordinator, ICOM-CC Working Group on Leather and Related Materials

======

Chers membres, chers collègues, Bonjour à toutes et à tous,

Je vous souhaite la bienvenue à Offenbach et me réjouis de vous voir si nombreux réunis aujourd’hui pour notre 10ème réunion intermédiaire.

Voilà 23 ans, en 1989, que se réunissait pour la seconde fois le groupe de travail Cuir de l’ICOM-Comité pour la Conservation. Déjà, il avait choisi de venir à Offenbach, marquant par là l’importance que peut avoir cette institution dans notre domaine d’activité. Dr Rathke, je vous remercie infiniment d’avoir accepté d’accueillir à nouveau notre groupe. Je sais le temps que Jutta et Nina ont consacré à son organisation et je tiens à les remercier chaleureusement pour leur remarquable travail, ainsi que l’ensemble de l’équipe de votre musée. Mes remerciements vont également aux auteurs qui ont proposé leur contribution et qui, je n’en doute pas, sauront nous transmettre leur connaissance tout en respectant le temps de parole qui leur a été réservé.

Cette rencontre est marquée par le sceau de la diversité. Mes assistantes Jutta Göpfrich, Carole Dignard et moi-même, n’avions pas souhaité privilégier de thème pour cette conférence afin que les auteurs se sentent libres de proposer des articles dans tous les domaines que couvre notre groupe de travail et aussi afin de rester au plus près de l’actualité qui est la leur. Les communications évoqueront donc les cuirs archéologiques, comme les parchemins, les objets ethnographiques, le mobilier et les cuirs dorés.

Notre programme pour les deux journées qui viennent est donc riche et divers. Il nous réunit autour d’un même matériau dont je regrette personnellement qu’il soit aujourd’hui encore trop délaissé dans les collections de musées. Combien de pays comptent un musée du cuir ? Trop peu encore et cette inexistence cache encore trop d’objets et d’œuvres à la vue du public. 7

ICOM-CC Leather and Related Materials WG Offenbach 2012

Notre groupe de travail contribue à les faire connaître et j’espère que vous trouverez ici les possibilités d’échanges qui n’existent souvent pas à l’échelle de nos pays respectifs. Je vous remercie d’avance de votre collaboration, en espérant votre participation active aux discussions. Et je vous invite maintenant à commencer ces journées que j’espère enrichissantes pour chacun. Céline Bonnot-Diconne Coordinatrice du groupe « Cuir et Matériaux Associés »

ICOM-CC Leather and Related Materials WG Offenbach 2012

Welcoming Address

Members, colleagues!

Leather and orchids have something in common. When referring to particularly rare and exotic university courses in German, in common parlance we often speak of so- called “orchid subjects”. Likewise, leather is anything but a common material in Conservation Science. Therefore we are particularly happy that the DLM - The German Leather Museum /German Shoe Museum in Offenbach has been chosen as the venue of the ICOM-CC Working Group on Leather and Related Materials for the second time in 23 years now. What may have served as an incentive for many colleagues, was the possibility to visit the collection –reopened in early 2012 following extensive renovation – for the first time, in particular the new section on treasury art from the 14th to the 17th century. At the same time, the collection of leather objects, unique in terms of its scope and diversity, offered enough material for comparison and focussed discussion. Furthermore, the DLM had prepared an exhibition on the restoration of leather objects, which was complemented by a bi-lingual exhibition catalogue.

Present restoration and research projects from the fields of ethnology, archaeology, arts and crafts and natural sciences took centre stage during the three-day professional exchange. As our guest of honour, Dr. Klaus Weschenfelder, the president of ICOM Germany, opened the meeting at the official evening reception. ICOM Germany had also been kind enough to provide financial support for the meeting. The supporting programme was intended to reflect the cultural spectrum of the ICOM CC Leather Working Group and as well as the collection of the German Leather Museum by means of the musical performances of the Mongolian band Egschiglen and the Bayern- und Gebirgstrachtenverein (Society for Bavaria and for traditional Alpine costumes).

On the third day, we undertook an excursion to Vollrads Castle and to Eberbach Monastery. Two different restoration projects were presented at Vollrads Castle. In addition to the talks on gilt leather from the day before, the restorers commented on the restoration measures carried out in situ on a gilt-leather wall covering dating from the 17th century. A rare tournament racing saddle (around1500) with a painted parchment cover in different colours was used to exemplify restoration problems commonly occurring in dealing with material combinations.

We are happy to be able to present to you the conference proceedings in printed form now. This is not least due to the generous contribution of a donor.

Jutta Göpfrich Assistant-Coordinator, ICOM-CC Working Group on Leather and Related Materials DLM- German Leather Museum/ Shoe Museum Offenbach

======

ICOM-CC Leather and Related Materials WG Offenbach 2012

Chers membres, chers collègues!

Le cuir et les orchidées ont quelque chose en commun. Ainsi, en allemand, dans le langage courant, quand on se réfère à des filières d’études particulièrement rares et exotiques, on parle de « sujets d’orchidées ». Le cuir, aussi, est un matériel marginal dans le domaine de la conservation. C’est pourquoi nous nous étions très heureux de voir le DLM - Musée Allemand du Cuir/ Musée de la Chaussure à Offenbach choisi pour la seconde fois en 23 ans comme le lieu du Symposium du groupe de travail « Cuir et matériaux associés » de l’ICOM-CC. La possibilité de pouvoir visiter pour la première fois la nouvelle section muséale consacrée aux trésors d’église du 14ème au 17ème siècle rouverte au début de 2012 après rénovation, a certainement motivé beaucoup de collègues à venir à Offenbach. Simultanément, la collection des objets de cuir – unique tant en termes de volume que de diversité – permettait de faire des comparaisons et d’avoir des discussions pointues. De plus, le DLM avait monté une exposition sur la restauration des objets de cuir accompagné d’un catalogue bilingue.

Le programme de la conférence, sur trois jours, avait pour sujets des travaux de restauration et de recherche dans les domaines de l’ethnologie, l’archéologie, l’artisanat d’art et les sciences naturelles. Comme invité d’honneur, le Dr Klaus Weschenfelder, président de I’COM-Allemagne, a ouvert la réunion lors de la soirée de gala. ICOM-Allemagne nous a généreusement aidés en apportant son soutien financier au congrès. Les événements annexes visaient à refléter l’éventail culturel du groupe Cuir de l’ICOM-CC et de la Collection du Musée Allemand de Cuir, avec les performances musicales du groupe mongol connu sous le nom de Egschiglen et celle du Bayern- und Gebirgstrachtenvereins (Société pour la Bavière et pour les costumes traditionnels de la zone alpine).

Le troisième jour, les participants ont fait une excursion au Château de Vollrads et au Monastère d’Eberbach. Au Château de Vollrads, deux projets de restauration différents ont été présentés. En complément aux exposés de la veille sur le cuir doré, des restaurateurs ont expliqué les mesures de restauration, exécutées in situ sur un revêtement mural de cuir doré daté du 17ème siècle. Prenant comme exemple une selle de tournoi très rare (datant des environs de 1500) recouverte de parchemin peint de diverses couleurs, on a pu discuter des problèmes de la restauration des matériaux composites.

Grâce à la générosité d’un donneur, nous sommes enchantés de pouvoir vous présenter les actes de la conférence sous forme imprimée.

Jutta Göpfrich Coordinatrice-Adjointe, ICOM-CC Groupe de Travail sur le Cuir et les Matériaux Associés DLM - Musée Allemand de Cuir / Musée de la Chaussure d’Offenbach

ICOM-CC Leather and Related Materials WG Offenbach 2012

Contents / Table des matières

Introduction...... 6

CÉLINE BONNOT-DICONNE ...... 7

Welcoming Address ...... 9

JUTTA GÖPFRICH ...... 9

Method of Use and Importance of Gilt Leather in Three Italian Palaces during the 16th and the 17th Century through the Study of Inventories ...... 15

CÉLINE BONNOT-DICONNE

A Suit of Lamellar Armour ...... 23

NINA FRANKENHAUSER

The Use of Correlation Methods to Detect Surface Deformation and Changes in the Leather Wall-Coverings in Jever Castle Museum ...... 29

ANGELIKA GERVAIS, PETER KÖNIGFELD

Treatment and Mounting of Medieval Parchment Fragments ...... 36

BARBARA HASSEL

The Restoration of the 17th Century Altar Frontal from the Oratory of Saint Dominic in Orvieto: Backing Reinforcement, Lining and Tensioning System According to Minimal Intervention ...... 43

SARA IAFRATE, ANNA VALERIA JERVIS, MARIABIANCA PARIS, MARCELLA IOELE, LAURA D’AGOSTINO

The Study and Conservation of a 12th Century Pair of Ceremonial Sandals with Arabesque Decoration from Castel Sant’Elia’s Collection of Liturgical Garments ...... 51

ANNA VALERIA JERVIS, SILVIA CHECCHI, ANTONELLA DI GIOVANNI, STEFANO FERRARI, MARIA RITA GIULIANI, MARCELLA IOELE, MICHAEL JUNG, MARICA MERCALLI, FEDERICA MORETTI

Transformation of Collagen into Gelatine in Historical Leather and Parchment Caused by Natural Deterioration and Moist Treatment ...... 61

RENÉ LARSEN, DORTE VESTERGAARD POULSEN SOMMER, KATHLEEN MÜHLEN AXELSSON, STEEN KRISTIAN FRANK

ICOM-CC Leather and Related Materials WG Offenbach 2012

Research on New Working Methods with Leather in Historical Artworks: Restoration Work of a Frontal Altar in Gilt and Painted Leather from Central Sicily ...... 69

LOREDANA MANNINA, ANGELA LOMBARDO

The Leather Drying Trial and Associated Analytical Work ...... 77

ANGELA MIDDLETON, KARLA GRAHAM, PAUL GARSIDE

Morphological Assessment of Shrunken Parchment Fibres at Microscopic Level ...... 90

KATHLEEN MÜHLEN AXELSSON, RENÉ LARSEN, DORTE VESTERGAARD POULSEN SOMMER, STEEN KRISTIAN FRANK

The Leather Furnishings in Palazzo Chigi in Ariccia: Documentary Sources .... 96

MARA NIMMO, MARIABIANCA PARIS, FRANCESCO PETRUCCI

Skin Clothing from the North: Research, Documentation and Preventive Conservation ...... 107

ANNE LISBETH SCHMIDT

Some Reflections on the Past Treatment of Gilt Leather, and Recent Remedial Work Where This Has Failed ...... 115

THEO STURGE

The Role of Leather Science and Technology in Heritage Conservation ...... 122

ROY THOMSON

Lubrication of Ancient Leather – Imperative or Impossibility? ...... 128

BERNHARD TROMMER, ANDREAS SCHULZE, HEINRICH FRANCKE

Reception of an 18th century Gilt Leather in the Historicism: Fortunate Exceptions or Characteristics Examples? ...... 137

WIVINE WAILLIEZ, ELOY KOLDEWEIJ

Leather and the Medieval Christian People of the Fourth Cataract Region of the Nile: Two Case Studies ...... 150

BARBARA WILLS

The Conservation of Archaeological Wet Leather at Schloss Gottorf During The Last 65 Years ...... 157

GABRIELE MARIA ZINK

ICOM-CC Leather and Related Materials WG Offenbach 2012

POSTER - Charisma Projects on Gilt Leather ...... 164

CÉLINE BONNOT-DICONNE, MARCELLA IOELE, CLAIRE PACHECO, MARIABIANCA PARIS, LAURIANNE ROBINET

POSTER - Les Collections Ethnographiques, Spécificités dans la Conservation ...... 165

MARINA REGNI

Participants List ...... 166

ICOM-CC Leather and Related Materials WG Offenbach 2012

Method of Use and Importance of Gilt Leather in Three Italian Palaces during the 16th and the 17th Century through the Study of Inventories

Céline Bonnot-Diconne

Abstract “Corame d’oro”, “cuoio dorato”, “pelle di Spagna”... gilt leather appears very often in the inventories of Italian palaces. By reading these archives, it is possible to deduce their use and their importance, even when they have all disappeared. The talk will focus on three of the most famous and richest palaces in Roma: Palazzo Ricci-Sacchetti, Villa Medici and Palazzo Farnese, dated between 1561 and 1644. Wall-hangings, door curtains, carpets and tablecloths of gilt leather were widely used and their disappearance is a great loss for the aesthetics and comprehension of the rooms’ decors and atmosphere.

Keywords Gilt leather, Spanish leather, pelle di Spagna, corami d’oro, Villa Medici, Ricci, Farnese, Italian, Rome, palace, inventory, archives, leather, wall-hanging.

Introduction

Sometimes the existence of interior decorations is only known through archival or historic records ; this is often the case for gilt leather wall-hangings. Although quite fashionable during the 16th century, they are seldom still present in situ. However, archives and inventories of Roman palaces provide information on their former ubiquity and use.

Dating inventories presents a complex maze of challenges that only archivists-paleographers are able to decipher. Except for the first entry date, an inventory is often a series of detailed checklists and a newer list often uses again the same terms used in a preceding list. The work involved in editing of such handwritten documents is long, tiresome and repetitive (Figure 1). The scribe does not always measure up, and errors and omissions are inevitable. The number of copies transcribed on various dates can also lead to hazardous interpretations. In Italy, an inventario is not necessarily made following the owner’s death. But the appointment of a new warehouse master meant that records relating to the goods of the palace would thus be updated. So, these residences, with all that they contained of useful pieces of furniture and valuables, take shape and are brought back to life along the pages of their inventories. We decided to concentrate on three prestigious buildings of Roma for which records are still being kept and which are also tied to key figures who built and decorated them.

Figure 1: Excerpt from an inventory in Villa Medici made on the 22d of June 1598 (Archivio di Stato di Firenze, Miscellanea Medicea 363 ins.II). Underlined: ‘Un paramento di còrame di Spagna d’oro e d’argento dipinto à grotesche ed pilastri darpiè et archi alto sei pelle ed dua sopra porte e una soprà finestra che para tutta detta stanza’. Author’s translation: “A gold and silver Spanish leather wall-hanging, painted with grotesque and pilasters of Harpies and arches, six skins in height and two over doors and one over windows which adorn all the room.”

ICOM-CC Leather and Related Materials WG Offenbach 2012

PALAZZO RICCI-SACHETTI

The Palazzo Ricci-Sacchetti owes its origins to cardinal Ricci di Montepulciano who was an exceptional figurehead of his time: a great traveler but also a great builder. Born in 1498, his career leads him, for the Church, in all of Europe and in particular, in the Iberian Peninsula where he stayed for many years (1545- 1550) (Deswarte-Rosa 2010). A passionate collector, cardinal Ricci played a significant role in the propagation of the “exotic” taste in Roma. The palazzo Ricci-Sacchetti, located in Via Giulia, became his true residence in Roma and the means of showcasing his success (Figures 2 and 3).

Figure 2: Localization of the three palaces in Roma.

It marks a stage in the history of the Roman cardinal palace, with its two courtrooms, one of winter and the other of summer, its series of rooms richly decorated with frescoes and tapestries, its cabinet of curiosities and its gallery, the whole abundantly decorated with antiques. Prestigious painters worked there like Salviati who, in 1553, painted the Stories of David frescoes for the living-room of the Globes. Begun in 1552, construction was completed in 1557. Having just overcome various financial difficulties, Ricci considered next, at that time, how to furnish his palace. He ordered a whole series of gilt leathers to decorate the walls, under the frescoes, following a well-established system in Roma. He decided to take on the services of a Spanish craftsman of origin, Giovanni Pietro da Corduba.

Figure 3: Plan of the Palazzo Ricci-Sachetti. In Schütze S., Colle E., 2003. Palazzo Sacchetti.

ICOM-CC Leather and Related Materials WG Offenbach 2012

The inventory of June 15611 gives a relatively complete idea of the palace’s furnishings. Nine rooms were decorated with gilt leather and for each one, it is specified the number of parts (pezzi) constituting the decoration. For each pezzo, the number of squares is specified. The rooms included between four and eight pezzi, of which the largest one counted fifty squares (or skins). Based on the standard size of a square, it is possible to estimate the size of the leather surfaces employed in the interior decoration: more or less 322 square meters.

We do not know what was represented on these leathers. The lists distinguish them by their color and the presence or not of friezes and columns. The gilt and silvered backgrounds are often enhanced with red and blue, which could refer to the crimson cardinal color and the blue of the sky of the Ricci coat of arms. It was usual in the 16th century to reproduce the owner’s armorials in blazons on leather door-curtains. And indeed, in the inventory, “30 cardinal insignia made of gilt leather among which some of the bishop” are mentioned.

The payment records that were kept state clearly that the auripellario (gilt leather maker) who manufactured them lived in Roma. His name is evocative: Giovan Pietro di Cordoba2.

The maestro was commissioned in 1557 for the sum of 69.5 scudi. Eighteen months later, in 1559, he received a first installment of 20 scudi. The balance is settled in 1560, that is to say two years and eight months later. We do not know if this contract related to the whole of the palace’s interior decorations. But the inventory of 1561 shows clearly the presence of leather in nine rooms.

Research carried out at the Archivio di Stato (State Archive) in Roma, made it possible to track down Giovan Pietro di Cordoba. Desiderius Bonaveva was his appointed notary at the Tribunale dell' Auditore Camerae. His archives enable us to follow the activity of the craftsman between 1554 and 1567. At first scarce, his contracts multiply in numbers and grow exponentially after 1560, which is the delivery year of cardinal Ricci’s leathers. In the following years, to fulfill his obligations, Pietro is constrained to pass agreements with other auripellarii (Jacob de Fulviis, Bartholomeus Rusconus, etc.)3.

The Palazzo Ricci had also other objects made of gilt leather (Table 1). The inventory of 1561 attests to other hangings in storage as well as to 14 door curtains which consist of 4 to 7 skins to which are added friezes. Listed as well are 13 leather tablecloths. The majority is red or blue but one is green. There are 16 cushions, including 13 “to put oneself at knees”.

Table 1: Objects made of gilt leatherat Palazzo Ricci in 1561.

ICOM-CC Leather and Related Materials WG Offenbach 2012

VILLA MEDICI

Giovanni Ricci di Montepulciano is also at the origin of another extraordinary building: the Villa Medici, located on the Pincio Hill (Figure 2). After purchasing it in 1564, he asks the Florentine architect Nanni di Baccio to modify the building. When he dies in 1574, the construction is not finished yet. But two years later, Ferdinando de Medici, son of Cosimo the first de' Medici, acquires it and asks the Florentine architect Ammannati to build a palace worthy of the prestige of the Medici family.

Born in 1549, cardinal at the age of 13, collector and sponsor, devoted to Antiquity, Ferdinando conceived his Villa as a museum. He added a gallery where he presented his collection of antique masterpieces (Figure 4). He inserted in the facade a series of antique bas-reliefs. At the time of the sale of the Villa in 1576, Ricci’s heirs asked for a sum of 1000 additional scudi for furnishings and paintings, among which “Pelle di Spagna numero mille in circa” (close to a thousand Spanish skins) in three rooms 4. It is estimated that they covered up to 250 square meters of walls.

Figure 4: Villa Medici today. Photo C.Bonnot-Diconne 2012.

When he was alive, the cardinal Ricci di Montepulciano was the friend of Ferdinando as well as his protector in Roma. During the summer 1569, Ricci makes his palace of Via Giulia available to Ferdinando. Ferdinando was able to observe there all the leather hangings. But he already knew very well this kind of decoration because the Medici family always highly appreciated gilt leather. In Firenze, they are mentioned in the inventories of the second half of the 16th century under Cosimo the first and Eleonora di Toledo.

At Villa Medici, for the interior decoration, Ferdinando called upon a Florentine artist, the painter Jacopo Zucchi. Decorations to the cardinal's rooms started in 1584, with a complex cosmologic iconography (Figure 5). Looking onto the gardens, the three rooms in a row form the cardinal’s apartment, and are richly adorned by the mannerist master and his studio. The first one is the Elements' room, the second one the Muses' room and the last one the room of Jupiter's loves. The coffered ceilings are typically Florentine, decorated with painted squared or hexagonal canvases. They are surrounded by friezes depicting stories of gods and goddesses meant to represent Ferdinando de Medici's horoscope. Figure 5: Room of the Elements at Villa Medici. Photo C.Bonnot-Diconne 2012.

Four years later, the inventory of 1588 indicates that the three rooms had a decoration rather similar of corame di Spagna with grotesque, pilasters, on gold and silver backgrounds. The wall-hangings of the Muses’ room and the one of Jupiter’s Loves comprised harpies, devastation divinities with a bird-winged body and a woman’s head 5.

ICOM-CC Leather and Related Materials WG Offenbach 2012

In another apartment (the Trinity’s one), two other rooms held gilt leather wall-hangings. These leathers were gilt, with silver and black backgrounds. These could be the hangings left by Ricci after his death. The Grand Salon, the most monumental room at Villa Medici, was also decorated with gilt leather. The inventory of March 1588 mentions a decoration with gold and silver backgrounds, with columns, friezes at top and bottom, and 22 arches.

In the rooms of the Cardinal, a restoration work on the frescoes has recently brought forth the discovery of nails, of which some had leather remnants on them. This reveals that the hangings were attached immediately under the frescoes friezes probably in a fixed way, as attested by the nails. In the rooms of the cardinal, the inventory states that the squares were 6 lengthwise, which, on 3.30 meters high, corresponds to a square of 55 cm.

The inventory of the Villa in 1588 mentions seats upholstered with leather in very significant numbers. The upholstery did not seem, however, to reproduce the design of the wall leathers, as was the practice in use the following century in France and Germany. Among the other objects, the inventory lists pavimenti di corame d’oro (‘gilt leather flooring’). These carpets are bedside rugs. In addition, the sopropanni (tablecloths) are rather systematically made of painted gilt leather, sometimes lined with silk. Twelve are in the rooms and there are four in storage (Table 2). Table 2: Objects made of gilt leather at Villa Medici in 1588.

In 1587, Ferdinando de' Medici was called to Firenze to replaces his brother Francesco on the throne of Tuscany. He left the interior decoration of the Villa partly unfinished. The most precious statues and part of the collections were moved to Firenze. Thanks to an exchange of letters between Ferdinando and his ambassador in Roma, we know that six months after his departure for Tuscany, Giovanni Niccolini packs up i costosi paramenti spagnoli di corame (the expensive Spanish leather wall-hangings) to send them by sea shipment to Firenze 6. Niccolini answers to the personal request of the Grand Duke. The Leathers shall be used and enjoyed again at Firenze’s Palazzo Pitti. The speed with which these leathers are brought back to Firenze shows how important these decorations were to Ferdinando.

PALAZZO FARNESE

Palazzo Farnese, today the French Embassy in Italy, is due to another cardinal, contemporary of Ricci and Medici: Alessandro Farnese. In 1495 Farnese buys a palace, near Campo dei Fiori and begins in 1517 the construction of a new building. In 1534, when cardinal Alessandro Farnese becomes Pope known as Paolo III, the construction is moving along well. In 1546, after the death of the architect Antonio da Sangallo, Michelangelo resumes the management of the construction site and the second floor is completed. Paolo III, who dies in 1549, will not see his finished project which ends only in 1589. The construction will have lasted about 75 years (Figure 6).

The palace is then inhabited by the great-grandson of Paul III, cardinal Edouardo Farnese. He is the one who calls the Carracci, painters from Bologna, to complete the palace’s interior decoration which had been started by Daniel da Voltera, Francesco Salviati and Taddeo Zuccari, the painters of the frescos of the Salotto Dipinto (painted living room).

ICOM-CC Leather and Related Materials WG Offenbach 2012

Figure 6: Palazzo Farnese by Giuseppe Vasi (1710-1782).

Before 1612, we unfortunately did not find any traces of commissions for gilt leather in the accounts of the palace. The oldest inventory found is unfortunately rather late, dating of 1644. It references nearly 7 320 objects of all kinds, among which many are made of gilt leather.

This inventory is interesting because it mentions at times the concepts of oldness: leathers are described as “antiques”, “old” or even “very old”. Only for hangings, more than 3 600 squares are mentioned in total (Table 3). Eight rooms are decorated with leathers, but there are also 1 200 squares of other hangings waiting in storage. The patterns are not precisely described but some details are mentioned such as: lily, gilt and silvered backgrounds, with a prevalence of the colors blue, red and crimson, sometimes associated between them. Two wall coverings in place are mentioned as being worked con borra, which shows the existence of flocked leathers at this date. At Palazzo Farnese, the leather rooms are essentially located on the ground floor, in the most beautiful apartments. Table 3: Estimate of the number of gilt leather squares at Palazzo Farnese in 1644.

But on the main floor, the Sala degli Imperatori (the Emperors’ Room) opens the series of rooms of great magnificence: twelve Roman busts and also twelve paintings by Annibal Carrache were presented ahead of the walls that were covered with gilt leather with unicorns and lilies, the Farnese’s emblems. This room is the first part of the Farnese residence where the decoration was carried out. According to the plans of Antonio da Sangallo, it has a sumptuous wooden carved ceiling, with a frieze of frescoes and cornices in stucco. Daniele da Volterra (1509-1566) is behind the decoration on the topic of Bacchus and the unicorn, probably carried out in 1547. So, gilt leather could possibly go back to this period. By their representation, the leathers would perfectly compliment the painted decoration and the colored and gold stuccos.

Going through the inventory reveals the presence of three altar frontals made of gilt leather, decorated with lilies. Two are unused in the wardrobe, but the last is in place in the chapel, where twelve cushions known as “new” are also found (Table 4). The gilt leather dishes include pitchers, flasks and glasses. These objects are often indicated as being decorated alla turchesca which could imply a Venetian origin. All the unused leathers are grouped together in the same cupboard in the wardrobe. As a matter of interest, a paesetto in corame d’oro, a small gilt leather landscape, is mentioned. There are 21 door curtains and they are often indicated as being of the same type as the squares of the wall coverings with which they are hung in the rooms. A cardinal’s armorials are illustrated in the center. The door curtains are sometimes lined with textile, sometimes with plain leather. It is indicated for one of them, that it is provided with iron hooks and gold rings with a cord to make it slip on a rod. The palace is also equipped with almost thirty-one tablecloths (coperta) of leather. The use of these tablecloths was lost but in the inventories, two types of objects stand out: the coperta (known also as panno) and the sopracoperta (‘a tablecloth going over a liner tablecloth’). The 20

ICOM-CC Leather and Related Materials WG Offenbach 2012

tablecloths were bordered of cascate in a style/color sometimes different from the top. Their dimensions are sometimes impressive. For the prestigious hard stone table of the Grand Living Room (now housed at the Metropolitan Museum of Art in New-York), there was a tablecloth, made up with forty-eight skins, with four edges of gilt punched leather, representing lilies (a second tablecloth is mentioned for the same table in the wardrobe). One can be astonished that such beautiful tables and of such great value were covered in such manner ?

Today, the Palazzo Farnese does not have in situ any more leather because it was gradually stripped in spite of the formal will expressed by the cardinal Alessandro.

Table 4: Objects made of gilt leather at Palazzo Farnese in 1644.

Conclusion

Two palaces, one Villa, three cardinals, artists, craftsmen and many gilt leathers... Because the men who worked for these palaces were often the same ones, the decorations’ arrangement was often identical. The inventories point out the place of the gilt leather on the walls of the most prestigious rooms associated with frescoes friezes and decorated ceilings. Gilt leather contributed to the creation of an exuberant, sumptuous and ostentatious environment. Like the other movable objects, it would contribute to the production of a framework of comfortable and luxurious life, corresponding to the status of their owners; all cardinals, yet animated by the same competitive spirit expressed by the magnificence of their palaces. Purple color (paonazzo) was very present in all decorations and did not leave any doubt about the owner’s high social rank. Today if frescoes and ceilings remain, the disappearance of leather wall-hangings, curtains and other such room furnishings alters our interpretation of these decorative cycles. Inventories can restore at least the knowledge of the existence of these disappeared items, and of their grand place within the luxurious interior settings of the wealthy and powerful.

Acknowledgments This study could not have been carried out without the help of the Académie de France in Roma and especially Dr A. Lemoine, assisted by A. Stahl and A.Gariazzo. We also thank for their invaluable information Dr M. Raspe (Bibliotek Hertziana in Roma) and Dr J-P. Fournet. We are most grateful to Carole Dignard for the corrections.

Endnotes 1. page 44, n°98, Fonti documentarie, La Villa Médicis, Volume 5, 2010

ICOM-CC Leather and Related Materials WG Offenbach 2012

2. page 42, n°91, Fonti documentarie, La Villa Médicis, Volume 5, 2010 (A.R.P., vol. 1769, f.n .n. (91SDR)) ; La Villa Médicis, volume 2, p.141, note 123. 3. We listed nearly 350 commitments over a period of twenty-five years. Examination and study of these archives are ongoing and planned for publication. 4. page 183, n°426, Fonti documentarie, La Villa Médicis, Volume 5, 2010 5. Harpies associated with architectonic elements like columns, are visible on a leather wall-hanging still preserved in situ in Torino, at Museo civico d’arte antica. 6. page 339, n°807-808, Fonti documentarie, La Villa Médicis, Volume 5, 2010

References Bourdon P., Laurent-Vibert R. 1909. « Le Palais Farnèse d’après l’inventaire de 1653 », Mélanges d’archéologie et d’histoire, T.29, 1909. pp.145-198; url: http://www.persee.fr/web/revues/home/prescript/article/mefr_0223-4874_1909_num_29_1_6997 Buranelli et alii. 2010. Palais Farnèse. De la Renaissance à l’Ambassade de France, Ed. Giunti, Rome. 479 p. Butter S., Fumagalli E., Deswarte-Rosa S. 2010. La Villa Médicis. Volume 5, Fonti Documentarie, Ed. École Française de Rome. 662 p. Chastel et alii. 1991. La Villa Médicis. Volume 2, Études, Ed. École Française de Rome. 664 p. Della Latta A. 2005. “Le pelli della corte: arredi di corame alla corte estense con qualche appunto mantovano (Court skins : Embossed Leather Furnishings in Este Palaces, including Notes on Manta)”, DecArt – Rivista di arti decorative. A Magazine for the Decorative Arts, n° 4, autunno/autumm 2005, Florence. pp.3-23. Fournet J-P. 2004. Les cuirs dorés anciens en France. Thèse de doctorat, Ecole du Louvre, Paris. Non publiée. Jestaz B. 1994. Le Palais Farnèse III, 3. L’Inventaire du Palais et des Propriétés Farnèse à Rome en 1644, Ed. Ecole Française de Rome, Rome. 408 p. Rossignoli G. 2009. Cuoio d’oro. Corami da tappezzeria, paliotti e cuscini del Museo Stefano Bardini, noédizioni, Florence. 158 p. Schütze S., Colle E., 2003. Palazzo Sacchetti, De Luca editori d'arte, Rome. 198 p.

Biography Céline Bonnot-Diconne is a leather conservator. She graduated in 1994 at the University Paris I-Pantheon-Sorbonne with a Master II in Conservation and a Master in Archaeology (from the University Lyon II). After 8 years as a leather conservator at the Regional Conservation Laboratory ARC-Nucléart in Grenoble, she established her own private practice in 2002, the Centre de Conservation et de Restauration du Cuir (2CRC). The 2CRC is dedicated to the conservation of any type of objects made of leather or skin, in the fields of archaeology, ethnology, furniture and decorative arts. She is a recipient of the French Academy in Roma award, and spent 18 months from April 2011 to September 2012, at the prestigious Villa Medici. Her main research subject was historic gilt leather techniques and uses.

Académie de France à Rome -Villa Médicis - Viale Trinità dei Monti 1 - 00187 Roma, Italia Web: www.villamedici.it - [email protected]

2CRC- Centre de Conservation et de Restauration du Cuir - Activillage - Centr'ALP - 235 rue de Corporat - 38430 MOIRANS, France - Phone: (+33) (0)4 76 66 14 45 Mob. (+33) (0)6 88 94 71 47 Web: www.bonnotdiconne.fr - [email protected]

Disclaimer These conference session papers are published and distributed by the International Council of Museums – Committee for Conservation (ICOM-CC), with authorization from the copyright holders. They are published as a service to the world cultural heritage community and are not necessarily reflective of the policies, practices, or opinions of the ICOM-CC. Information on methods and materials, as well as mention of a product or company, are provided only to assist the reader, and do not in any way imply endorsement by the ICOM-CC.

ICOM-CC Leather and Related Materials WG Offenbach 2012

A Suit of Lamellar Armour

Nina Frankenhauser

Abstract In 1929, the DLM-German Leather Museum/ Shoe Museum bought a suit of lamellar armour that was collected in Tibet and is presumably of Naxi origin. The armour is made up of hundreds of leather and rawhide or semi-tanned lamellae, which are arranged in dense overlapping rows and are laced to one another with leather and rawhide straps. The armour covers the whole body and also protects the upper arms. All the lamellae are coated with lacquer, probably all with shellac as bonding medium, and some are also painted.The suit of armour shows many signs of changes and repairs, but there is nothing known about any conservation treatment after 1929. After years in the storage room, it was sent to an exhibition in 2012. During its preparation some evident problems were noticed, like loose lacquer and broken or lost laces, so that conservation treatment became necessary. The conservation treatment included a careful cleaning, the consolidation of loose lacquer and damaged laces, as well as here and there the replacement of lost laces and the construction of a suitable mannequin.

Keywords Armour, lamella, lamellar, lacquer, shellac, consolidation, mannequin, Tibet, Naxi, leather, rawhide, semi-tanned skin, treatment, cleaning, repair, replacement, German Leather Museum, Shoe Museum

Introduction

In 1929 the DLM-German Leather Museum/ Shoe Museum bought a suit of lamellar armour (Figure 1) from the German company Umlauff that specialized in the trade with ethnographic objects and supplied museums all over Germany. After decades in the museum´s storage rooms the armour was lent to an exhibition in 2012. In that context there was the opportunity to spend more time on the examination and conservation of this extraordinary object.

Figure 1: Suit of Lamellar Armour. Before conservation: back

The suit of armour was bought in Tibet, at Wassu, but its origin is not proved so far and also not its age. It was presumed that it could be Mongolian, but, on the basis of comparisons with lamellar armours in other collections, it is now assigned to the Naxi. This people lives in the south west of China, in the provinces of Sichuan and Yunnan and is thought to originate from Tibetan nomads, who immigrated hundreds of years ago (Fülling 2012; Hauser, Häring 2005).The armour cannot be compared with Japanese Samurai armour. There exist some quite similar suits of armour made of metal lamellae with Mongolian or Tibetan origin, but not many with leather lamellae (Grant 2010; LaRocca 2006; Robinson 1967).

Description

A lamellar armour consists of horizontal rows of overlapping plates, the so-called lamellae, that are joined by leather lacing or related materials, as in this example. The lamellae are laced only to one another and there is no lining or other support material to which they are fastened. The lamellae typically overlap upwards, contrary to scales or roofing tiles, which overlap downwards.

ICOM-CC Leather and Related Materials WG Offenbach 2012

The suit at the German Leather Museum consists of the armour for the body and shoulder defences for the arms. It is open only on one side, probably the front, from top to bottom. Due to this rigid pattern, it would not be suitable for equestrian use. The asymmetrically attached sleeves might indicate that it was worn by an archer.

The 1558 lamellae of this suit are mainly made of hardened leather and some probably of parchment. The sizes differ, but they all have rectangular shapes with slightly rounded corners and are pierced with 9, 10 or 12 holes. These holes are sometimes round and sometimes quadrangular. The average lamella has a thickness of 2-4 mm. The dimensions vary from 7 to 9 cm in length and 2 to 4 cm in width.

A variety of materials was used for the laces, with three main types: the first and main sort is made of rawhide or semi-tanned hide, sometimes with hair on it. The second, also frequently appearing, type is a yellowish leather. The third sort of laces are of vegetable tanned leather.

All the surfaces of the lamellae are coated with lacquer, and many are decorated in a colourful way. The main colours are black and red, but there is also yellow, brown and orange to be seen. Most do not have fine surfaces, as known from Samurai armours. On many of the lamella, irregularities are found, originating from the application of the lacquer.

Analysis One of the frequently occurring types of lamellae has been analysed by the Microanalytical Laboratory of Prof. Dr. Elisabeth Jägers in Bornheim (IRS and EDX). The sample is composed of thin layers of lacquer with pigments and shellac, as bonding agent and layer, and at the very top, silver-leaf coated with shellac (About shellac see Baur 1966; Koller, Baumer 2000. About the use of shellac on armour see LaRocca 2006).

Description of the analysed sample: Layer on the surface: shellac, strongly aged Fifth layer: shellac with silver-leaf Fourth layer: first layer: red lead, cinnabar, iron oxide-red pigment, lac dye and shellac second layer: cinnabar, lac dye and shellac Third layer: shellac Second layer: cinnabar, iron oxide-red pigment, calcium carbonate, barium sulfate, probably lac dye, and shellac as a bonding agent First layer: calcium carbonate with iron oxide and shellac as bonding agent

On the examined lamella, the silver cannot be seen anymore, because the shellac-surface today is too strongly aged. Whether all of the lamellae have a layer of silver-leaf cannot be verified, but there are some more lamellae where this is possible. Also, it can only be presumed that all the lamellae are lacquered with shellac as the bonding agent.

There are 20 sorts of decoration to be distinguished. The main types are: - black/brown surface layer with red colour beneath (like the analysed sample) - black ground with red crescent or drop - red ground with black crescent, sometimes looking like a comma - flowers on black ground - spots on black or red ground

On some lamellae there are special decorations, such as a drawing that looks like a face, and on other lamellae lines that are reminiscent of letters. On the reverse, there is also one lamella with a carving on the surface that is presumed to be a seal. On the inside the colour of the lacquer is mainly brown and yellow and without decoration.

ICOM-CC Leather and Related Materials WG Offenbach 2012

The Manufacture of the Suit of Armour

It is not so easy to determine how the armour could have been manufactured, because there are many old repairs and damages. But repairs and damages are also a chance to see details that otherwise would have been hidden.

As far as one can say at present, the manufacturing process consisted of the following steps:

1. The lamellae were cut with sharp knives and the holes for the laces pierced with punches. 2. The lamellae were then probably joined in overlapping horizontal rows. 3. Then they were lacquered. This is the reason why often one third of a lamella is not lacquered and decorated. The lacquer was applied on the leather with some kind of brush leaving the typical brush strokes. 4. After that the rows were joined vertically. The armour consists of 18 rows on the front and 19 rows on the reverse. The sleeves which are fixed by rawhide laces onto the lamellae around the neck consist of seven rows.

Around the bottom edge of the coat and also of the arm defences, rectangular tabs made of leather or semi- tanned leather, are attached to each lamella. The suit was probably closed with a belt around the waist. This belt is lost, but the lamellae of the seventh row, that is near the waist, are slightly curved, perhaps caused by a belt.

Re-use Some peculiarities may indicate that to manufacture this suit, parts of two or more other suits were re-used. Firstly, the armour is really colourful with lots of different designs on the lamellae, but perhaps this is not unusual. Secondly, there are rows of lamellae where the lacquer-surface shows that the lamella was once covered by the neighbouring lamella on the other side. And finally, there are three ways in which the lamellae are connected by the laces vertically. Pattern one is mainly used at the top of the suit and on one of the sleeves. Pattern two is used further down, beginning with the eighth row and on the other sleeve. The third pattern is not very often to be seen. However, you can sometimes find the same decorations on all three patterns, for example the decoration with flowers.

Repairs One can also see smaller repairs, especially on the parts around the front opening of the suit. Probably lost or damaged lamellae were replaced by other, larger lamellae. Often lacing materials that were different from the surrounding ones were used and the repair was not done in an accurate way. There are also some lamellae where the lacing holes were broken and have been repaired. Deep cuts in some of the lamellae possibly had been caused by a weapon.

Condition

In more recent years the armour was always stored in the dark, lying horizontally in an archival museum box, but in an old storage room, which sometimes had climatic problems. The dirt on the surfaces consisted of loose dust from decades of storage, and some kind of grey, quite firm dirt attached to the surface that sometimes looked like sand or soil. The lacquer on many of the lamellae was loose (Figure 2).The condition of the lacquer in some areas was so bad, that the object could not be moved. This mainly applies to the uppermost layer, which, on the analysed lamella, was determined as strongly aged shellac layer. As well, some lamellae had all of their lacquer layers laying loose as a compact whole sheet. But not all lamellae were damaged. Some rows of lamellae were completely in good condition.

ICOM-CC Leather and Related Materials WG Offenbach 2012

Concerning the laces, many of them were broken, or had parts of the laces lost (Figure 2). Especially when they connected two lamellae horizontally and the two lamellae were gaping apart, this was not only an aesthetic problem but above all it adversely affected the stability of the armour. One particular damage only concerned the laces made of the yellowish leather. The grain side of these laces showed broad horizontal tears. The reason for this damage could be the weight of the lamellae that are hanging on these laces.

Figure 2: Lamellae with loose lacquer and damaged laces. Treatment Rationale

The armour was planned to be sent to a non-local exhibition and will become part of the new permanent China and Tibet room at the museum. This meant that the conservation treatment had to be designed so that the object could, not only withstand the stress of the transport, but also so that it could be suitably and aesthetically displayed.

Treatment

The conservation treatment consisted of the following steps: - examination - careful cleaning - consolidation of loose lacquer - support of broken laces and replacement of lost laces that are important for the stability of parts or the whole object - manufacture of a suitable mannequin

The suit of armour consists of many different types of skin and leather lamellae, with some also coated with lacquer. To take fibres from the skin or leather of only a number of the lamellae, with the aim of measuring the shrinkage temperature and then to decide about the condition of the whole armour, would not have been useful; as well, the same representativity limitations would have occurred if we would have analysed the lacquer of only a number of the lamellae. The most useful way we found to proceed, was to be careful and to observe each surface before and during any treatment carried out. The only parts of the suit of armour where it was really useful and possible to measure the shrinkage temperature were the heavily damaged laces of yellowish leather. Here the shrinkage temperature Ts was 55°C (samples taken from different laces).

Cleaning and Consolidation The cleaning involved some problems. First, it was necessary to determine how far the cleaning should go. The second problem was that due to the loose layers of lacquer, the cleaning could only be done simultaneously with a consolidation. It was decided to only remove the loose dust and to maintain the firm grey dirt that seemed not to have a damaging effect on the lacquer surfaces.

Consolidation of the loose lacquer was inevitable, because in its present condition it was not possible to move the object in any way. The consolidation medium had to be suitable for different purposes. On the one hand, there were thin loose parts of lacquer, consisting only of the uppermost lacquer layer. On the other hand, large fragments of many lacquer layers had broken off. These fragments needed to be adhered down rather than consolidated. The aim was not to use different materials for the consolidation and the glueing of the lacquer. The material should give strong bonds and have a certain degree of flexibility. Alcohol should not be used as a solvent because of the shellac, although aged shellac does not easily solubilise in alcohol.

ICOM-CC Leather and Related Materials WG Offenbach 2012

Solvents were tested on diverse small lacquer samples that had fallen off and were found laying in the museums storage box with no possibility to relocate them. Of course these samples could not be assumed to be representative for the whole object. As water seemed to have no negative effect on the lacquer at all, parchment glue and sturgeon glue were considered and also acrylic dispersion adhesives, but in the end sturgeon glue was chosen. It was suitable for the consolidation of the lacquer, where a fine brush or a syringe was used, and also in a gelled texture to glue the larger fragments. Many different solutions rangeing from 3 to 10 % were used, depending on what had to be consolidated or glued. During the drying process, these parts were fixed by means of cardboard, clips and clamps and Hostaphan (polyethylene) foil as an interleaf material, depending on what was needed. Where there are rawhide laces, the suit is quite stiff and difficult to move. That is why it was often not easy to fix the loose layers.

Cleaning and consolidation were done simultaneously row after row, starting on the sleeves where no consolidation was necessary. Cleaning was carried out overall with a vacuum and fine brushes, with a “Wallmaster” cleaning sponge also used for the parts in good condition and where the lacquer was in a very good condition. The cracks on the yellowish leather were adhered with rice starch paste.

Replacing Lost Laces Some of the lamellae needed to have their lost laces replaced by some kind of support material. Leather or a similar material was avoided to make it distinguishable from all the diverse original hide and leather materials on the armour. Strips of thick Japanese tissue paper were found to be suitable. To make them stronger, they were soaked with sturgeon glue and air-dried. These strips were then each inserted into the holes of both lamellae and the ends of the strips glued together. From a distance the strips are visually similar to the rawhide laces. On close examination, they are recognizable as paper and thus clearly distinguishable from the original laces (Figure 3). Figure 3: Lost laces replaced with Japanese tissue

paper strips. Construction of a Mannequin The suit of armour weighs a total of 15 kg. Since it is presented standing in a showcase, it is important that the entire weight does not hang from the shoulder area of the suit but is instead evenly distributed upon an appropriately constructed mannequin. This was achieved by means of attaching hoops to the main frame at varying heights. The hoops have holes and always one row of lamellae is supported by one hoop and can be fixed to it by means of, partially padded, nylon cords, so that the weight is distributed.The mannequin is made of powder-coated steel and aluminium and is adjustable in height. It can be turned without first having to remove the object (Figure 4).

Display Recommendations To conserve the lacquer and the leather a sparing illumination of a maximum of 70 lux and a constant climate are absolutely necessary.

Figure 4: After Conservation with new mannequin: front

ICOM-CC Leather and Related Materials WG Offenbach 2012

Conclusion

During the conservation treatment, considerable new information could be gained concerning the technology and history of this exceptional object, but many questions remain unanswered. The aims of the conservation treatment were firstly to alter the original object as little as possible, so that for example the greyish dirt was left on the lamellae surfaces, and secondly to make each necessary change clearly visible, such as the replaced laces made from Japanese tissue paper. The consolidation of the lacquer was carried out because it was judged to be necessary for the preservation of this object. Now, after decades in the storage room, the suit of armour will be part of the new permanent China and Tibet exhibition.

Acknowledgments Thanks to Dr. Christian Rathke and to all the colleagues at the DLM-German Leather Museum/ Shoe Museum, especially Jutta Göpfrich, for their support. Prof. Dr. Jan Bemmann and Prof. Dr. Peter Schwieger, University Bonn, Dr. Michael Schmauder, LVR- Rheinisches Landesmuseum Bonn and Jeanette Werning M., Reiss-Engelhorn-Museen Mannheim are thanked for all the helpful information. All Images © DLM-Deutsches Ledermuseum/Schuhmuseum, Offenbach; Corinna Perl-Appl, 2012.

Materials Sturgeon glue, sturgeon glue pellets: Störleim Manufaktur, Eva Przybylo, D-59929 Brilon Japanese tissue paper: Japico GmbH & Co.KG., D-Dietzenbach Hostaphan® PE foil: Pütz GmbH + Co. Folien KG, D-65232 Taunusstein Rice starch: Gabi Kleindorfer, D-Vilsheim Wallmaster® cleaning sponge: natural latex 100%, Deffner & Johann GmbH, D-Röthlein/Schweinfurt

References Baur P. 1966. Aus der Frühgeschichte des Lackes – Laksha + Gummi Forniß, Farbe + Lack, Zentralblatt der Farben- und Lack-Industrie und des Handels, 72. Jahrgang, Hannover, pp.5-12 Fülling O. 2012. China – der Süden, DuMont Verlag, Ostfildern, pp. 426 Grant R.C. 2010. Krieger, Kämpfer und Soldaten – von der Antike bis Heute, Dorling Kindersley Verlag GmbH, München, pp.92-93 Hauser F., Häring H. 2005. China-Handbuch – Erkundungen im Reich der Mitte, Trescher Reihe Reisen, Trescher Verlag, Berlin, pp.406 Koller J., Baumer U. 2000. Historical Black Lacquers in Europe. Black Lacquers of the 17th and 18th Century, East Asian and European Lacquer Techniques, International conference of the Bavarian State Department of Historical Monuments and the German National Commitee of ICOMOS together with the Tokyo National research Institute of Cultural Properties Munich, 11-13 March 1999, Michael Kühlenthal ed., Bayerisches Landesamt für Denkmapflege, Karl M. Lipp Verlag, München, pp.157-162 LaRocca D.J.2006. Warriors of the Himalayas – Rediscovering the Arms and Armor of Tibet, The Metropolitan Museum of Art, New York, pp. 51 Robinson Russel H., F.S.A., 1967. Oriental Armour. The Arms and Armour Series. W.J.Mackay co Ltd, Chatham, Kent, pp.126-166

Biography Nina Frankenhauser is an Objects Conservator with a Diploma in Conservation and Restoration of Archaeological, Ethnological and Craft objects from the State Academy of Fine Arts Stuttgart, Germany, 2002. Then employed at a Free- Lance Conservation Studio, the Landesmuseum Württemberg Stuttgart and the Bomann Museum Celle. Since 2006 employed at the DLM-Deutsches Ledermuseum/ Schuhmuseum Offenbach. DLM-Deutsches Ledermuseum/ Schuhmuseum, Frankfurterstr.86, 63067 Offenbach am Main, Germany - [email protected]

ICOM-CC Leather and Related Materials WG Offenbach 2012

The Use of Correlation Methods to Detect Surface Deformation and Changes in the Leather Wall-Coverings in Jever Castle Museum Angelika Gervais, Peter Königfeld

Abstract At Jever Castle there are deformations in the leather wall coverings primarily due to the way they are fixed and because of the effects of the sewing technique used and the room climate. Parallel to optical deformation measurements taken at the castle, studies on the leather samples were carried out in the climate chamber in cooperation with the Institute for Restoration of the Hildesheim/Holzminden/Göttingen University of Applied Sciences. The laser speckle technique was used for the first series of measurements on the basic deformation behaviour of one historical and one present-day sample of leather.

Keywords Leather, Jever Castle, Friesland, 18th century, speckle correlation method with laser light, digital image correlation with white light, vibrometer measurements with laser light.

Introduction

It is common to find leather wall coverings housed in historic settings that are deformed due to the way they are fixed and because of the effects of the sewing technique used and the room climate. A pilot project was proposed by the ZMK (North German Centre for Materials Science for Cultural Heritage) and sponsored by the DBU (The German Federal Foundation for the Environment). The entire project was carried out in close cooperation with the Lower Saxony State Office for Monument Preservation. The objective was to develop a cost-saving restoration system for treating, maintaining and presenting historical leather objects damaged by the environment. It was imperative to consider the local conditions which directly affected the leather material as well as matters concerning the material (aging leather, original pigments and binding agents, sewing materials and conservation and restoration materials).

Description of the Project

Fundamental restorative and scientific assessments were carried out on two sample objects - the altar antependium from the Old Church of St. Alexander in Wallenhorst, near Osnabrück (made up of differing pieces of old leather from the baroque period) and leather wall coverings in Jever Castle (Edzard room), in Friesland from the 18th century.

Mechanical strains on the leather account for considerable damage to the leather wall coverings: tension at the seams and nail holes, cracks and sometimes shearing of the mounting layers. The origin of these deformations and how they relate to changes in the environmental conditions – especially the climate - was investigated by K. Hinsch (2003), University of Oldenburg, with the aid of non-contact measurements of deformation fields, and also by H. Hinrichs (2003), Hochschule Bremerhaven, with test specimens in a simulation chamber – see list of literature: Physics.

In both examinations, the testing of optical measuring methods was included in the pilot project (2001-2004) to establish their suitability for such examinations. Using these methods was expected to result in evidence to assess the aforementioned damage processes and also solve the problem of mounting the wall coverings as well as providing recommendations for a moderate room climate.

Within this work programme, two sets of examinations were carried out on:  The mechanical behaviour of leather specimens under controlled climate conditions in a simulation chamber;  The identification of characteristic movements and deformations of wall coverings on site.

ICOM-CC Leather and Related Materials WG Offenbach 2012

In order to record the influence of humidity and room temperature under defined conditions pre-tensioned leather specimens were exposed to predetermined cycles in a climatic chamber. One specimen was made of new calfskin (10 x 10 cm²), another one was from a stored remnant stemming from the painted leather from Jever Castle; this specimen was divided in the middle and sewn together again in order to examine the behaviour at a seam.

The completed humidity and temperature cycles (40° C, increase of humidity from 30 to 50%), where the specimens were nailed to the top edge of a wooden lath and pulled down with a typical weight (1kg tensile load) corresponding to roughly ten times the weight of the approximately 3m long and 50 cm wide strips of leather wall covering, indicated that the leather wall covering had retained its elasticity. The reversibility of stretching and bulging in the leather of the damp area in situ could also be tested using this method.

After the initial measuring campaign the specimens were left to age in the chamber for two months and then the measurements were repeated.

While H. Hinrichs was carrying out these measurements, C. Meier (2002) was using a climatic chamber for conducting restorative examinations on the stretching behaviour of leather with historical and restorative stitching on the leather wall coverings in Jever.

Speckle Correlation Method with Laser Light

In the measuring method used by Hinrichs, the surface of the prepared specimen (nailed to the lath at the top, at the bottom wedged in a profile groove but freely movable to right and left, no force exerted) was illuminated by laser light and photographed (Figure 1). This results in a phase-sensitive speckle pattern which reacts very sensitively to changes in the surface microstructure (in the micrometer range). In sequence, images are recorded with the changing climate. The correlation method shows the local surface change as small structures from patterns in the first image have changed slightly in the second image which is calculated by mathematical correlation. If the pattern not only changes in the structure, but has also been displaced in plane, the displacement is obtained in the form of a vector as an additional local information.

Figure 1: Typical structure of speckle correlation procedure. The surface to be examined is lit up by laser light. The resulting speckle-pattern is recorded by a high-resolution camera and compared locally with previous records.

ICOM-CC Leather and Related Materials WG Offenbach 2012

Figure 2: Evaluation of speckle-correlation measurements on 10x10 cm² leather samples during a humidity change in the climate chamber. Horizontal and vertical axes indicate position in the sample in arbitrary units, the coloured area representing roughly the leather piece. Colour plots the correlation coefficient in the speckle patterns which indicates surface change from no change (value 1 - white) to large changes at low values (black). The blue arrows give the displacement vector, maximum arrows corresponding to approximately 150 µm. Top A: Single piece of present-day calf-skin leather, artificially aged. Bottom B: Historical leather from Jever Castle consisting of two pieces stitched together in the middle in a horizontal seam.

Experiments

Test Conditions The humidity at 40°C was increased from 30% to 50%.

ICOM-CC Leather and Related Materials WG Offenbach 2012

Samples A (top image in Fig. 2): Artificially aged present-day calf-skin leather sample. B (bottom image in Fig. 2): Historical leather from the museum at Jever Castle consisting of two pieces stitched together in the middle in a horizontal seam.

Depiction of the Leather Response: In-plane lateral displacements of the leather are shown by vectors (blue arrows), while changes in the surface microstructure are scaled in colour. White means no change small to large changes are shown by an increase from yellow to orange to red; the most severe surface changes are indicated in black.

Results:

 Examination of test specimens in laser speckle correlation experiments (Figure 2) In the present-day leather sample A the strong decorrelation indicates a pronounced response of the leather to the increased humidity. This can be attributed to the unprotected surface that allows the water to change the leather microstructure. The historical sample B, on the contrary, has a robust decorative paint cover and responds much less. Concerning the displacement values, the soft present-day leather has been stretched by the load and experiences little additional strain from soaking with water resulting in a pattern of rather irregular and small displacement vectors. The historical sample B, however, presents a nice regular displacement field showing a downward motion under the external load – even across the seam.

 Digital image correlation with white light As early on-site experiments with laser light (laser speckle correlation) failed because of strong decorrelation, i.e., the effects were of such extent that the speckle patterns of subsequent images showed no similarity any more , image correlation in ordinary white light proved to be ideal. This does no longer rely on detecting displaced speckle patterns but utilizes minute image structures in the object or produced by the illumination. Furthermore it was shown that the displacements to be measured were of the right order (in the millimetre range) for this method. The selected area was observed from the tiled stove for 11 days in constant measurement. Parallel to that, Berling (2002) recorded humidity and temperature and additionally how the rooms were used. It became clear that the wall covering mainly moved downwards whereby the displacement (max. within one day of 2,7 mm) was zero at the lower fixture. There was no evidence of abrupt changes at the seams. Perhaps the wall covering expanded along its whole length as a result of the increasing humidity during the observation period. As it is fixed at the bottom it could have been tilted out of plane, possibly around the horizontal seam (Figures 3 and 4).

 Vibrometer measurements with laser light When doors and windows are suddenly closed, shock waves arise, which in this particular case are directed outwards from the plane of the wall covering. In order to gain an impression of typical signals, in addition to Berling’s airflow measurements, further measurements were taken at two positions in the passageway to the audience hall. A laser vibrometer is an instrument that allows to measure single-point displacements. Its laser beam was always focused onto the measurement point in the leather, so that its diameter there was little more than one millimetre. The movements of the wall covering after sudden airflow movements (banging doors, opening and closing windows) and also mechanical strains on the material were found to range up to 100 micrometer and more (Figures 5 and 6). Shape and magnitude of the oscillations depend on the type of source (door/window) and the geometry of the architecture (shock waves may even be reflected). In summarizing it can be stated that such disturbances cause considerable impact on the historical leather.

ICOM-CC Leather and Related Materials WG Offenbach 2012

Figures 3 and 4: Average movement (left y axis, in mm) of the leather piece within the measuring area. Red: Vertical displacement, black: horizontal displacement. The x-axis shows the time (in days) starting at day 6.5, up to day 18.5. On top of that the violet line shows the relative humidity (top right y axis, expressed in %) measured and the green line shows the air temperature (bottom right y axis, expressed in degrees C). All displacements refer to day 8.5 when a new reference image was taken. Conclusion of the measurements: The leather stretches and pulls out of shape the most when the relative humidity changes the most (from 55% to 65%) within a short period of time (1 day).

ICOM-CC Leather and Related Materials WG Offenbach 2012

Figures 5 and 6: Typical laser vibrometer readings of displacement (vertical-axis) at a position (“Messort 1=measurement location 1”) on the leather wall covering when a door (top) or a window (bottom) is suddenly closed. The resulting shocks produce oscillations in the leather with amplitudes of up to several 100 µm. The horizontal-axis shows the time in 0.5 sec intervals the direction of motion towards the vibrometer is indicated by the yellow arrow (“Bewegung zum Messkopf”= motion to probe).

Conclusion

The leather tapestries in Jever Castle do not require any further conservation procedures such as those required in the Old Church of St. Alexander in Wallenhorst, but they need the climatic conditions to be stabilised with a reduction in the effects of the outside climate, balanced climatic conditions in front of and behind the wall coverings, the prevention of increased changes of air and air movements as well as

ICOM-CC Leather and Related Materials WG Offenbach 2012

exposure to the sun. In the sample axis brought about in in Jever Castle the strip of wall covering was suspended freely and fixed at the top and bottom with the sides sliding in anchorage points so as not to impede the shape deformations of the material (increase in length– shrinkage) due to relative humidity variations. The example examinations in the pilot project showed among other things that using optical procedures for measuring etc. The methods have sufficient sensitivity to provide significant measurement data.

Acknowledgments We are thankful to Prof. Dr. Klaus Hinsch and to Prof. Dr. Heiko Hinrichs and to the DBU (The German Federal Foundation for the Environment), Osnabrück, supporting the DBU-Project no. 18321. For translation we thank Pamela Seidel, Laatzen.

References Climate Berling, H.: Nahfeldklimamessungen im Bereich des Edzard-Zimmers im Schlossmuseum Jever, Abschlussbericht, Braunschweig, 10.12.2002 (masch.schr.) Physics Hinrichs, H.: Abschlußbericht der Teilaufgabe: Einsatz von Korrelationsverfahren zur Untersuchung von Oberflächenverformungen und Strukturveränderungen von Lederproben in einer Klimakammer, Varel 2003 (masch.schr.) Hinsch, K.: Abschlußbericht der Teilaufgabe: Einsatz von Korrelationsverfahren zur Untersuchung von Oberflächenverformungen und Strukturveränderungen des Leders, Oldenburg, Februar 2003 (masch.aschr.) Restoration and object history Meier, C.: Die barocken Ledertapeten in Schloss Jever, Objekt- und Restaurierungsgeschichte. Hausarbeit zur Fachprüfung Kunstgeschichtliche Grundlagen, Institut für Restaurierung, Fachhochschule Hildesheim/ Holzminden/ Göttingen, Juli 2002 (masch.schr.) Meier, C.: Restauratorische Untersuchungen zum Dehnungsverhalten von Leder mit historischen und restauratorischen Vernähungen an den Ledertapeten im Schlossmuseum Jever. Vordiplom im Hauptfach, Institut für Restaurierung, Fachhochschule Hildesheim/ Holzminden/ Göttingen, Sommersemester 2002 (masch.schr.)

Biographies Dipl.-Geol. Angelika Gervais studied Geology, scientific trainee and research assistant at the Lower Saxony state Museum in Hanover. Since 1992 working in monument preservation research, various projects to preserve cultural heritage, amongst other things with the aim of using scientific methods for restoration. Member of VDR, ICOMOS ZMK e.V. - Hinüberstrasse 19 - D-3075 Hannover, Germany - Phone 0049 (0) 511 – 27900833 [email protected]

Dr. Peter Königfeld trained as a restorer, studied History of Art, then until 2003 worked at the Lower Saxony State Office for Monument Preservation as head of the central restoration workshops. Specialist advisor and supervisor of important restoration measures. Interdisciplinary work to provide the scientific principles for conserving cultural heritage. Warmbüchenstrasse 10 a - D-30159 Hannover, Germany [email protected]

ICOM-CC Leather and Related Materials WG Offenbach 2012

Treatment and Mounting of Medieval Parchment Fragments

Barbara Hassel

Abstract This paper deals with the conservation of 44 fragments of medieval parchment manuscripts, which were found together in a ceiling-floor interspace in the former Cistercian monastery, Kloster Eberbach. Cleaning, reshaping, repair, mounting and housing are described.

Keywords Parchment, manuscripts, conservation, treatment, cleaning, reshaping, repair, mounting, Kloster Eberbach

Introduction

Unique fragments of medieval parchment manuscripts were found in 2009 in the cavities of a cross-ribbed vault under the wooden floor of the attic of the Hospital building at Eberbach Abbey during reconstruction work. Kloster Eberbach, a former Cistercian monastery with impressive Romanesque and early Gothic buildings, is sited near Eltville on the Rhine in Germany. How the rather randomly discovered manuscripts and things of daily religious life found their way to this hidden place has not yet been investigated, but the manuscripts may be significant as the Abbey library was robbed during the 1618-1648 war and the contents spread all over the world. The findings were pre-selected and sorted in 38 plastic folders according to the finding location by the archaeologist in charge. Thus 179 items came to be preserved. The 131 paper manuscripts, from the 17th to the 19th century, are personal letters, invoices or receipts, court files or inventory listings, paper seals and two 1910 newspaper prints. Two fragments are from leather bindings; a silver tooled 1602 goat skin front cover and an unusually small chamois suede leather girdle book pouch. The 44 parchment fragments which will be discussed here, include pages from prayer books, records, charters, an architectural drawing and blank scraps.

Work

Object Description The 44 parchment manuscript fragments were tentatively dated by a historian to the 13th-15th century according to inscriptions on the manuscripts, the style of lettering or their content. The most homogenous groups are sections from two or three prayer books. Eleven single folded sheets of 15 x 20 cm came in two sections of five and six (Figure 1). These manuscripts are written in black ink and rubricated with blue and/or red initials. The others are seven, small, single-folded sheets of about 6 x 11 cm, forming two sections and some individual sheets of about the same size. Then there are single-folded pages of a letter book with dimensions of 13 x 20 cm. Six more or less fragmented charters, one of which is dated 1420, an architectural drawing of a gothic building structure, two pieces of a so-called rotulus, a long rolled parchment strip, used for recording taxes and debits, and a few smaller pieces bearing as yet unidentified text and some blank vellum scraps add to the collection. The parchments prepared for use as manuscripts or prayer books are scraped and written on both sides, those for Figure 1: Objects 16-21 in plastic folder on arrival charters are one-sided. at the workshop ( Hassel, 2010)

ICOM-CC Leather and Related Materials WG Offenbach 2012

Condition Most of the parchment manuscripts were in very poor condition: soiled with incrustations of earth and sand, stained by dust and water damage, blocked with adjacent pages, brittle, inflexible and deformed (Figure 2). One reason for the bad condition was the unprotected environment in which they must have been for years, between the floor of the attic and the ceiling of the church-like room below. They were exposed to continuous fluctuations of humidity and temperature, the best sort of artificial aging. Insects and rodents had eaten away at the organic material. Microorganisms profited from the substrate and seemingly, salts had crystallized on the surfaces. The black and brown inks had become friable and were powdering or flaking off. In some instances, the ink letters could be lifted off the parchment with a spatula. The red and blue initials had migrated or washed out into the surrounding area or had copied to adjacent pages. The smaller manuscripts were less deformed and soiled, but their inks were even more friable. One of the charters was severely infested by microorganisms and had suffered from high humidity; the ink seemed to have dissolved the velvety white flesh side unless silverfish have eaten the letters. The architectural drawing was purple stained from micro-organisms.

Figure 2: Object 16 unfolded before further conservation treatment ( Hassel, 2010)

Technical Examination Shrinkage temperature (Ts) measurements were performed on selected items to obtain an estimation of their hydrothermal stability (Larsen, 2002). Two sets of fibers from four objects were observed at 32x magnification on the Linkam TH 100 micro-heating equipment, heated by a Eurotherm 2132 at a rate of increase of temperature of 2°C per minute. The measurements were performed by Ines Jesche, student at the BA/MA program "Studiengang Konservierung und Restaurierung von Graphik, Archiv- und Bibliotheksgut" at the Staatliche Akademie der Bildenden Künste in Stuttgart, Germany. Samples from very deteriorated areas disintegrated on first contact with water, which is a sign for maximum deterioration, because these fibers denaturate to gelatin at 20°C, room temperature. The fibers from areas with obviously better conditions showed a Ts of around 50 °C, which means they are still quite stable but definitively more deteriorated than new parchment which has shrinkage temperatures of 55-64°C (Thomson, 2006). From the results it was clear that aqueous treatments or water based adhesives were not advisable.

ICOM-CC Leather and Related Materials WG Offenbach 2012

Treatment / Results

Dry Cleaning The parchments were mechanically separated from each other where they were blocked and surface cleaned by the use of a scalpel, soft and hard brushes and a museum vacuum cleaner at a distance (Figure 3). Personal protection was used when removing mold infestation and microorganisms. Further disinfection was not performed as a dry storage is foreseen. Where the inks and pigments were flaking or powdering off easily at contact with the tools, cleaning was performed around the letters. Less fragile areas were dry cleaned by dabbing with a Wallmaster latex sponge. Most dust, sand, earth incrustations, wood, straw and metal particles, insect pupae and salt deposits could be removed from the surfaces. Water stains were not further reduced.

Figure 3: Object 11 during dry cleaning and removal of earth incrustations ( Hassel, 2010)

Reshaping Humidification was required to soften the brittle folded manuscripts and reshape the distorted documents so that they could be mended, mounted and used. The least possible humidification was desired and no contact with liquid water, as this accelerates ink corrosion and the hydrolytic breakdown of aged collagen fibers (Hassel, 2003). The humidity chamber is a plastic tray covered with an acrylic sheet. A wet cotton towel, sprayed with 70% ethanol-water solution is placed on the bottom of the tray and covered with a thick layer of polypropylene felt, such that no liquid water can touch the object. Finally, a GoreTex membrane is placed on top such that the water vapor diffuses in the direction of the object, which is supported by Hollytex to aid the transportation from the chamber to the working table (Figure 4). In the chamber 92% relative humidity was reached after an hour. The objects were positioned and their suppleness regularly checked. Most objects could gently be smoothed out after one to six hours humidification, some areas needed repeated sessions of humidification and working. The alcohol was meant to hinder mold growth during the session, but the smell rising from the tray on opening was in no way sterile.

Figure 4: Humidification chamber set-up ( Hassel, 2012)

The best method to reshape parchment is to dry it under tension. In this case it was not possible because of the numerous lacunae and the locally powdery degraded state of the material. Therefore the objects were placed in a soft padded drying stack between Hollytex, polypropylene and woolen felts and blotters and left to dry under wooden boards and weights for 5 weeks. Soft materials were chosen for the drying sandwich to prevent the parchment from compressing and becoming transparent. The result was satisfactory as complete flattening was not intended. The aim was to make the text areas accessible by evening out distortions and creases. A nice side effect was that some flaking inks rebonded to the parchment surface, probably due to reactivation of their binding media.

ICOM-CC Leather and Related Materials WG Offenbach 2012

Repair For the repairs, Japanese paper was chosen. Parchment or goldbeater`s skin were excluded as mending materials as they have their own characteristics and were thought to be liable to create too much tension on the weak originals. Furthermore the numerous lacunae were more easily patched with torn Japanese paper than with scraped parchment.

A water free adhesive with acceptable adhesive strength is KlucelGF, a hydroxypropyl cellulose ether soluble in and reversible by organic solvents (Feller, 1990); Lascaux, 2012). It was selected as the application of water based adhesive was to be avoided, although methyl cellulose, gelatin and isinglass are otherwise preferred in parchment conservation. A tissue which could be reactivated was produced from Japanese paper RK0 and 5% Klucel solution and was tested, however the adhesive strength was not satisfactory. Therefore direct brush application was chosen. The mold damaged, powdering areas of the parchments were consolidated with a 2% KlucelGF solution in isopropanol and if necessary, locally reinforced by pasting 5g/m2 RK0 Kozo fiber tissue to the surface. If necessary, text areas were secured by the more transparent 2g/m2 Berlin Tissue. Missing areas were filled with Ino Shi, a 18g/m2 Kozo fiber paper, dyed with Luganil paper colors. The outlines of the missing areas were transferred on a light table to the Japanese paper with a needle or awl. The edges of the inserts were torn to overlap by 1-2 mm and pasted to the manuscripts’ reverse side with 5% Klucel GF in isopropanol. If necessary, the overlapping zone was further stabilized from the other side with Berlin tissue and a 2% Klucel solution (Figure 5). The edges of the manuscripts were normally not filled back to a rectangle but to their “natural” outlines after the missing areas had been treated. Tears were mended by use of 5% Klucel GF and small strips of Japanese paper RK0.

Figure 5: Object 20, insertion of a missing area and reinforcement of the overlap  Hassel, 2011. Mounting The manuscripts were to be prepared for safe use, storage and exhibition. Frequent handling by staff in the monastery, researchers and archivists was to be expected. Therefore due regard was given to the mounting of all items. It had to be suitable for giving safe access to both sides of the manuscripts without having to touch the fragile surfaces. In addition the documents were to be preserved for non- intrusive handling on exhibition and extended storage. These requirements were fulfilled by mounting each fragment in a double sided window mat, slid into a Mylar sleeve (Figure 6). The outer measurements of 30 x 40 cm were decided on with respect to standardized formats of frames, sleeves and archival boxes.

Figure 6: Mount of the parchment objects ( Hassel, 2012)

ICOM-CC Leather and Related Materials WG Offenbach 2012

The exact outline of the mended parchment object is transferred to a sheet of 240g/m2 archival cardboard by pencil. Two of these boards are laid on top of each other and cut simultaneously so that the remaining frame is 2 mm larger than the object. The object is placed on the frame of the lower board and hinged to it with strips of 18g/m2 Kozo fiber paper Ino Shi. The hinges are placed every 3-5 cm all around the object. The strips, cut to 1 x 5 cm, are pasted with 5% Klucel to the edge of the object, overlapping 3-5 mm. The other end of the Japanese paper strip is pasted 5 mm wide to the cardboard with wheat starch paste (Figure 7). The intermediate 4 cm of each hinge is left unfixed so that it can slightly expand and contract when the object reacts to changes in relative humidity. The second board is positioned to cover the hinges and is fixed at the longer top or left side with two V-shaped hinges of gummed archival paper.

Figure 7: Object 2, Japanese paper hinges Figure 8: Object 2, mounted in the double sided attached ( Hassel, 2011) window mat ( Hassel, 2011)

Two 1,6 mm archival matboards 30 x 40 cm are prepared by cutting identical square windows at a distance of 5 mm to the largest object outline. These window mats are placed on both sides of the mounted object and fixed again on one side with gummed paper strips. They serve as spacers between the object and housing unit. While each of the larger objects is mounted in its own window mat, two or three of the smaller fragments are mounted together in one mat (Figure 8).

The objects are mounted such that they can be looked at from both sides immediately. They float in their mounts and are able to react to minimal climatic changes. However they float like a trampoline and therefore are endangered by local pressure. Double sided acrylic glazing was an idea to protect the sheets from physical damage. However costs, weight and required storage space were against this. An acceptable compromise was to slide each sandwich into a DIN A3 standard size 75 Mylar polyester sleeve welded on three sides and tucked in on the fourth side. Thus the objects are well protected against dust, oils and dirt on handling and may yet be removed from the sleeve if necessary.

Housing All 178 parchment, leather and paper objects were mounted and housed in the same way, except that the paper objects had no window mats as spacers. The one hundred sleeves were placed in ten folders, custom made from corrugated archival board, joined at the left long side with black Iris linen cloth. The folders were placed in three standard size archival boxes which were modified to Solander boxes for better handling. The boxes were supplied by the Hessisches Staatsarchiv Wiesbaden, where the collection is to be stored for an extended period (Figure 9). Figure 9: Object 10 verso and 11 recto in their double sided mats and Mylar sleeves in an archival box ( Hassel, 2012)

ICOM-CC Leather and Related Materials WG Offenbach 2012

The collection was delivered to the monastery where it is temporarily deposited in an office space without climate control. After the selection of the items which will go on permanent exhibition, it will be transferred to the above mentioned archive.

Exhibition A selection of items will be on permanent exhibition in the Abbey Kloster Eberbach and may be exchanged with other items on a regular basis. A showcase, humidity controlled by Art Sorb, is yet to be constructed. The objects may be positioned in the showcase in their mats and sleeves with the option to view the reverse side of the manuscripts. Alternatively they may be framed in double sided 30 x 40 cm standard size frames with glass or acrylic glazing and be presented so that both sides are visible. The exhibition is scheduled for the end of 2012.

Documentation Digital images were taken of the plastic folders containing the bundles of fragments as presorted by the archaeologist. The hand written numbers on the plastic folders mark the coordinates of the finding places and were registered in a table together with the number and type of objects in one folder. Each fragment was given a serial object number as it came out of the folder. Therefore the numbers are deliberate as to type or quality of the object and only later was it possible to associate fragments with each other to create larger units. After separating and counting the objects, pictures of both sides of each fragment were taken before and, later, after the treatment. Close ups of interesting features and steps of the treatment were recorded. The pictures were taken at daylight with a Konika Minolta Dimage A2.

Relevant information of each object such as serial number, dimension, date, material, ink, content, condition, treatment, methods and materials are listed in an Excel file. The print-out for each object contains a documentation sheet and four images and is, together with a digital version, delivered to the client in 3 copies for the archive, monastery and construction management.

Discussion

No conservator is employed in the monastery and the objects may be shown around to the public on different occasions by people who are not familiar with this type of material. For this purpose, window mats and Mylar sleeves provide a good protection if the objects are touched only at the mat frame. The objects are however endangered by careless handling, as they float in their mount frames on local hinges. The protective polyester sleeves may create a micro-climate and increase the risk of mold growth on water damage or may melt on to the objects in case of fire. On rubbing the surface, the Mylar may create electrostatic attraction to the objects. Newer archive regulations demand the separation of organic material and plastics.

The water free Klucel may not be as strong an adhesive as animal glue, but the degraded areas of the parchment were to be protected from water based adhesive and strong tension. Japanese paper as filling and reinforcing material proved to be a good choice with respect to the handling and type of damage.

Conclusion

The parchment, leather and paper fragments are cleaned of surface dirt and incrustations, stabilized and preserved in mounts and protective sleeves of archival quality, and housed in acid free folders and boxes for long term storage. The mounts are suitable for exhibition in standard size frames, the boxes meet the requirements of the storage shelves in the archive. Both sides of the objects are immediately visible to the interested. Use of archival materials provides a long term stability in an adequately controlled environment.

ICOM-CC Leather and Related Materials WG Offenbach 2012

Acknowledgments My special thanks is to Ines Jesche for help with the practical work. Wolfgang Riedel, Stiftung Kloster Eberbach and Hessisches Baumanagement I thank for entrusting me with this project and Roy Thomson for proof-reading this article.

Materials Archival matboard 0842, 240g/m2; www.klug-conservation.de Berlin Tissue 2g/m2, Kozo and Mitsumata fibers; [email protected]. Corrugated archival board; www.klug-conservation.de GoreTex ePTFE membrane; www.gmw-gabikleindorfer.de Gummed archival paper; www.klug-conservation.de Hollytex 31g/m2 polyester non woven fabric; www.gmw-gabikleindorfer.de Iris linen cloth; www.leos-nachfolger.de Iso propyl alcohol, isopropanol, 2-propanol; www.carlroth.com Japanese paper Ino Shi 30% Japan Kozo 70% Manila Hemp; [email protected] Japanese paper RK0 5g/m2, Kozo fiber paper; www.gmw-gabikleindorfer.de Klucel GF cellulose ether; www.gmw-gabikleindorfer.de Luganil paper colors; Braunwarth & Lüthke, Ickstattstr.3, 80469 München Matboard 1,6 mm H12; www.klug-conservation.de Mylar D75 polyester sleeve DIN A3; www.anton-glaser.de Wallmaster dry cleaning latex sponge; www.gmw-gabikleindorfer.de Wheat starch paste; www.gmw-gabikleindorfer.de

References Feller, R., Wilt, M. Evaluation of Cellulose Ethers for Conservation, Research in Conservation Series, Getty Conservation Institute, 1990, p. 94 Hassel, B. 2003. ‘Heat Damaged Parchment’, PapierRestaurierung 3 (4), pp.31-38 Hessenschau website 2012. Funde im Kloster Eberbach restauriert. http://www.hr- online.de/website/rubriken/kultur/index.jsp?rubrik=5986&key=standard_document_45584323&mediakey=fs/hessenschau/2012 08011930_Funde_im_Kloster_Eberbach_restauriert____2669&type=v Larsen, R., Poulsen, D.V., Vest, M. 'The Hydrothermal Stability (Shrinkage Activity) of Parchment Measured by the Micro Hot Table Method (MHT)', in: R. Larsen (ed.), Microanalysis of Parchment, Archetype Publications, London, 2002, pp. 55-62. Lascaux website 2012. Klucel : http://lascaux.ch/pdf/de/produkte/restauro/5_cellulosen_staerken_polysaccharide.pdf Thomson, R., Kite, M., ‘The Nature and Properties of Leather’, Conservation of Leather and Related Materials, Butterworth-Heinemann, 2007, p. 2 Biography Barbara Hassel runs her own conservation studio in Frankfurt/Main as a free-lance since 1998. Trained as a hand bookbinder, she specialized in book and paper conservation and led the book conservation department at the Badische Landesbibliothek Karlsruhe from 1986 to 1993. For five years she worked as head of workshop at the paper conservation study program at the Stuttgart Academy of Fine Arts. In 2001 she received her degree of Master of Science in Conservation in Copenhagen, Denmark with a research study on heat damaged parchment. Barbara Hassel, Book and Paper Conservation, Auf der Platte 20, D- 60435 Frankfurt/Main, Germany - Phone: 0049- (0)69-54840721. [email protected]

ICOM-CC Leather and Related Materials WG Offenbach 2012

The Restoration of the 17th Century Altar Frontal from the Oratory of Saint Dominic in Orvieto: Backing Reinforcement, Lining and Tensioning System According to Minimal Intervention

Sara Iafrate, Anna Valeria Jervis, Mariabianca Paris, Marcella Ioele, Laura D’Agostino

Abstract The Saint Dominic altar frontal from Orvieto originally decorated the oratory of Saint Dominic’s confraternity and is now part of the Cathedral’s Museum’s collection. It presents the traditional technique of gilt leather artefacts, with decorative patterns painted with pigments and lacquers in an oil medium. Conservation treatment was led by the Istituto Superiore per la Conservazione e il Restauro (ISCR) Leather artefacts laboratory, which formed an interdisciplinary research group to investigate technical, historical and conservative aspects of the altar frontal. The artefact’s poor conservation conditions required the study and the application of specific intervention procedures, inspired to reversibility and minimal intervention criteria.The leather support was particularly thin and structurally compromised by numerous losses and tears. Some areas were especially brittle and fragile. Conservation treatment focused on three main aspects, strictly related to one another: 1) Reconstruction of the artefact’s structural integrity, using different techniques and adhesives for backing reinforcement and infilling inserts, in order to introduce a “sacrificial layer” that would permit the removal of the insert without damaging the original support. 2) Reintegration of missing parts, mainly located along peripheral areas. 3) Providing a new supporting structure and an extensible tensioning system, that should be adequate to the poor structural condition of the leather. Observing the principle of minimal intervention, it was decided not to reline the artefact completely and to apply a loose lining consisting of a supple synthetic fabric.

Keywords Gilt leather, infilling insert, backing reinforcement, sacrificial layer, tensioning system.

Introduction

Gilt and painted leather artefacts, such as hangings, paintings and altar frontals, were widely used to decorate Italian churches between the 16th and the 18th century. Although many altar frontals are no longer in their original locations, a considerable number is still preserved today, and can be found in museum collections. The Saint Dominic altar frontal from Orvieto represents a significant example of this artefact’s category (Figure 1).

Figure 1: The Saint Dominic altar frontal from Orvieto, © ISCR, 2012.

A field survey of extant gilt and painted leather artefacts in the territory of Central Italy, conducted by the Leather conservation laboratory of ISCR in the 1990’s, provided useful information and helped classify our leather altar frontal in type, technique and in floral decorative patterns. Conservation treatment still in 43

ICOM-CC Leather and Related Materials WG Offenbach 2012

progress at ISCR’s Leather laboratory stimulated an interdisciplinary research, in order to thoroughly investigate technical, historical and conservation aspects of the altar frontal. Aesthetical and treatment choices were based on previous studies concerning conservation materials and adhesives, and specific methods based on principle of reversibility and minimal intervention.

Object Description

Altar frontals, or antependia, are a particular kind of liturgical furnishing widely used in the past to decorate the front of the altar. They could be made of different materials, such as stone, wood, precious metals, textiles and leather. The first examples of this kind of artefact date back to the 12th-13th century, and with the earliest altar frontals having a more narrative than ornamental decoration, based on the stories of the Bible.

In time structures and dimensions of antependia were codified, particularly after Counter-Reformation, when many books were written containing specific and pragmatic rules regarding liturgical life. Both for its dimension (97x214cm) and for its architectural structure, Saint Dominic’s altar frontal seems to follow the instructions written in one of the most important of these books, Instructiones fabricae et supellectilis ecclesiasticae, by Carlo Borromeo in 1577.

Our altar frontal is composed of a top trimmed on the lower border, two lateral wings and a central medallion bearing the image of Saint Dominic, to whom the altar was dedicated. The central panel is richly adorned by a floral pattern. The naturalism and free arrangement of the flowers are typical of 17th century’s decorations. It is also during this period that many of the oriental botanic species represented in the floral motifs, such as fritillaria and tulip, were first introduced to Europe (Masson G. 1970). The floral pattern is tooled in relief on a punched gold field imitating the ornate of more precious textile frontals that were used during solemn rites services. According to the precepts in Instructiones, instead, leather antependia were to be used only during ordinary liturgical services.

Historical Data The altar frontal originally decorated the oratory of Saint Dominic’s confraternity in Orvieto. The confraternity was suppressed with many other clerical orders after Italian Unification in 1861. According to the documents in the archive, the altar frontal became part of the Cathedral’s Museum collection in 1899 and since that time it had been in the museum’s storage.

Materials The altar frontal is composed of eleven pieces of leather cut into different sized squares depending on the layout of the decoration and sewn together. In order to correct natural defects and gaps in the skins created during the working process, they were patched and glued with inserts of gilded and painted leather. Other inserts were glued along the edges of the skins in order to regularize their rectangular shape. Spot tests on micro samples of collagen fibres were made in order to determine the tanning process. Results showed the presence of hydrolyzed tannins and free gallic acid that possibly derive from the hydrolyzation of condensed tannin. Tests for determining alum content were also carried out, resulting in only small quantities of the element, which could indicate that the leather tanned twice possibly to improve its mechanical properties.

SEM-EDX analysis on cross-sections revealed the presence of silver leaf applied to the surface of the leather and of two superimposed layers of gold varnish, mixed with lead pigments to help drying. The floral pattern on the gold varnish was painted using pigments and lacquers in an oil binder. Particularly interesting is the use of organic colours, such as indigo to paint the petals of the blue flowers and copper resinate for the leaves and stalks. The floral pattern is characterized by a decorative motif tooled in relief by pressing the gilded leather onto a wooden mould into which the negative of the decoration was carved.

Condition When the altar frontal arrived at ISCR Leather Laboratory, the leather support appeared very fragile and affected by different kinds of damage and deterioration processes. 44

ICOM-CC Leather and Related Materials WG Offenbach 2012

In order to study deterioration of the collagen fibres’ structure, measurements of the shrinkage temperature (Ts) were carried out, on samples taken from different skins of the artefact. The average value was between 44°C and 55°C, showing considerable destabilization of the fibrous tissue.

From a macroscopic point of view, the leather support is very thin when compared to other similar artefacts (0,4-0,6 mm thickness). There were many tears and losses of different sizes over the entire piece, caused by inappropriate storage, manipulation and by biological attacks. The leathers’ edges beyond the stitching were very thin and fragile and had many tears that caused the loss of small fragments of leather: moreover, many of the original leather inserts were detached. Fading of colours and loss of fragments of the grain layer were visible particularly along the edges and along the stitching, where mechanical stress was particularly heavy, and caused thinning and weakening in localised areas of the leather.

Stiffness and fragility in limited areas were also noticeable, caused by humidity and past interventions on the back of the artefact. During such interventions strips of canvas were applied with animal glue in order to strengthen thin and weak areas (Figure 2). Figure 2: Strips of canvas applied with animal glue during previous interventions, © ISCR, 1989.

The leather support was also affected by deformations created mechanically by crease lines in the frontal having been stored folded for over a long period of time. Other deformations were probably caused by shrinkage of the fibres: a damp patch on the back of the left side may suggest that the artefact came in contact with moisture. Moreover, inappropriate stitching during previous interventions caused other deformations.

Treatment / Results

Conservation treatment was mainly focused on regaining the artefact’s structural integrity. The choice of materials and methodologies was based on previous studies conducted by ISCR’s Leather Conservation Laboratory on materials and adhesives for leather structural repair.

Following removal of inappropriate stitching that restrained the leather’s movement, the artefact underwent localised and total humidification in order to eliminate principal deformations. Humidification treatment was carried out using a synthetic membrane (Gore-tex) permeable to water vapour and keeping the altar frontal flat under weights during the drying process. Structural repair and regaining the frontal’s complete planarity, was achieved first by removing the animal glue from the back of the artefact. It was decided to soften this layer of glue by swelling it with water whereby avoiding any mechanical pressure in areas where the leather support was very thin and fragile. To prevent excessive wetting of the support, a rigid gel was utilized, made of esopolisaccaride (Kalcogel), commonly used in paper conservation.

Considerable and extensive structural damages notwithstanding, it was decided to intervene minimally and to treat locally by applying backing reinforcements and infilling inserts rather than relining the entire frontal. The most appropriate material for localised backing reinforcement, chosen on the basis of previous studies (Iafrate S. et alii, 2011) was a polyamide non-woven fabric (Cerex 1.0) glued to the back of the artefact with Beva Film. The same materials were used to mend tears and to reinforce the edges of the leather weakened by the stitching and by consequential mechanical stress.

ICOM-CC Leather and Related Materials WG Offenbach 2012

Figure 3 (on the left): Thin and fragile indented borders of lacunae with tears and detached fragments, © ISCR, 2012 Figure 4 (on the right): “Sacrificial layer” of Cerex glued with Beva Film, © ISCR, 2012

Alum tawed leather was chosen for infilling inserts, for its mechanical and aesthetic characteristics (Iafrate S. et alii, 2011). Gluing leather inserts directly to the edge of losses, in this specific case, posed a few critical problems with respect to important conservative issues, first of all reversibility. The edges of missing areas often presented very thin and fragile indented borders with tears and detached fragments (Figure 3). These fragile conditions would not allow for easy removal of inserts without damaging the original leather during future possible treatments, and for this reason a “sacrificial layer” was created to sit between the alum tawed leather of the infilling and the original leather. This “sacrificial layer” consists of an appropriate backing material attached with a different adhesive than the insert, in order to permit separate removal if necessary.

More specifically, the “sacrificial layer” is made of a polyamide non-woven fabric (Cerex 1.0) adhered with Beva Film (Figure 4). This adhesive is water-resistant and reversible in non-polar solvents, and acts as a barrier against moisture when gluing the alum leather insert with a water-based adhesive. The Cerex also acts as a localised backing reinforcement for these particularly fragile parts around the losses. The leather inserts were thus glued with a water-based adhesive, a mixture of Evacon R (2 parts) and Tylose 4% solution in distilled water (8 parts) and pasted to the edges of the infilling insert (Figure 5). Some dynamometric tests were carried out on specially prepared samples in order to determine how the “sacrificial layer” could affect the resistance of the joint to shear strength. As shown in the graph (Figure 6), the “sacrificial layer” weakened the resistance of the joint so that its breaking load become lower than that of ancient leather, represented by the red line. Moreover, the presence of the “sacrificial layer” would permit the removal of the insert applying a low peeling strength, without need of humidity to weaken the resistance of the joint by swelling the adhesive.

Figure 5: Infilling insert of alum tawed leather, © ISCR, 2012

ICOM-CC Leather and Related Materials WG Offenbach 2012

Figure 6: Breaking load of alum tawed leather/ancient leather joint with and without interposing the sacrificial layer.

Not all the losses were infilled. Careful evaluation was made by considering their dimensions, where they were located and by trying to determine their possible aesthetical impact on the final general appearance of the frontal. Losses along the peripheral areas, where the artefact was previously nailed to the frame, were not reintegrated, preferring to maintain the fragmentary aspect of the whole and to avoid arbitrary reconstruction of the bottom edge, which is almost completely missing. Other losses were reintegrated to prevent mechanical stresses.

Once the frontal’s structural integrity was adequately regained, it was decided to construct a new supporting structure, a rigid panel, that would allow hanging the altar frontal without compromising planarity obtained during treatment, and would offer better protection to the frontal’s back. In fact, many documents make reference to frames onto which frontals were tensioned, such as Carlo Borromeo in his Instructiones, and a rigid structure would correspond to the artefact’s previous mounting, as it was probably originally affixed to a wooden frame.

The mounting of the artefact to the panel has been planned, but is yet to be carried out. A panel made of composite material acts as the new supporting structure. Considering the particular fragility and conditions of the leather support, a tensioning system that would avoid internal mechanical stress possibly causing renewed tears and structural damage was required. An extensible tensioning system was studied, that affixes the frontal to the new structure through a network of helical springs positioned along the perimeter of the panel (Figure 7).

This extensible system allows for controlled and homogeneous tensioning of the frontal, and not restrain the movements of the leather. The springs’ yielding value was chosen according to the tension applied to the artefact (40g/cm) to maintain planarity (Nimmo M. et alii, 1996). To facilitate leather movements and to eliminate friction between the panel edges and the artefact, a system of rollers on ball-bearings was placed along the perimeter of the panel (Figure 7). Figure 7: Section of the structure of the panel, the spring system and the loose lining.

ICOM-CC Leather and Related Materials WG Offenbach 2012

To connect the leather support to the spring system the frontal will be provided with a loose lining glued to the back of the leather and connected to the spring system. The fabric will be glued to the back of the artefact just along the borders with Beva film (Figure 7). Research was conducted in order to select a proper material for lining that would be both sufficiently resistant and extensible, and that would comply with the leather’s movements without causing tensions and deformations. A polyamide fabric made of hexagonal celled structures (tulle) was found to be a suitable choice. Tulle offers a rich choice in its range of colours, and has a pleasant aesthetical appearance that seems compatible with gilt leather. In addition to its aesthetical qualities, its suitability for conservation purposes is presently being determined through mechanical tests.

The particular hexagonal cells’ structure in the fabric seems to characterize its considerable extensibility. In this way the lining should be able to follow the movements of the leather without restraining it. Dynamometric tests are being carried out on specifically prepared samples in order to study the mechanical behaviour of this material exposed to different stress strengths. Mechanical tests in response to shear strength have already been conducted. As shown in the line graphs (Figures 8-9), tulle’s stress-elongation diagram is quite similar to leather’s: the first part of the line indicates an alignment of the hexagonal structure along the stress direction. In the following segment the gradient of the line changes and the yielding of the material decreases as fibres start to resist the load strength. In the last segment the gradient changes again becoming lower, because fibres start to break until complete rupture of the sample occurs. These tests also revealed that tulle, as well as leather, doesn’t have an isotropic behaviour, as its mechanical characteristics (yielding and breaking load) change according to the direction of load application. Anyway, its breaking load is similar to that of ancient leather.

Before lining, other tests will need to be performed on aged samples in order to have more information on the mechanical behaviour of polyamide tulle in time: shear strength tests after ageing and creep tests will need to be performed before and after ageing on prepared samples. Test results should be compared with mechanical behaviour of aged gilt leather, in order to have more information on the suitability of the fabric for conservation purposes.

Figure 8: Tulle stress-elongation diagram; load direction perpendicular to the hexagonal structure. Figure 9: Tulle stress-elongation diagram; load direction parallel to the hexagonal structure.

Conclusion

During this intervention we tried to apply the principle of “minimal intervention” throughout all phases. This guiding principle seems important as the artefact is particularly fragile and structurally deteriorated. Localised backing reinforcements, the introduction of a “sacrificial layer” in the joints and a controlled extensible tensioning system helped to avoid complete relining, reducing invasive treatments and improving the reversibility of the intervention.

ICOM-CC Leather and Related Materials WG Offenbach 2012

Acknowledgments The authors would like to thank Manuela Andreano, Alessandra Cannistrà, Antonio Iaccarino Idelson, Angelo Rubino, Ulderico Santamaria.

Materials Alum tawed calf leather; J. Hewit and Sons - 12 Nettlehill Road - Houstoun Industrial Estate, Livingston - West Lothian EH54 5DL, Scotland - Telephone: +44 (0)1506 444160 Cerex ®30, 34 g/m2 (non-woven nylon fabric); CEREX Advanced Fabrics - 610 Chemstrand Road – Cantonment - Florida 32533, USA - Telephone: 850-937-3321 [email protected] Evacon R (ethylene vinyl acetate) ; Conservation By Design, Timecare Works - 5 Singer Way - Woburn Road Industrial Estate - Kempston, Bedford, MK42 7AW - Telephone: (01234) 846300 info@conservation- by-design.co.uk Tylose MH300P (methyl hydroxyethyl cellulose); CTS Srl - Via Piave 20/22 - 36077 Altavilla Vicentina (VI), [email protected]

References Nimmo, M., M. Paris, L. Rissotto, P. Cappa, and F. Bonetti. 1996. «Tensioning gilded and painted leather », ICOM-CC 11th Triennial Meeting Preprints, Edinburgh, 1–6 September 1996, ed. J. Bridgland, 751–758. London: James & James. Iafrate S. et alii. 2011 «Research study on support materials and adhesives for the restoration of gilt leather: first results», ICOM-CC 16th Triennial Meeting Preprints, Lisbon, 19-23 September 2011 Borromeo C., Instructiones fabricae et supellectilis ecclesiasticae, libri 2, Milan, 1577 Masson G. 1970, <>, in “Arte illustrata”, n. 3

Biographies Sara Iafrate received her degree in paintings conservation in 2009 at the Istituto Centrale per il Restauro in Rome (now ISCR). In 2007 she received her triennial degree in Conservation of Cultural Heritage at Perugia University. Now she’s attending the Tuscia University in Viterbo to obtain her degree in art history. As a freelance conservator she worked with private and public institution such as ISCR in Rome, Venaria Reale in Turin and The Vatican Museums. L’Aquila, Italy - [email protected]

Anna Valeria Jervis received her degree in paintings conservation in 1982 at the Istituto Centrale per il Restauro in Rome (now ISCR). In 1994 she received her degree in art history at La Sapienza University in Rome. As a freelance conservator she participated in various projects, including the conservation treatment of Luca Signorelli and Gentile da Fabriano’s wall paintings in the Orvieto Cathedral. From 1999 to 2005 she worked in the Conservation Department of the Istituto Nazionale per la Grafica in Rome. Since 2005 Jervis has been working in the Leather objects laboratory of the ISCR in Rome, and in 2006 she was placed in charge of the laboratory. She teaches conservation theory and procedures at the ISCR School and university courses. Her publications include articles on the history and methodology of conservation and restoration and on the professional profile of conservator-restorers. Istituto Superiore per la Conservazionee il Restauro (ISCR), Rome, Italy - [email protected]

Marcella Ioele is director of the chemistry laboratory at the Istituto Superiore per la Conservazione e il Restauro (ISCR) in Rome. Following studies in Chemistry (1990), Ioele obtained a PhD in Chemistry (1994), from La Sapienza University in Rome. Until 2000, she continued her research work on mechanisms of organic reactions with post-doctoral fellowships at the Bologna CNR Research Area and at the Max Planck Institute of Mülheim (Germany). Since September 2000 she has been employed by ISCR as conservation scientist in charge of leather, paper, textiles and waterlogged wood departments. She has been involved in several research projects in the field of conservation chemistry and in national and international conservation workshops. Ioele is the author of circa 50 publications in national and international journals, books and conferences. She teaches undergraduate and Masters courses at the ISCR School. Istituto Superiore per la Conservazionee il Restauro (ISCR), Rome, Italy - [email protected]

Mariabianca Paris has a diploma in Painting Conservation at the Istituto Centrale per il Restauro in Rome (presently ISCR) in 1980, with a one year post-diploma course. Degree in Medieval and Modern Art History at “La Sapienza” University in Rome in 1986. Conservator at ISCR; among other duties, she has ongoing responsibility for research, training, treatments ad consultancy in the Leather Conservation Laboratory that she contributed to establish in 1989. 49

ICOM-CC Leather and Related Materials WG Offenbach 2012

Member of the ICOM-CC working group ‘Leather and Related Materials’ since 1990, she coordinated it in the period 2008-2011. She is the author of several technical publications. Istituto Superiore per la Conservazionee il Restauro (ISCR), Rome, Italy - [email protected]

Laura D’Agostino received her degree in art history at La Sapienza University in Rome in 1979. Since 1983 she has been working for the Ministery of Cultural Heritage as art historian. Since 2003 she works at ISCR as art historian and director. She is specialized in the study of historic techniques and artistic aspects regarding artefacts made of paper, leather and textiles. She is also specialized in the field of modern and contemporary art. [email protected]

ICOM-CC Leather and Related Materials WG Offenbach 2012

The Study and Conservation of a 12th Century Pair of Ceremonial Sandals with Arabesque Decoration from Castel Sant’Elia’s Collection of Liturgical Garments

Anna Valeria Jervis, Silvia Checchi, Antonella Di Giovanni, Stefano Ferrari, Maria Rita Giuliani, Marcella Ioele, Michael Jung, Marica Mercalli, Federica Moretti

Abstract This pair of ceremonial sandals from the 12th century is one of three that belong to the collection of medieval liturgical garments from Castel Sant’Elia, near Viterbo, Italy. This footwear, made of many different materials, such as leather, textile fibres, metallic elements and parchment, was probably produced in a Sicilian workshop by artisans directly influenced by Islamic culture. The shoes are similar in form and bear a striking resemblance in decoration to the Stavelot sandals, from Stavelot Abbey, today in Brussels’ Musées d’Art et d’Histoire. The preservation to the present day of objects made of such fragile materials is quite exceptional. For this reason, the Istituto Superiore per la Conservazione e il Restauro (ISCR) created an interdisciplinary work team to determine the conservation treatment approach, which comprised the study of the original materials and subsequent additions within their historical context. The footwear was thoroughly examined through optic and electronic microscopy and through a variety of analytical means such as XRF, FT-IR, SEM-EDS, and micro-chemical analysis. Conservation treatment included consolidation and cleaning. Further treatments to preserve their shape and protect them during storage, handling and transport are planned and discussed.

Keywords Pontifical sandals, pontifical, sandal, ceremonial, footwear, shoes, catholic, church, religious, medieval, 12th century, gilt leather, silk fibres, Castel Sant’Elia, liturgical, garment, leather, silk, XRF, FT-IR, SEM-EDS, Istituto Superiore per la Conservazione e il Restauro, pseudo-Kufic.

Introduction and Aim of the Project

In order to study and treat this pair of pontifical sandals (Figure 1), an interdisciplinary working group was created in the Istituto Superiore per la Conservazione e il Restauro (ISCR) which included the participation of historians, conservation scientists and conservators specialized in different materials (leather, textiles, metals, archeological organic materials). Figure 1: The pontifical sandals before conservation treatment (photo by Marcello Leotta, © ISCR, 2012) The purpose of the project was: - To extend the knowledge regarding these objects, described in recent studies as “footwear with pseudo-Kufic decoration”. Studies are being carried out in the context of Castel Sant’Elia’s collection of liturgical garments. Historical comparisons are also being made with other contemporary ceremonial sandals from other European collections. - To treat the footwear, to restore and maintain their shape and to protect the vulnerable organic materials from which they are made. One of the purposes of the conservation treatment is also to minimize stress whilst exhibited, stored, handled and transported.

Historical Background and Origin

Among the footwear in Castel Sant’Elia’s collection, this is the second pair being treated by the ISCR. A thorough study of the collection was recently carried out during the restoration of the garments. The artefacts are being treated to go on display in the new museum, which itself has been conceived according to modern 51

ICOM-CC Leather and Related Materials WG Offenbach 2012

conservation standards (Mercalli Checchi (ed) 2012). The survival and presence of three pairs of footwear in the collection is remarkable, and reflects the important role that the basilica and monastery of Sant’Elia had between the 12th and 13th centuries (Lalli 2012). These sumptuous examples, used during solemn occasions such as Pontifical Mass celebrations, probably date from this period.

What at first appears quite remarkable is the resemblance of these shoes to the pair from the Belgian 12th century Stavelot Benedictine monastery (Lalli 2006; Balace De Meûter 2006). Stavelot’s historical events could be useful to date Sant’Elia’s pair: it is known that in 1161 Stavelot’s abbot, Erlebald, received the privilege, usually consented only to bishops, to wear pontifical footwear from the anti-pope Victor IV. Such information helps to tentatively date the pair that is today in Brussels’ Musée d’Art et d’Histoire and, probably, also this Sant’Elia pair. Further research on potential connections between the two Benedictine monasteries (Stavelot and Castel Sant’Elia), and of both with Montecassino’s abbey, and also on the political role of their respective abbots, could reveal additional interesting information.

As with most of the collection, it is believed that these sandals were made in Sicily, by craftsmen directly influenced by Islamic culture.

Direct Observation and Scientific Analyses

In order to gain more information on the materials and their state of conservation, the footwear was thoroughly examined through optic and electronic microscopy and different analytical tests. Such tests included non-destructive examination by X-ray fluorescence (XRF) (Table 1); others required examination of micro-samples (Optical Microscopy (OM), infrared spectroscopy (FT-IR), scanning electron microscopy coupled with energy-dispersive X-ray spectroscopy (SEM-EDS), micro chemical analysis (MC) ) (Table 2). Note that, since it is not possible to tell the right shoe from the left, the less complete shoe (shown on the right in Figure 1) is referred to as “shoe A”, and the other “shoe B”. The test results are incorporated below within the discussion on this pair of objects.

Table 1: XRF results (counting Anode W, 33 KV, 0, 15 mA, acquisition 150 sec). Measure Description Results (a) (b) 1 Metal rivet PbL(1433) AgK(29) SnK(57) ZrK(773) (a) (b) 2 Metal rivet PbL(1032) AgK (281) SnK(87) ZrK(889) (a) 3 Metal rivet AuL(123) PbL(994) AgK (29) SnK(57) (b) ZrK(771) (a) 4 Metal rivet FeK(24) PbL(1461) AgK(94) SnK(70) (b) ZrK(739) 5 Gilt leather stripe of the snakes decoration CaK(44) FeK(57) AuK(246) PbL(30) AgK(27) (b) ZrK(791) 6 Gilt leather stripe of the snakes decoration CaK(38) FeK(42) AuK(199) PbL(36) (b) ZrK(775) (b) 7 Parchment CaK(45) FeK(79) AuK(57) ZrK(711) (c) (a) 8 Blue fabric , next to the metal rivet CaK(36) FeK(37) PbL(445) AgK(52) SnK(27) (b) ZrK(680) 9 Gilt leather of the open-worked decoration CaK(35) FeK(43) AuK(167) PbL(51) (b) ZrK(716) (b) 10 White skin that covers the sole CaK(21) FeK(19) ZrK(716) (b) 11 Undersole CaK(53) FeK(37) ZrK(710) (b) 12 Blue fabric on back part of the upper of shoe CaK(39) FeK(46) ZrK(638) B

13 Gilt leather on back part of the upper of shoe CaK(30) FeK(54) AuK(119) PbL(29) (b) B ZrK(686) (b) 14 Head of the rivet PbL(1223) AgK(46) SnK(61) ZrK(676) (b) 15 Side of the rivet PbL(914) AgK(42) SnK(41) ZrK(889) (b) 16 Part of the rivet inside the shoe FeK(38) PbL(1376) SnK(63) ZrK(682) 52

ICOM-CC Leather and Related Materials WG Offenbach 2012

(a) Sum of SnK and AgK (b) Instrumental (c) The measure spot includes the metal rivet

Table 2: Description of micro-samples and analyses Sample Location Analyses* 1 Gilt leather stripe of the snakes decoration (shoe A) OM, SEM-EDS, MC, FTIR 2 Upper (shoe B) OM, MC, Ts 3 Gilt leather of the open-worked decoration (shoe B) OM, SEM-EDS, MC 4 Insole (shoe A) OM, MC, Ts 5 Dirt deposit on the gilt leather of the open-worked decoration FTIR (shoe A) 6 Parchment (shoe B) OM, SEM-EDS, MC, FTIR, Ts 7 White leather on the profile of the sole (shoe B) OM, MC, Ts 8 Black material on the undersole (shoe A) OM, MC, Ts 9 Red fabric (separated fragment from shoe B) OM 10 Blue fabric (separated fragment from shoe B) OM, SEM-EDS, FTIR 11 Weft thread of the snakes decoration (shoe A) OM 12 Yellow thread (shoe B) OM 13 Pink-orange thread (shoe B) OM 14 Stitching thread of upper-sole-outsole (shoe A) OM * OM = optical microscopy SEM-EDS Scanning Electron Microscope with X ray microprobe FTIR Fourier Transformed Infrared analysis MC- Micro-chemical analyses for the identification of tanning materials Ts Shrinkage temperature

Description of the Different Parts, Materials and Construction

A study of the shoes reveals a high level of technique and refinement in their assembly. A description of the two main parts that constitute the sandals (upper and sole) can be made starting from the external elements of the upper: - Rivets, - Gilt leather decoration, - Parchment support of the snakes’ decoration, - Blue silk fabric, - Structural part of the upper.

The Rivets Numerous rivets are distributed across the entire surface of the upper. Each rivet is formed of two parts: one is placed on the inside of the upper, and the other is in the form of a head that was probably assembled by beating it into the other part. The XRF has shown that the part placed inside is made of a tin and lead alloy, while the head shows the presence of silver. By means of the optical microscopy it was possible to observe what seem to be silver fragments on the surface of the heads. SEM examination would give a more certain response regarding the presence of silver as external plating or as a component of the alloy, but we would have had to remove one of the rivets from the shoe, and such operation has been considered unnecessarily destructive.

The Gilt Leather Decoration The leather decoration, gilded with gold leaf, is cut from a very thin vegetable-tanned skin, probably split or thinned, gilded and subsequently open-worked. Complex needlework secures this layer to the upper with two different colors of silk thread: yellow and orange. The blue silk fabric is thus sewn between the gilt layer and 53

ICOM-CC Leather and Related Materials WG Offenbach 2012

the upper. The stitching to the gilt leather open-work creates the shoe’s decorative pattern which shows up against the blue silk fabric. Such complex design leads one to conclude that the whole decoration was planned and drawn in detail before its execution.

Joseph Braun, who first studied these objects (Braun 1899), recognized “an Arabian inscription with great Kufic” letters in the decoration of the back part of the upper of shoe B (Figure 2), and was confirmed by other scholars, until Giusy Lalli (Lalli 2006) for the first time redefined its pseudo-Kufic character. Recent conservation treatment enabled a more precise examination of the decoration. As a result, we find it at present more appropriate to define the decoration as arabesque decoration (and not as a pseudo-Kufic inscription).

Figure 2: Drawing of the decoration of the back part of the upper Figure 3: The snakes’ decoration on the front part of the of shoe B (graphic documentation by Federica Moretti, © ISCR, upper of shoe B (photo by Marcello Leotta, © ISCR, 2012) 2012)

Gilt leather was used in decorating shoes for high ranking persons in Coptic and Byzantine cultures, following a tradition that Arab-Sicilian craftsmanship contributed to transmit to Europe. The technique of leather gilding with gold leaf is described in European medieval sources, such as the Mappae Clavicula and Theophilus’ Diversarum Artium Schedula (Jervis et alii 2010).

XRF proves that gold leaf was used for gilding; SEM-EDS analyses show that 10 to 15 percent of silver is present: gold was often used in alloy with silver. It must be remembered that gold leaf was usually obtained by beating gold coins; studies have been conducted on the gold/silver alloys of Islamic coins in medieval time (Jung et alii 2012).

The decoration on the front part of the upper represents two interlocked snakes (Figure 3). A very similar one is visible on Stavelot’s pair. The completeness of these sandals allows us to presume that, on our pair as well, the interlacement of the two reptiles once continued onto the frontal strip of the upper, now missing, which probably extended to the instep. Such decoration is composed of a fabric where the weft and the warp are made respectively of silk thread and very thin strips of gilt leather. Once produced, this element was sewn onto the upper using different coloured, degummed silk threads to characterize the two snakes: yellow for one, and probably orange, or pink-orange, for the other, where the different colour tones of the two snakes originally showed up clearly. The thread that today appears beige (but is pink-orange on the inside of the upper, where it was protected from light) is the same one that secures the whole gilt leather decoration to the shoe.

Both the gilt leather and the leather of the shoe upper are vegetable-tanned, as evidenced by the micro- chemical tests for the recognition of tannins. A micro sample of gilt leather examined by means of SEM-EDS determined the presence of aluminium as well; this leads to assume that the leather was tanned twice.

ICOM-CC Leather and Related Materials WG Offenbach 2012

The Parchment Support of the Snakes’ Decoration This decoration, manufactured separately, was stitched by means of the weft silk thread to the edges of a semi-rigid parchment support; the reinforced decoration was then stitched to the shoe. Pieces of parchment used as support can frequently be found in historical embroideries, garments and clothing.

The SEM-EDS analysis of a parchment micro-sample shows the presence of calcium and considerable amounts of aluminium, potassium and silica that could be residues of pumice stone. According to traditional techniques, the parchment was stretched and scraped with special half-moon blades and pumice stone. It is otherwise possible that the embroidery support could have been made re-using an old piece of manuscript or drawing; in this case some sort of ground layer with a mineral content, e.g. white clay, could have primed the surface.

The Blue Silk Fabric The whole upper was covered with a plain silk samite, whose blue colour was certainly obtained with indigo or woad and sewn between the gilt layer and the upper.

The Upper As in Stavelot’s pair, this one was ankle high, ending with buttonholed strips, with a leather or textile string securing them around the ankle. This shape can be found in different ceremonial shoes of the same period, like those in Vienna’s Weltliche Schatzkammer, extensively re-worked, probably for the occasion of Emperor Ferdinand II’s coronation in 1619. Similar are also those of Trier’s Bishop Arnold I, who died 1189 (Göpfrich Dreyspring 2005).

The upper is shaped assembling different pieces of very thin vegetable-tanned goat leather, whose juxtapositions are hidden by the silk and gilt leather layers. To our surprise, for some reason no traces of lining were found: the inside of the upper is thus visible, and the reverse of the silk seams as well (Figure 4).

Figure 4. Inside of the upper of shoe A (photo by Marcello Leotta, © ISCR, 2012)

Insole, Sole and Outsole The sole is made of a thick piece of cork that was lined, on the outside, with two different layers: the first made of an alum-tawed leather that was once covered, along the edge, with red silk taffetas of which only very small fragments remain.

The outer sole is of vegetable-tanned leather, coloured with a thick black layer. This element is very likely the original outer sole, as the seam holes seem to testify that it was never replaced. Examination by optical microscopy revealed that the thread is made of hemp fibers. On the outer sole small holes are visible, where wooden pegs were once inserted to secure it to the sole.

Among Castel Sant’Elia’s garments the presence of a mitre, with silk fabric and metal thread appliqués should be highlighted. The blue and red fabrics of its decoration, respectively a blue samite and a red taffeta, are very similar to those on these shoes. It is nevertheless difficult to confirm if they are actually the same fabrics used on both artefacts. Period and region of manufacture possibly match, so that this hypothesis seems plausible, although it must be corroborated by further elements.

ICOM-CC Leather and Related Materials WG Offenbach 2012

Conservation Condition

Large areas of the upper are missing on both shoes, which created difficulties in identifying the original shape of the sandals. The correct reading of the decoration too, has been hindered by the folds and creases of the back, still in place on shoe B. The high fragility of the organic materials must be taken into account, as the survival of similar objects through the ages is certainly rare.

The Leather Parts The Ts (shrinkage temperature) was tested on leather micro-samples from different parts of the sandals, and its values have confirmed the poor condition of the leather. The gilt leather (samples 1 and 3) appears to be already contracted. Through SEM-EDS examination of the sample from this layer it was possible to observe the gold leaf, which appears to be extremely thin and fragmented.

Among the different leathers, that of the upper (sample 2) seems to be better preserved, with a Ts around 49°C. Ts of the alum-tawed leather (samples 4 and 7) is considerably lower (between 35° and 43° C), but this is in part explained by that particular kind of tanning (Table 3).

Table 3. Characterization of tanning materials and shrinkage temperature (Ts) Location of sample Presence of Description of sample Ts vegetable tannins 1 Gilt leather stripe of the snakes YES Thin fibres in fairly good ND decoration conditions 2 Upper YES Thin fibres in fairly good 49 °C conditions 3 Gilt leather of the open-worked YES Thin fibres in fairly good ND decoration conditions 4 Insole NO Short fibres No contraction above 35°C 6 Parchment NO Short fibres stuck Fairly perceptible together probably already contraction at 45°C contracted 7 White leather on the profile of the NO Thin fibres separated one 43°C sole from the other 8 Undersole YES Short fibres probably Fairly perceptible already contracted contraction at 44°C ND = not determined

The Silk Fibres All silk fibres are very deteriorated; the yellow seam is probably the one in better conditions. On the outside of the upper, they appear more deteriorated and discoloured than on the inside, which is more protected from light and abrasion. The thread that today appears pink-orange (Figure 5), in particular, was certainly darker and more vivid in colour, possibly even bright red. The fragment of the blue samite observed with OM and SEM denotes conspicuous degradation and localized discoloration of the fibres (Figure 6). On the same fabric, the FT-IR spectra shows a particular shape that is typical of silk oxidation (Figure 7). Of the red taffetas only tiny fragments are still in place, hidden in the recesses between the stitches (Figure 8); on this fabric as well OM shows conspicuous degradation of the fibres. Examined samples of the blue samite and of the red silk fabric were taken from fragments found already detached in the shoe’s box.

ICOM-CC Leather and Related Materials WG Offenbach 2012

Figure 5: Micro-optical image Figure 6: Micro-optical image of the silk pink-orange thread (© ISCR, 2012) of the silk blue samit (© ISCR, 2012)

Figure 7: Micro FT-IR analysis of the silk blue samite (© ISCR, 2012)

Figure 8: Micro-optical image of the red taffetas once Figure 9: Detail of the tin rivets (© ISCR, 2012) covering the profile of the sole (© ISCR, 2012)

ICOM-CC Leather and Related Materials WG Offenbach 2012

The Rivets The head of the rivets shows advanced corrosion that has produced a black and opaque colour. This is particularly dark on the peripheral areas of the heads, where fragments of silver leaf are still in place, while in the middle, where the silver leaf is mainly missing and the tin/lead alloy directly visible, the surface shows a greyer tone and a structure that appears more evidently coarse-grained than where the silver is present (Figure 9).

Superimposed Layers Examination by optical microscopy showed a thin layer of greasy dust and dirt that covers the whole surface of the artefacts, and it appears to be thicker in the interstices and gaps. Dark stains are clearly visible on the alum-tawed leather of the insole. FT-IR examination confirmed the presence of a fatty component in the dust/dirt layer, and that the stains on the insole are composed of beeswax.

We have information regarding two previous conservation treatments: the first dates from the beginning of the 20th century, while the second was conducted in 1956 (Mercalli 2012). It is possible that the presence of paraffin wax and animal glue, identified by FT-IR on some spots of the upper, refer to materials used during such treatments.

Conservation Treatment

As much information as possible regarding these sandals was documented by use of graphics. Conservation treatment is still in progress, and is expected to finish by Fall 2012. The principal goals of the treatment are:  To secure fragile parts and elements that were at risk of loss;  To clean and remove deposits, fatty dirt and stains;  To restore and maintain the correct form of the upper;  To prepare a support that can: o sustain and maintain the correct shape; o facilitate handling and transport, minimizing direct contact and other possible risks.

Fragile and detached parts of the gilt leather layer were supported with woven-unwoven polyamide fabric (Cerex), using Beva Film as an adhesive. Tears were mended by adhering the adjacent margins with a cellulose ether (Tylose MH300P) diluted in water to an 8% concentration.

Low-pressure vacuuming was performed on both shoes. Although dirt deposits were rather thick in certain areas, it was chosen not to carry out this operation thoroughly everywhere, as certain parts were too fragile. The gilt leather decoration was cleaned with a mild detergent solution (deionized water 99,4%, Tween 20, non-ionic surfactant 0,2%, CMC carboxymethylcellulose 0,2%). The non-ionic surfactant is intended to bring organic grime into solution, while CMC helps hold it in suspension while also minimizing the amount of water absorbed by the object. This operation was performed under the optical microscope, using very small cotton swabs and being careful not to wet the silk fibers of the seams (Figure 10).

Figure 10. Detail of the gilt leather open-work after cleaning (photo by Marcello Leotta, © ISCR, 2012) 58

ICOM-CC Leather and Related Materials WG Offenbach 2012

The cleaning of the rivets was also carried out with tiny swabs under the microscope, removing first dirt and fat with an ethanol and acetone mixture. Subsequently, in order to reduce the corrosion layer, a DPA solution (diethylenetriaminepentaacetic acid) was used in a 5% water solution, using sodium hydroxide to bring the pH value to 7 and then removing it with ethanol.

During the entire treatment, it was necessary to proceed with eliminating local distortions, especially in order to mend tears accurately. Local humidification was then performed, using Gore-Tex membrane and moist blotters, and small temporary supports clamped together with magnets.

It was decided not to compensate gaps and losses, as the incompleteness of these objects is to be taken into account and preserved in its historical meaning. Nevertheless, maintaining the upper in its correct position has both a conservation as well as an aesthetic purpose.

The stage of conservation treatment that addresses the shoes’ shape and protection is currently in progress, but some decisions have already been considered and made. One option considered was to put some sort of filling inside each shoe, in order to suggest the shape of the foot. This solution, however, would interfere with the visibility of the reverse of the upper and of the seams, which is clearly visible with the missing lining. A different solution was then sought. The tip of the uppers is to be filled with a soft filling, not visible from the outside. The back part that survives in shoe B needs a support to remain upright, as the leather is very soft and thin. For this part, a transparent thin molded support made of Vivak has therefore been proposed, of exactly the same shape as the original, to be placed against the inside. This element is to be fixed to a tray on which the sandals will be placed. The junction between support and tray has been thought in such a way that this would also become a useful protection during transport, as the tray itself can be manipulated without touching the shoes.

Conclusion

After finishing the restoration of this pair of liturgical sandals, the Leather Conservation Laboratory of ISCR is going to restore also the third pair in Castel Sant’Elia’s collection. We hope that completing the conservation treatment of this small group of historical shoes, and continuing its study, could bring some new piece of information to the knowledge of medieval footwear and liturgical garments.

Acknowledgments The authors wish to thank Laura D’Agostino, Jutta Göpfrich, Marcello Leotta, Marina Marchese, Mariabianca Paris, Lidia Rissotto, Valentina Rossi Sagaria. Where not otherwise stated, photographs are by the authors.

Materials Beva Film T (ethylene vinyl-acetate copolymer):CTS Srl, Via Piave 20/22, 36077 Altavilla Vicentina (VI), Italy; [email protected] Cerex ®30, 34 g/m2 (non woven nylon fabric): CEREX Advanced Fabrics, 610 Chemstrand Road, Cantonment, Florida 32533, USA. Telephone: 850-937-3321. Fax: 850-968-0688. [email protected] Tylose MH300P (methylhydroxyethylcellulose): CTS Srl, Via Piave 20/22, 36077 Altavilla Vicentina (VI), Italy; [email protected] Tween 20 (non-ionic detergent): CTS Srl, Via Piave 20/22, 36077 Altavilla Vicentina (VI), Italy; [email protected] CMC (carboxymethilcellulose): Bresciani Srl, via Breda 142, 20126 Milano, Italy; [email protected] Gore-tex© (expanded polytetrofluoroethylene): W.L.Gore & Associates Inc, Elkton, MD 21921, USA. Diethylenetriaminepentaacetic acid: Fluka Analytical, Menichelli Tito s.r.l, Via Acuto, 30, 00131 Roma, Italy; [email protected] Vivak© (thermoplastic copoliester sheet): Bayer MaterialScience S.p.A., 05035 Nera Montoro, Italy; [email protected]

References Braun, J. 1899. «Der Paramentenschatz zu Castel S. Elia», Zeitschrift für Christliche Kunst, X, pp.291-301. Göpfrich, J., Dreysping, B. 2005. «Fabulous beasts - leather, silk and gold: recent research on and conservation of 12th century footwear from the Episcopal tombs in Trèves Cathedral», ICOM Committee for Conservation 14th Triennial Meeting, Le Hague. 2005. Preprints, London, I, pp. 243-248. 59

ICOM-CC Leather and Related Materials WG Offenbach 2012

Balace, S., De Meûter, I. 2006. «Sandali liturgici», Nobiles Officinae. Perle, filigrane e trame di seta dal Palazzo Reale di Palermo, I, Catalogue of the exhibition, M. Andaloro, Palermo (ed.), p. 258. Lalli, G. 2006. «Sandali liturgici con iscrizioni pseudo cufiche», Ibid., pp. 254-257. Jervis, A.V., Giuliani, M.R., Ioele, M., Jung, M., Mercalli, M., Moretti F. 2010. «Stepping across the Mediterranean: Conservation of a pair of pontifical sandals of the thirteenth century», in IIC Triennal Meeting, Istanbul, pp.113-119. Jung, M., Moioli, P., Pierdominici, F., Seccaroni, C. 2012. «Techniques and pigments used for the wall paintings of the Masgid-I Jom’e at Isfahan. A first preliminary review», Proceedings of the 7th International Congress on the Archaeology of the Ancient Near East, II, Wiesbaden, pp.405-423. Jervis, A.V., Jung, M., Moretti, F. 2012. «I sandali pontificali», I paramenti liturgici di Castel Sant’Elia. La loro storia e la cronaca del restauro, Marica Mercalli e S.Checchi (ed.), Rome, pp.88-99. Lalli, G. 2012. «I paramenti liturgici di Castel Sant’Elia», I paramenti liturgici di Castel Sant’Elia. La loro storia e la cronaca del restauro, Ibid., pp.48-55. Mercalli, M. 2012. «I preziosi paramenti. Storia di una collezione e della sua conservazione», Ibid., pp.18-29.

Biographies Anna Valeria Jervis received her degree in paintings conservation in 1982 at the Istituto Centrale per il Restauro in Rome (now ISCR). Since 2005 Jervis has been working in the Leather objects laboratory of the ISCR in Rome, and since 2006 she is in charge of the laboratory. [email protected]

Silvia Checchi received her degree in paintings conservation in 1989 at the Istituto Centrale per il Restauro (now ISCR). Since 2001 she has been working as conservator in the Textile objects laboratory at the Rome ISCR. silvia.checchi@beniculturali

Antonella Di Giovanni received her degree in objects conservation in 1995 at the Istituto Centrale per il Restauro (now ISCR). Since 2007 she has been working at the ISCR where she conducts research, directs and participates in conservation treatments of artefacts for museums and institutions. [email protected]

Stefano Ferrari received his degree in objects conservation in 1995 at the Istituto Centrale per il Restauro (now ISCR). Since 2002 he has been working at the archaeological site of Mozia (Sicily) for the Department of Historical, Archaeological and Anthropological Studies of Rome’s University La Sapienza. [email protected]

Maria Rita Giuliani received her degree in biological sciences at La Sapienza University in Rome. She is director- biologist at the Rome ISCR. She gained experience on biology applied to the preservation of organic materials of cultural heritage. [email protected]

Michael Jung is curator of the Islamic section of the Museo Nazionale d’Arte Orientale of Rome. He has participated in numerous projects in Islamic countries and has published contributions on Islamic art and architecture and on the Museum’s collections. [email protected]

Marica Mercalli received her degree in art history at La Sapienza University in Rome. Since 1984 she has held a permanent position working for the Ministry of Cultural Heritage. Since 2012 Mercalli was appointed Superintendent for Historic and Artistic Works of the provinces of Venice, Belluno, Padova and Treviso. [email protected]

Federica Moretti has received her degree in paintings conservation at ISCR School in 2003. Her professional qualifications and experience cover several types of artifacts: stone objects, mural paintings, panel and canvas paintings. [email protected]

ICOM-CC Leather and Related Materials WG Offenbach 2012

Transformation of Collagen into Gelatine in Historical Leather and Parchment Caused by Natural Deterioration and Moist Treatment

René Larsen, Dorte Vestergaard Poulsen Sommer, Kathleen Mühlen Axelsson, Steen Kristian Frank

Abstract It is known that, for parchment and leather tanned with condensed types of vegetable tannins stored under acid conditions, chemical deterioration may lead to a transformation of the collagen into a gelatine-like substance that may dissolve in water even at ambient temperature. In a previous study, eight major morphological features of the breakdown process have been characterized as typical and representative for degraded historical parchments. In most cases, one or more of these characteristics of damage are present to various degrees in a sample, and often even in a single fibre sample, reflecting the complexity of breakdown. Damage was most frequently observed as a process of unfolding into flat bands, splitting, forming of pearls on string-like structures and/or fraying of the fibres. In some cases, this developed further to a fragmentation of the fibres and finally to a gelatine-like substance that may dissolve on contact with water or even at room temperature at relative humidity levels normal used for storage in archives, libraries and museums. Morphological transformation of fibres upon wetting at room temperature can be observed partly by the naked eye, and characterized more fully by microscopic examinations, as well as by hydrothermal shrinkage activity measurements. Examinations of degraded historical parchment documents and leather objects show that the gelatinisation may take place as part of the natural deterioration. However, especially parchment documents which have undergone conservation and restoration water-based treatments have been co-related with the frequent presence of some level of gelatinisation damage, especially in the case of degraded parchment, a finding which is consistent with the fact that direct exposure to water can cause gelatinisation of such degraded fibres. Caution is needed with water or alcohol based-treatments since they pose a risk of accelerating the morphologic transformation of degraded collagen fibres, especially in the case of parchment.

Keywords Leather, parchment, collagen fibres, morphology, gelatine, gelatinisation, denaturation, shrinkage temperature, treatment, water, moisture, deterioration, damage

Introduction

Previous experiments and studies have shown that the chemical mechanisms causing deterioration of the collagen structure in leather and parchment are acid hydrolysis and oxidation and that the degree of deterioration can be correlated to the hydrothermal stability of the material.1,2 For deteriorated parchment and leather tanned with condensed types of vegetable tannins, moist treatment or acid polluted storage may lead to a transformation of the collagen into a gelatine-like substance that may dissolve in water even at ambient temperature. Based on visual, IR spectroscopy and X-ray diffraction studies, Weiner et al. were the first to report on gelatinised historical parchment. Their diffraction studies indicated a measured collagen to gelatine ratio varying from that of new parchment and down to a ratio comparable to new gelatine.3 Larsen et al. reported on a historical parchment that although degraded, the fibres of which were apparently intact, seemed to be in a pre-gelatine state making them dissolve immediately on exposure to water at room temperature.4

Dissolving of fibres from condensed vegetable-tanned (mimosa) leather in a pre-gelatine state at 40oC by heating in water has been reported.5 The leather was about sixty years old and heavily damaged by storage in the polluted London atmosphere. Antique Nubian and Egyptian leathers were found to have already gelatinised (they were already in the form of a hard, brittle glass-like material without fibre structure) and to dissolve immediately on contact with a drop of water, respectively.6 Together with the above mentioned studies on parchment, these observations clearly show that there is no direct relation between the age of the material and

 The present paper is a slightly revised version of that given with kind permission from the 5th Freiberg Collagen Symposium, where it was part of the program and proceedings (ISBN 978-00-039421-8). The Symposium was held on September 4-5, 2012, in Freiberg, Germany and was organized by Forschungsinstitut für Leder und Kunststofbahnen, FILK, Freiberg, Germany. More information can be found at http://www.filkfreiberg.de

ICOM-CC Leather and Related Materials WG Offenbach 2012

its hydrothermal stability that Burton et al. hoped to find.7 The impact of storage conditions and treatment of the leather and parchment is far greater and any correlation between the material and its physical condition may only be found for material of same origin and quality stored and treated under the same conditions through history. Even a short history may be disastrous to relatively high quality material.

As described the so-called pre-gelatine state of the collagen fibres may occur at a relatively high level of hydrothermal stability for both leather and parchment. Thus, a dissolving temperature of 40oC has been observed for deteriorated vegetable tanned leather. The pre-gelatine state of the fibre structure constitutes a serious problem in practical conservation and storage of cultural heritage leather and parchment objects. However, microscopic studies of the reaction of small fibre samples in water, in combination with micro determination of the hydrothermal stability, are relatively simple diagnostic methods for determining the degree of fibre degradation, and the leather’s sensitivity to moist treatments and even to room temperature at moderately humid environments in storage.

In this study, a proposed gradual sequence of gelatinisation in the form of morphological transformation of the fibres at microscopic level is described in detail for parchment samples. So far our studies have mainly focused on parchment fibres, as the problem of gelatinisation of this material is massive compared to leather. However, examples on gelatinisation of leather are also included.

Method

The leather and parchment fibres presented in this paper were prepared taking a sample of about 1 mm2. The sample was soaked in excess demineralised water for 10 minutes on a microscope slide with concavity. The fibres were carefully separated in the water with fine preparation needles and placed on a flat microscope slide in excess water, covered with a cover glass and examined under a light microscope using a magnification of 100 times. For observation of specific details, magnification up to 640 times was used.8

Results and Discussion

The Morphology of Collagen Fibres Transforming from Intact to Gelatine

The gelatinisation process seems to follow certain typical denaturising patterns that can be observed as characteristic morphological transformations of the collagen fibre structure at the microscopic level. Eight major characteristic morphological breakdown features have been identified as typical and representative for historical parchments in various degrees of deterioration: split, frayed, flat, pearls on a string, butterfly, bundles, gel like fragments and hard fragment structures which exceeds no or almost no reaction by heating in water .9 Figure 1 shows a possible sequence of transformation of intact fibres (Fig. 1a) to the solution of gelatine-like fibre fragments in water at room temperature. In general, the morphological transformation occurs as a process of unfolding into flat bands (Fig. 1b), splitting, forming of pearls on string-like structures (Fig. 1c) and/or fraying of the fibres. This may develop further into a fragmentation of the fibres into the butterfly structures and finally to a gel-like substance that may dissolve on contact with water. The gradually morphologic deformation is normally related to hydrothermal denaturation and modification of the amino acid distribution and.10-12 Moreover, during shrinkage measurements we have observed that the most abundant denaturation sequence is that the flat band-like fibres (Fig. 1b) are transformed into the pearls on a string structure (Fig. 1c).8 The latter may again fragmentise into the butterfly-like structure (Fig 1d). However, this has only been observed a few times in connection with the shrinkage measurement.

Hydrolysis and oxidation are likely to take place in the charged regions of the peptide chains that swell when in wet condition.13,14 These large, charged, bulky regions containing reactive amino acids like Arg and Lys of the molecules are located between the twists in the regions and form the “pearls” in the denaturised fibre at microscopic level.8 The chemical cleavage in these areas in combination with physical swelling, leads to unfolding, formation of “pearls on a string structure”, “butterfly structures” and other fragments observable at the microscopic level, finally leading to a reduced molecular weight distribution comparable to gelatine. As known 62

ICOM-CC Leather and Related Materials WG Offenbach 2012

gelatine is in general defined as a polymer with a molecular weight distribution from around 15,000 to around 250,000 which dissolves in hot water at around 40oC15, and studies of historical parchment using Steric Exclusion Chromatography (SEC) and SDS-PAGE and 2D-Electrophoresis confirms a similar molecular weight distribution of the soluble fraction of historical parchment.16,17

Figure 1: Possible (proposed) sequence of transformation of intact fibres to their dissolution as gelatine-like fibre fragments in water at room temperature: intact fibres in a solid rope-like structure (a). Unfolded, flat bands (b). Pearls on a string structures (c). Fragmentation with formation of butterfly structures (d). Gel like butterfly fragments that may appear sticky in water (e). Gel like fragments dissolving (f).

Examples of Gelatinisation of Historical of Leathers and Parchment

In the following we present examples on leather and parchment in different stages of gelatinisation, hoping that it may contribute to the diagnosis and characterization of the phenomena. Figure 2 shows parts of an ancient Nubian quiver, dating c. 370 AC – 410 AC, in a pre-gelatine state dissolving by contact with water at room temperature (Figure 2 top)6; and a Coptic shoe sole leather, c. 1336 BC -1279 BC, which has completely transformed into a brittle glass-like structure (comparable also to classic animal glue) (Figure 2 bottom).

Figure 2: Top: Ancient Nubian quiver: (Left) the leather is very brittle and the fibres are in a pre gelatine state, which dissolve (Right) by contact with water at room temperature. Bottom: Coptic shoe sole leather: (Left) The one end is completely gelatinised, hardened and brittle; (Right) the cross section of the broken part shows that the whole structure has transformed into a glass-like non fibrous structure. 63

ICOM-CC Leather and Related Materials WG Offenbach 2012

The degraded ancient leathers that we analysed all had significant modified amino acid profiles (decreased content of Arg, Hyl, Hyp, Pro and Lys and increased content of Asp and Glu as well as the presence of breakdown products in the form of new amino acids like β-ala and γ-amino butyric acid ) indicating a high degree of oxidative deterioration.14,18 Thus, the chemical deterioration, acid hydrolysis and oxidation (as indicated by the mentioned forms of breakdown products), may result in the cleavage of the peptide chains to fragments of the size of gelatine fibres. The chemical process of gelatinisation of leathers may take place gradually and under normal storage conditions of room temperature and relative humidity as in the case of the Coptic shoe sole leather shown in Figure 2 (bottom). As for this leather, the gelatinisation has been observed starting in the one end (probably the most deteriorated part of the leather) where the fibres are transformed first into a dark or black, sticky mass which then dries into the glass-like structure. Following this a new area of fibres next to the glass area becomes sticky and then dries. Over time this can be observed as a movement of the sticky zone over the surface area of leather with the glass area becoming larger and the fibre area becoming smaller till the leather has been completely transformed into a glass-like solid glue structure. It should be noticed that this leather has not been exposed to or treated with excess of water. The gelatinisation has taken place and can be observed to develop during storage.

Like for the leather the chemical degradation of parchment is caused by hydrolysis and oxidation and in addition to Arg, Hyl and Lys, the amino acids most sensitive to oxidative modifications in parchment are Tyr and His.11 Furthermore, we have observed that gelatinisation of parchment seems to take place gradually, often starting in and around remaining hair holes if these are present (normally these are shaved away especially on parchment for writing purposes where an even and smooth surface is wanted) also in the case of parchments which have not been treated with moisture or show sign of having been exposed to excess of water.12,19 Figure 3a shows an example of this type of initial surface gelatinisation, with the follicle and its nearest surrounding appearing yellower. Observation of the surface and of the sampled yellow substance under the microscope, dry and in water, show no signs of fibre structure and in case with Domesday book, Codex Sinaiticus and other newer parchment documents studied, the gelatinised substance bears the ink or paint layers which may gradually be lost by flaking. 9,12,19, 20 A known phenomenon is gelatinisation of the overlap zone between the new restoration and the original parchment which may occur only a few years after the treatment (Figure 3b) The observational and statistical studies of these more than two thousand historical parchment in the form of documents and book bindings of different age, history and origin have revealed that the use of water during conservation or restoration treatment may initiate or accelerate the gelatinisation drastically within a few decades or maybe less like for example in the case of Domesday book the 887 manuscript leaves of which were restored and rebound in 5 volumes. It should be remarked that neither we nor the many colleagues that we have discussed this phenomena with have experienced serious yellowing of the parchment surface and overlap zones between the original and new restoration parchment immediately in connection with the restoration and humidification in connection with flattening. The partly gelatinisation, for example in connection with hair holes, may spread with time and may lead to loss of text and paint layers when the inflexible gelatine surface cracks and falls off (Figure 3c). Over time the full surface may have gelatinised and the gelatinisation may have proceeded into the inner structure leaving the whole parchment in a non fibrous gelatine form as shown at this parchment fragment originating from the 17th century (Figure 3d). Figure 3a-c is from parchment leaves of the Codex Sinaiticus dating from the 4th century.20 The rate at which the gelatinisation spreads depends on the storage conditions and treatment as the study of the vegetable tanned leathers from the British long term trial and accelerated ageing of parchment clearly shows.12, 21-24 Thus, heavy pollution and/or oxidation in the form of heat, light and oxidative pollutants etc. as well as high relative humidity most probably speeds up the chemical processes leading to the gelatinisation or a molecular condition (pre gelatinisation) that leads to immediate dissolution and transformation into gelatine by exposure to the critical amount of moisture and heat. The latter seems to be dependent on the degree of chemical degradation (fragmentation) of the collagen and can as described take place spontaneously over time at room temperature and maybe at relative low RH (accelerated ageing experiments have shown gelatine formation on parchment surfaces at 80 % RH at 80oC and at 75 % RH in 400 mg/m3 acetic acid vapour at room temperature23,24).

ICOM-CC Leather and Related Materials WG Offenbach 2012

Figure 3: (a) Gelatinisation of parchment: example of initial surface gelatinisation starting in remaining hair follicles and characterized by an enhanced yellow colour. (b) Gelatinisation of the overlap zone (arrow) between the new restoration (on the left) and the original parchment (on the right. (c) The gelatinisation has spread into the text which has started to fall off (arrow). (d) The remaining grain surface has gelatinised completely and fibres, apparently intact, are sticking out from the lower corium part of the parchment.

According to our experience, full gelatinised structures always dissolve on immersion in water between at around 40oC and down room temperature or even below. In dry condition these may appear as an already “dissolved” non-fibrous (glass like) structure or in what seem to be an intact fibre structure (what we define as the pre-gelatine state. Less damaged fibre fragments may become rubber-like and sticky in excess water. On the other hand we have not observed parchment developing sticky areas as an intermediate stage during normal storage like we have for ancient leather.

However, although gelatinisation seem to be a major route of degradation of parchment, it is important to note that in some cases the breakdown leads to the formation of brittle and hard fragments which exhibit a very low and slow hydrothermal activity which may continue at relatively high temperatures (around 60oC to 80oC).9 This may be due to the formation of a high extent of chemical cross links and/or that the intact structures in parts of these fibres and peptide chains left possessing shrinkage properties may be the hydrophobic regions (folding regions) of the collagen molecules.1 Moreover, this type of fragments (appears visually as the so called “red rot powder” which is easily scraped from the deteriorated leather) in general seems to be the most typical for very weak, heavily deteriorated vegetable tanned leather.6

Impact of Conservation and Restoration

Previous12 and more recent experiments with moist treatment (including starch paste) of deteriorated historical vegetable tanned leather and water and alcohol treatment of historical parchment ) have shown to result in morphologic transformation of the fibres in the form of unfolding, splitting and other denaturation characteristics

ICOM-CC Leather and Related Materials WG Offenbach 2012

indicating a progressing development towards gelatinisation immediately after drying of the samples. The samples were wetted with excess solvent or paste in an amount normally used in restoration and left to dry. Fibres from these were then assessed and compared with fibres from untreated reference samples according the procedure described above and showed significant morphological changes caused by the extra wetting compared to the references thus strongly indicating the potential harmful effect of moist and solvent treatments. These observations raises serious doubts that many used treatment methods may be more harmful than beneficial. It also shows us that there is a need for greater knowledge about the type and at which level of degradation the fibre structure is adversely affected by these treatments, immediately after application and whether there is a deleterious ‘hidden’ consequence that is manifested in the long term.

Conclusion

Studies have shown that the collagen fibre structure of historical leather and especially parchment may transform into a glue or gelatine-like substance by natural deterioration. Moreover, moist treatment and storage may accelerate the transformation process since these affect chemical and physico chemical reactivity of degraded collagen. In some cases, this seems to proceed via a pre-gelatine stage where the fibre structure may seem relatively intact making diagnosis of the condition of the fibre structure difficult. Microscopic studies of leather and parchment fibre morphology in water at room temperature combined with visual observation and measurement of their hydrothermal stability by the MHT method has shown to be relatively simple and effective tools for diagnosis. At present the methods are used in the EC MEMORI project as tools in the study of the impact of volatile organic acids on leather and parchment. 21 However, in order to improve the preservation of our cultural heritage leather and parchment in general, further studies and trials are necessary to achieve better and more accurate information about how different types and degrees of degradation and the associated morphological fibre characteristics respond to different treatments and storage conditions. Therefore, it is our hope that other colleagues also will take part in this development.

Acknowledgments We are very grateful to Frau Jutta Göpfrich, The German Leather and Shoe Museum in Offenbach, Gemany for supplying photos and micro samples; to our colleagues at the British Library in London, UK for supply of photos in connection with the Codex Sinaiticus project and to our colleagues and partners in the EC IDAP project at the Royal Library, Copenhagen, Denmark, for supply of photos. The many years of good cooperation with these and other fine colleagues have been essential for reaching the present state of knowledge on the deterioration of cultural heritage leather and parchment.

References 1. Larsen, R.: Experiments and Observations in the Study of Environmental Impact on Historical Vegetable- Tanned Leathers. Thermochimica Acta 365 (2000) (special issue: Cultural Heritage and Environmental Implications), 85-99. 2. Larsen, R.; Poulsen, D. V.; Odlyha, M.; Nielsen, K.; Wouters, J.; Puchinger, L.; Brimblecombe, P.; Bowden, D.: The use of complementary and comparative analysis in damage assessment of parchments. In: Micro Analysis of Parchment. Archetype Publications Ltd. René Larsen (ed.), London 2002, 173-175. 3. Weiner, S.; Kustanovich, Z.; Gil-Av, E. And Traub, W.: Dead Sea Scrolls parchments: unfolding of the collagen molecules and racemisation of aspartic acid. Nature, Vol. 281 (1980), 820-823. 4. Larsen, R.; Poulsen, D. P. and Vest, M.: The Hydrothermal Stability (Shrinkage Activity) of Parchment Measured by the Micro Hot Table Method (MHT). In: Micro Analysis of Parchment. Archetype Publications Ltd., René Larsen (ed.), London 2002, 55-62. 5. Larsen, R.: Evaluation of the Correlation between Natural and Artificial Ageing of Vegetable Tanned Leather and Determination of Parameters for Standardization of an Artificial Ageing Method: STEP Leather Project. European Cultural Heritage Newsletter on Research 7 (1993) 1-4, Special Issue, 19-26. 6. Larsen, R.; Vest, M.; Poulsen, D. V. and Kejser, U. B.: Determination of Hydrothermal Stability by the Micro Hot Table Method. In: Environment Leather Project. European Commission DG XII, Research Report no. 6., The Royal Danish Academy of Fine Arts, School of Conservation, René Larsen (ed.), Copenhagen 1996, 145-165. 66

ICOM-CC Leather and Related Materials WG Offenbach 2012

7. Burton, D.; Poole; J. B. and Reed, R.: A new approach to the dating of the Dead Sea scrolls. Nature 184 (1959), 533−534. 8. Mühlen Axelsson, K.; Larsen, R. and Sommer, D. V. P.: Dimensional studies of specific microscopic fibre structures in deteriorated parchment before and during shrinkage. Journal of Cultural Heritage 13 (2012) 128–136. 9. Larsen, R.: Introduction to damage and damage assessment of parchment. In: Improved damage assessment of parchment (IDAP), Assessment, data collection and sharing of knowledge. European Commission, Research Report no. 18, European Communities, René Larsen (ed.), Brussels 2007, 17-21. 10. Larsen, R.; Poulsen, D. V.; Minddal, K.; Dahlstrøm, N. and Fazlic, N.: Damage of parchment fibres on the microscopic level detected by the micro hot table (MHT) method. In: Improved damage assessment of parchment (IDAP), Assessment, data collection and sharing of knowledge. European Commission, Research Report no. 18, European Communities, René Larsen (ed.), Brussels 2007, 69-72. 11. Larsen, R.; Poulsen, D. V.; Minddal, K.; Dahlstrøm, N. and Fazlic, N.: Molecular damage of parchment studied by amino acid analysis. In: Improved damage assessment of parchment (IDAP), Assessment, data collection and sharing of knowledge. European Commission, Research Report no. 18, European Communities, René Larsen (ed.), Brussels 2007, 111-114. 12. Larsen, R.; Sommer D.V.P. and Mühlen Axelsson; K.: Scientific approach in conservation and restoration of leather and parchment objects in archives and libraries. In: New approaches to book and paper conservation-restoration, ed. Engel, P., Verlag Berger, Horn/Wien, 2011, 239-258. 13. H.T. Müller, E. Heidemann, Untersuchung der Gesetzmässigkeiten für den Säureabbau von Hautkollagen und identifizierung der Kollagenspaltstellen beim sauren Gelatineprozess. Das Leder 44 (1993) 69–79. 14. Larsen, R.: Fundamental aspects of the deterioration of vegetable tanned leathers. PhD thesis, University of Copenhagen, The Royal Danish Academy of Fine Arts, School of Conservation (Publ.), Copenhagen 1995. 15. Singh, S.; Rama Rao, K.V.; Venugopal, K. and Manikandan, R: Alteration in Dissolution Characteristics of Gelatin-Containing Formulations. A Review of the Problem, Test Methods and Solutions. Pharmaceutical Technology (2002), 36-58. 16. Juchauld, F. And Chahine, C.: Determination of the Molecular Weight Distribution in Parchment Collagen by Steric Exclusion Chromatography. In: Micro Analysis of Parchment. Archetype Publications Ltd., René Larsen (ed.), London 2002, 123-131. 17. Larsen, R.; Poulsen, D.V. and Vest, M. SDS-PAGE and 2D-Electrophoresis. In: Micro Analysis of Parchment. Archetype Publications Ltd., René Larsen (ed.), London 2002, 133-147. 18. Larsen, R.; Vest, M.; Poulsen, D. V.; Kejser, U. B. And Jensen, A.L.: Amino Acid Analysis. In: Environment Leather Project. European Commission DG XII, Research Report no. 6., The Royal Danish Academy of Fine Arts, School of Conservation, René Larsen (ed.), Copenhagen 1996, 39-68. 19. Nielsen, K: Visual Damage Assessment. In: Improved damage assessment of parchment (IDAP), Assessment, data collection and sharing of knowledge. European Commission, Research Report no. 18, European Communities, René Larsen (ed.), Brussels 2007, 45-51. 20. http://codexsinaiticus.org/en/project/conservation_parchment.aspx 21. Vest, M.; Jacobsen, J. and Larsen, R.: Accelerated ageing: effect of heat and relative humidity. In: Improved damage assessment of parchment (IDAP), Assessment, data collection and sharing of knowledge. European Commission, Research Report no. 18, European Communities, René Larsen (ed.), Brussels 2007, 67-68. 22. Odlyha, M.; Theorodorakopoulos, C.; de Groot, J.; Bozec, L. and Horton, M.: Thermoanalytical (macro to nano-scale) techniques and non-invasive spectroscopic analysis for damage assessment of parchment. In: Improved damage assessment of parchment (IDAP), Assessment, data collection and sharing of knowledge. European Commission, Research Report no. 18, European Communities, René Larsen (ed.), Brussels 2007, 73-87. 23. Della Gatta, G.; Badea, E.; Maṧic, A. and Ceccarelli, R.: Structural and thermal stability of collagen within parchment: a mesoscopic and molecular approach. In: Improved damage assessment of parchment (IDAP), Assessment, data collection and sharing of knowledge. European Commission, Research Report no. 18, European Communities, René Larsen (ed.), Brussels 2007, 89-51.

ICOM-CC Leather and Related Materials WG Offenbach 2012

24. Odlyha, M.; Bozec, L.; Hackney, S.; Colombini, M.P.; Bonaduce, I.; Di Girolamo, F.; Larsen, R.; Mühlen Axelsson, K.; Sommer, D.V.P.; Scharff, M.; Grøtoft, T.; Dahlin, E.; Chelazzi, D. and Baglioni, P.: Non- destructive and minimally invasive test methods for the evaluation and assessment of movable organic- based heritage materials. In Cultural Heritage Preservation. Proceedings of the 2nd European Workshop on Cultural Heritage Preservation (EWCHP-2012). Elin Dahlin (ed.), NILU – Norwegian Institute for Air Research Kjeller, Norway, 24th – 26th September, 2012, 12-19. 25. www.memori-project.eu

Biographies René Larsen is associate Professor at the School of Conservation in Copenhagen. Between 1996 – 2012 rector of the School of Conservation in Copenhagen. Has a background as a Professional Bookbinder, a Master in the Science of Conservation (1986) and a PhD in Biochemistry (1995). Has been coordinator of the following EC projects: STEP Leather; Deterioration and Conservation of Vegetable Tanned Leather (Environment Leather project); Micromethods for the Analysis of Parchment (MAP); Improved Damaged Assessment of Parchment (IDAP). Organizer of and tutor in several IDAP assessment courses. 2002 – 2003 member of the Committee for Physical Cultural Heritage for reporting to the Danish Parliament. 2004 – 2005 member of the working group regarding recommendations for handling, storage and exhibition of cultural heritage objects, the Danish National Cultural Heritage Authority. 2004 – 2009 member of the national working group on mass deacidification of paper, The Danish National Library Authority. Leading scientist in leather and parchment and Head representative for the School of Conservation in the ongoing EC project MEMORI.

Dorte Vestergaard Poulsen Sommer holds a Master of Conservation-Restoration Science (MSc) from the School of Conservation in Copenhagen. Has been assistant professor at the School of Conservation in Copenhagen since 1996. Has participated in the following EU research projects: Environment Leather project (Research Assistant), MAP (Research Assistant), IDAP (Researcher and assistant coordinator). Is currently working as a researcher in the ongoing EC MEMORI project. Co-organiser and workshop tutor at several IDAP courses in Europe.

Kathleen Mühlen Axelsson holds a Master of Conservation-Restoration Science (MSc) from the School of Conservation in Copenhagen. Currently working as a Research Assistant and PhD student on leather and parchment in the ongoing EC project MEMORI (Measurement, Effect Assessment and Mitigation of Pollutant Impact on Movable Cultural Assets - Innovative Research for Marker Transfer) at the School of Conservation in Copenhagen. Co-organiser and workshop tutor at several IDAP (Improved Damaged Assessment of Parchment) courses in Europe. Has been working as a Research Assistant on various leather and parchment projects for the School of Conservation in Copenhagen since 2007. Worked as a conservator at the Regional Archives in Lund, Sweden between 2002 - 2007, where she was Head of the Conservation and Bookbinding Department between 2004 - 2007.

Steen Kristian Frank holds a Bachelor degree in Biology (BSc) from Copenhagen University and has been working in the field of biology for several years. Has volunteered in conservation projects in Greece (sea turtles and chameleons) and Honduras (collection of insects). 2005 – 2006 worked as a research assistant for BioBasis at The Natural History Museum of Copenhagen with identification of insects and is currently working at the same place as a guide. Began studies of Conservation-Restoration Science 2007 at the School of Conservation in Copenhagen where he is now a master student (MSc). Other current project is the work as Research Assistant in the ongoing EC MEMORI project.

The Royal Danish Academy of Fine Arts E-mail 1st author: [email protected] Schools of Architecture, Design and Conservation E-mail 2nd author: [email protected] School of Conservation E-mail 3rd author: [email protected] Esplanaden 34 E-mail 4th author: [email protected] 1263 Copenhagen K Denmark

ICOM-CC Leather and Related Materials WG Offenbach 2012

Research on New Working Methods with Leather in Historical Artworks: Restoration Work of a Frontal Altar in Gilt and Painted Leather from Central Sicily

Loredana Mannina, Angela Lombardo

Abstract The present paper is an abstract of the first author's graduate thesis. The research and the conservation work were carried out during two years (2008-2010). This research concerns two aspects: an historical research of artefacts in leather on the Sicilian territory and their distribution on one side, and a scientific research about products and material useful for the conservation of artworks in leather on the other side. The study of the restoration of the frontal altar in gilt and painted leather from the Sanctuary of the Santissimo Crocifisso di Papardura in Enna began with a historical and stylistic examination of the work of art attributed to the XVIII century. Our understanding of the frontal altar leather and its decoration has been deepened thanks to specific scientific analysis. We wanted to be sure to respect the original material during the restoration work. For this purpose, we decided to use a diagnostic procedure in order to provide a clear picture of the materials (type of leather, silver leaf, pigments and binder), technique and state of preservation. After such careful analytical study the conservation work of the frontal altar we carried out a more conscious opera because the preventive study had given the most important data, thanks to whom was possible draw up a restoration design.

Keywords Sicilian gilt leather, Sicily, gilt, leather, aqueous cleaning, aqueous system, cleaning, elastic tensioning frame, frame, tension, tensioning.

Introduction

The recent revaluation of artistic artefacts made in leather has lead to the rediscovery of an extensive heritage of artworks that has been left to degrade and that was virtually unknown until now. This rediscovery has made it possible to start an important search of the places where these sorts of artefacts are located and to record their state of preservation in order to foster their appropriate conservation, after their cataloguing [1,2].

The starting point of our search was the survey in the Sicilian territory in order to identify the location where the leather artefacts were being preserved. This search has led to interesting information showing that in Sicily, there are lots of artefacts made of decorated leather. Especially in the oriental part of the island there are frontals, aristocratic furniture and sacred furnishings, tapestries, and more. Unfortunately, often many of them are not well conserved. The more interesting data acquired by searching archives concerned the production of these artefacts. From the acquired information, we understand that there was a great demand of such artefacts, especially of liturgical furnishing between the 17th and the 18th Century. This high demand is in regards to not only imported artefacts from Venice, but also to the Sicilian production of gilt and painted leather. Those findings attest the presence of artisans that worked leather in Sicily. This is important news that should be further researched with a more extensive study.

The Frontal Altar in Gilt and Painted Leather with the Crucifixion Theme

The production of decorated leather was widespread in Sicily, as found in this historical research. By way of demonstration, the frontal altar herein examined is part of a series of the five Antependia depicting "The Five Sorrowful Mysteries of the Passion of Jesus". Only recently have artefacts made in artistic leather become credited as ‘artwork’; before then, there was little interest in such articles, and because of this, it has been very difficult to find information about the story of the frontals of the Sanctuary of Enna. Such type of artworks were considered as an expression of the "minor arts" and the people that should have taken care of this heritage did not carry out their responsibilities, taking no interest in the conservation of these leather artworks and their cataloguing.

ICOM-CC Leather and Related Materials WG Offenbach 2012

Our research focused on records from the Archive of Enna. This way we were able to obtain some information that has allowed us to put forward a hypothesis about the period in which the frontals were produced. The terminus ante quem is the year of the building Sanctuary (1656). This hypothesis is based on the fact that the iconography of the five frontals is inspired to the iconography of the Sanctuary (the passion of Jesus) [3]. It is clear that the frontals were made for that place and not for any other one. In addition, only in this period the theme of the Passion of Christ was popular to decorate the frontals. The terminus poste quem could be considered 1736. At that time a local wood carver, Paolo Guglielmaci was commissioned to design and make the wooden frames for the frontals [4]. We conclude that for these reasons, the confection of the Sanctuary frontals dates between 1656 and 1736.

Materials, Manufacture and Iconography

The frontal altar in gilt and painted leather with the Crucifixion theme comes from the Sanctuary of the Santissimo Crocifisso di Papardura in Enna (Figure 1). It synthesizes features of manufacturing that in Venice were called "cuoridoro" (‘Golden Hearts’). It is the result of the union of more than one type of skin with different form and measure (three large skins and seven smaller pieces) joined either by gluing or by overlapping parts together. The frontal altar had a leather lining kept together by three large stitched skins. The decorative technique is the traditional one: on the leather surface, silver leaf was applied, then the surface was gilded with gilt varnish (Mecca) in a way to emulate true gold. On this gilded background the decorative motif was stamped using woodblocks well visible in the relief of the central image frame. Oil paint was then applied onto this design. The frontal altar was finished with eight kinds of different punch marks (lines and dots, circles, squares, etc.). The decorative repertory includes classic motifs such as acanthus leaves, and motifs with Sicilian taste, in a perfectly symmetrical composition where the centre of the composition is occupied by the sacrum composition: Jesus crucified between Mary and Saint John. These are only the information inferable by the careful visual observation, but these were not sufficient for the correct identification of the technique of execution. Diagnostic work was carried out in order to acquire a more clear understanding about materials and techniques.

Figure 1: Frontal in gilded and painted leather, 18th century, Sanctuary of the Santissimo Crocifisso di Papardura, Enna (after treatment).

Diagnostic Work: Study of the Technique of Execution

The diagnostic work was performed by the authors with the support of the scientific research departments of the University of Palermo and the chemical laboratory at the Regional Centre for Design and Restoration of Palermo. In the leather artistic work, the original craftsman’s search for specific features, connected with the possibility of receiving decoration, was the cause for the historical use of only a few animal species. Every animal species is characterized by a particular kind of tissue, and these change in thickness and size of fibres. For this reason, it is possible to recognize the species by carrying out a microscopic analysis on the leather surface. In this study was possible to recognise the species thanks to a little exposed portion, where 70

ICOM-CC Leather and Related Materials WG Offenbach 2012

the silver leaf had left the leather surface visible. The investigation was carried out with a digital microscope capable of taking instant photos. These were compared with available bibliographic data, and therefore it was possible to ascertain that the leather used was sheepskin.

The study of the techniques of execution started with the observation of the painted surface through a digital microscope. The macro-photographs provided important information, especially useful to identify more precisely the points where future investigations could be carried out. XRF analysis was subsequently conducted on all coloured areas. Thanks to its non-destructive feature, the XRF is an instrument of choice in the analytical field for the identification of pigments. The results obtained from XRF analysis were then collated with results from a second technique, the Raman Spectroscopy, which was used to further compile more comprehensive information, especially in the case of some pigment’s chemical composition which the XRF analysis could not perfectly identify, for example in the case of the lighter elements. In addition, the leather substrate and the presence of the silver leaf can interfere with the measurements. This is why it was decided to complement with another type of analysis when the pigment was not recognized by XRF. Raman spectroscopy was associated as an auxiliary technique; for this technique however, it was necessary to take some representative micro-samples. Data was compared and processed in a single test, integrating the results of the analyses of the cross section, carried out with SEM in order to know the stratigraphy and to identify the presence of some overpaint, and also performing the micro-analysis (EDS) to confirm some hypotheses about material composition. To this end, cross-sections were prepared of two representative samples of the areas that were important to investigate. The position of the samples was carefully chosen during the preliminary screening, thanks to which it was possible to understand the starting situation and to limit such destructive analysis to a few points. The analyses have given important information about the nature of pigments and have allowed identifying the entire paint palette used by the artist. The blue and red pigments were more difficult to identify, this is why it was necessary to study their composition with more than one diagnostic technique. For example the contextual reading of XRF, Raman and SEM data has permitted to understand the nature of red pigment. The fluorescence measurements of XRF didn't permit to identify the artificial or natural composition because the impurities are invisible for the instrument. The microanalysis made with the SEM has shown that it was an artificial pigment because the spectrum hasn't registered the presence of impurity. In this way the red pigment was identified as Vermilion. The analyses of the cross section confirmed the bibliographical data and the information available from visual observation of the artwork. In fact, as demonstrated by the study of Fioravanti [5], the decoration is constituted by (Figure 2): - Leather as a first layer, as a material used for the decoration; - The silver leaf was applied evenly on the leather (the glue layer was not visible where the silver leaf was applied. Nevertheless the micro-photos, taken with SEM, showed the fibres more close near the silver layer). - The silver leaf was used as support material for the decoration (painted surface)

Figure 2: (a) Macro-photography of the cross-section taken by SEM; (b) Cross section of the gilded and painted leather where it is possible to observe: 1) leather; 2) Silver leaf (EDS); 3) Mecca (gilt varnish); 4) Painted layer. 71

ICOM-CC Leather and Related Materials WG Offenbach 2012

State of Preservation

When the frontal altar in gilt and painted leather arrived at the laboratories of the Regional Centre for Design and Restoration of Palermo, it appeared to be in a considerably bad state of preservation. It showed signs of both mechanical damage and degradation in the decorative layers. The biggest numbers of problems were due to the use of a wooden rigid frame where the leather was nailed around its entire perimeter. This wooden frame was probably used as a method to hang the frontal at the altar when it changed its destination use and it was re-employed as a painting in the Sacristy of the Sanctuary (when the silver altar was re- discovered). The rigid system had led to splits and tears, losses and abrasions, making the leather less resistant to mechanical stress.

For their part, the decorative layers were characterized by alteration in colour, cracking and oxidation of the metal leaf. The painted surface have shown the evident sign of neglect: a thick layer of dust, soiling and surface deposits covered the entire painting to such an extent that, at that moment, it was impossible to determine if the central image had survived (Figure 3). In addition, insect damage and microbiological attack were well observable.

Figure 3: The main degradations of the painted layer: repainting, degradation and yellowing of the final varnish, surface deposit with greasy and oily consistency.

Diagnostic Work: Study of the Preservation State

Other preliminary analyses were necessary to evaluate the preservation condition of the leather, before starting with the conservation work. Measurement of pH helped to determine if there were chemical degradation processes such as acid degradation or demineralization of the material. Both the microbiological and entomological analyses were necessary because the leather showed signs of biological attack. In particular, little white and powdery spots and insect grazing tracks were observed. These scientific analyses were very important in order to determine the preservation state and to ensure appropriate decisions concerning restoration work.

A sample was manually taken from a piece of leather almost detached. The sample with a weight of 0.5 gram was put in a flask with deionized water (pH 7.0) and hermetically sealed. The flask with sample was shook up and then it was let it stand. Twelve hours later the pH of the watery extract of the sample was measured with a traditional pH meter. Some studies have shown that the pH safety range for artistic leather is between 3.5 and 6.0 [6]. In this case, the pH of the extract was measured as being 5.9; this was considered adequate.

The entomological and microbiological analyses showed a precarious state of preservation. Through careful observations as a first step, and afterwards through the use of a digital microscope, we investigated whether there was a dangerous entomological attack actively taking place; in particular there were signs of larvae tunnelling activities for two kinds of species of bugs (wood boring beetles and clothes-moths). For this reason it was indispensable to proceed with an anoxic disinfestation before starting the restoration.

ICOM-CC Leather and Related Materials WG Offenbach 2012

Conservation

Following a well-established procedure, the intervention has started only after a careful preliminary phase of written, graphic and photographic documentation. We made a metric survey drawing and a careful mapping of the techniques of execution (for example every sort of punch marks was surveyed). We carried out three different mappings about the state of preservation of leather, of decorative layer and of the lining. This phase has been very important because it represented the starting point to draw up the restoration design of the frontal altar.

During the entomological analysis, it became clear that an insect attack was still active; therefore, it was necessary to start the conservation treatment with an anoxic disinfestation. So, after the removal of all nails (about a hundred) and the wooden frame, we built an anoxic chamber with an inert gas, nitrogen. The leather was left in the chamber per a period of thirty days with a data logger as a means of evaluating the environmental condition inside the chamber. After that period, the frontal altar was submitted to an accurate dusting-phase with a macro-vacuum cleaner, and with the use of a micro- vacuum for the interstices.

Choice of Cleaning Method for the Gilt and Painted Leather

As well as the scant bibliography, the complications in working with this sort of work of art lie in the necessity to consider contextually both the leather's features and its decoration. The leather substrate has intrinsic qualities that require superficial work in order to prevent the fibres from felting, or the tannins from solubilising. The decoration (which includes silver leaf, pigments and varnish) requires specific solutions. In addition, little knowledge remains about the way to work in cases where it is necessary to remove overpaint or varnish from a paint layer, itself fragile due to its composition and thinness.

As well as the surface, completely occulted by deposits, the painted and gilt leather had more layers of repainting in the perimeter zone, and a thick coating was degraded visibly. In the treatment of leather, it is always preferable not use solvents because it can cause the fibres to dehydrate or may provoke chemical processes in the silver leaf and the pigments. Taking cues from recent studies [7], it was decided to test the "Water-Based Cleaning Systems" (as first developed by Richard Wolbers) in the frontal altar cleaning. These cleaning systems allow one to use the water's cleaning quality while lessening its intrinsic risks, thanks to the addition of particular "activities", good for improving some important features:

- Choosing and controlling the pH with the use of buffer solutions; - Adding chelating agents, able to improve the solvent power; - Making the solutions denser with gelling agents, useful to reduce the scant power of water to wet the surfaces (hydrophobe surfaces, like varnish, oil painting), but also useful to reduce the high power of water to wick in under the surface.

Among the better aspects of this cleaning method is that is possible prepare all solutions in the laboratory, with the characteristics needed.

The polychrome surface of the frontal altar was seriously compromised in its legibility. In addition a thick layer of deposit having a greasy and oily consistency had veiled the shine of the gold varnish. After testing with the complete method, the more satisfying system was the use of a buffer solution with an acid pH (5.5) in order to remove the surface deposit and the use of a buffer solution with a basic pH (8) in order to remove the overpaints and thin down the varnish layer. The solutions were prepared with the addition of chelating agents (the weak chelating agent TAC for the buffer solution with 5.5 pH, and the strong one, EDTA for the buffer solution with 8.0 pH), mixed with gelling agents (Klucel) (Figure 4).

ICOM-CC Leather and Related Materials WG Offenbach 2012

Figure 4: The final result obtained by the use of Water-Based Cleaning Systems. Before and after cleaning

Top left: Figure 5: Integration of losses. Top right: Figure 6: Lining with partial system. Bottom left: Figure 7: Construction of the elastic tensioning frame. Bottom right: Figure 8: Calibration of the frame to the correct degree of tensioning.

ICOM-CC Leather and Related Materials WG Offenbach 2012

The Tensioning Frame and Last Part of the Conservation Treatment

After the cleaning phase, humidification of the frontal altar was necessary in order to re-give the correct flatness to the leather and to relax the leather fibres. After assessing traditional humidification systems, we decided to use a GoreTex® membrane creating a localised chamber in order to permit a correct procedure of humidification. The losses were integrated with the use of new leather, compatible in elasticity and thickness with the original. The adhesive chosen for this operation was a mixture of Plextol B500 and Tylose 3%. We added the Tylose (proportion Plextol B500: Tylose, 9:1) in order to thicken the acrylic resin. In this way we reduced the risk of seepage in the leather (Figure 5). Lining was achieved by applying only reinforcing bands and avoiding a total lining that could have create problems. The junctions where the skins were glued to form the entire sheet were unglued. This problem was caused by the incorrect tensioning state of the wooden frame. This is why we preferred to use a partial lining with covering bands over the junctions. The bands were obtained by the use of only the dermis part of the skin in order to have the elasticity but not the rigidity of the skin (Figure 6). The integration of colour was done with watercolour applied on the new leather and varnish-colour over the silver leaf.

The restoration work was completed with the construction of an elastic tensioning frame where the tensioning was assigned to a sprung-stretcher system (Figures 7-8). We know that if the relative humidity in the ambient environment is high, normally the fibres of leather lengthen themselves, and conversely if the ambient environment is dry the fibres shrink. This can cause tears. To this end, after studying almost every kind of conservation-frame, we opted for this sort of frame. Indeed, it can change automatically its tensioning degree thanks to the work of the springs. The spring can lengthen or shrink itself in an automatic way, following the leather’s movement. This framework makes sure that the leather can move without rigid constraints that could lead to tears in the leather [8].

Conclusion

This Sicilian painted artwork is now well preserved and has better odds to survive for future generations as a testimony of Sicilian leather craftsmanship and great art (Figure 9).

A display case was designed for it with controlled microclimate and illumination in order to provide this exquisite piece of artwork with well-deserved, high quality conservation protection against adverse environmental conditions while safeguarding it in-situ in its location and context in the Sanctuary. Figure 9: Before and after the conservation work.

ICOM-CC Leather and Related Materials WG Offenbach 2012

Acknowledgments The authors wish to thank the Regional Centre for Design and Restoration of Palermo for the help given throughout the duration of the project. The authors also appreciate the helpful assistance of Dr. Cosimo Di Stefano, from the chemical laboratory of the Palermo's CRPR and would also like to thank Dr. Bartolo Megna, a materials engineer from the Laboratory of Materials for Restoration and Conservation, Dept. of Materials Engineering, University of Palermo. The enthusiastic collaboration of Leonardo Borgioli from the CTS is warmly acknowledged.

Materials Tris-EDTA (Salt of Ethylenediaminetetraacetic acid): Sigma-Aldrich, www.sigmaaldrich.com Klucel G (non-ionic hydroxypropylcellulose): CTS srl. www.ctseurope.com Plextol B500 (thermoplastic acrylic resin): CTS srl. www.ctseurope.com TAC (Tri-ammonium citrate): Sigma-Aldrich, www.sigmaaldrich.com TYLOSE MH 300 P (methylhydroxyethylcellulose): CTS srl. www.ctseurope.com

References [1] Mannina L. «Ricerca e sperimentazione di metodi d’intervento su cuoio in manufatti di interesse storico- artistico. Il restauro del Paliotto in cuoio dorato e dipinto del Santuario del SS. Crocifisso di Papardura.», Graduate thesis. Palerm 2010 [2] Berardi M.C., Nimmo M., Paris M., «Il cuoio dorato e dipinto, ricerche e conservazione», Rome ICR 1996. [3] ASEN, notary Francesco Planes, Vol. 1237, c. 415, act of 1-3-1671. [4] ASEN, notary Vito Planes, Vol. 1338, c. 197, act of del 21-10-1736. [5] Fioravanti L., «Dell’arte dei corami d’oro e sua manifattura», Dello Specchio di Scientia Universale, Venice 1564., Ch XLI. [6] Kite M, Thomson R. «Conservation of leather and related materials», Oxford 2006. [7] Cremonesi P., «L’Uso di Tensioattivi e Chelanti nella Pulitura di Opere Policrome», Padoue 2001. [8] Berardi M.C., Nimmo M., Paris M., «Tensionamento controllato per dipinti su cuoio: dati sperimentali», bulletin ICR, n. 4, 2002.

Biographies Loredana Mannina obtained a master's degree in conservation and restoration of Cultural Heritage at the University of Palermo with an experimental thesis about the leather restoration. It won the first prize in the thesis competition organized by IGIIC and sponsored by the Droghetti foundation in 2011. Thanks to this competition, the thesis is in course of publishing as an e-book for Nardini Editors. After some professional experiences with the restoration of wooden sculptures and archaeological ceramics, she's going to acquire the title of textile and leather restorer. Leather and textile conservator - 16 Via Elio Vittorini, 91011 Alcamo (TP), Italy - +39 329 6326988 - [email protected]

Angela Lombardo, Vice president of the Esperia association, charter member of the Polo Tessile Mediterraneo Melilli foundation, has been the handler of the restoration laboratory of organic materials at the Regional Centre for Design and Restoration for Sicily and she teaches Conservation, Maintenance and Restoration at the University of Palermo since 2003, coordinating the restoration laboratories of leather and textile materials and of book and archival materials. She was the organiser in the celebrations of the eighth centenary of the Federico II birth. On behalf of the chair of the region Sicily, she has coordinated the international project about the minimal opening of the Emperor Federico's tomb in the Cathedral of Palermo. She took part (1995) in the project for the institution of a Centre for the conservation and restoration of the historical-cultural Heritage of the North-Occidental China. Supervisor of the textile laboratory - University of Palermo - 46 Via Emerico Amari - 90139 Palermo, Italy - +39 3289162706 - [email protected] - [email protected] - www.angelalombardo.it

ICOM-CC Leather and Related Materials WG Offenbach 2012

The Leather Drying Trial and Associated Analytical Work

Angela Middleton, Karla Graham, Paul Garside

Abstract The aim of this study was to compare and evaluate different treatment and drying techniques for wet archaeological leather; using parameters such as shrinkage, flexibility, appearance, time, and costs. Treatments included 20% glycerol,

20% PEG and no impregnation. Half the samples in each treatment category were pre-treated using 5% Na2EDTA. The leather was dried using vacuum freeze drying, non-vacuum freeze drying, air drying and controlled air drying. As part of this project leather samples were examined using SEM and FTIR. Scanning Electron Microscopy (SEM) aimed to look at the differences in the fibre structure following treatment. Fourier-transform infrared spectroscopy (FTIR) analysis aimed to examine the changes in the leather as a result of the different treatment methods. Focus was put on the leather that had received EDTA treatment in order to examine whether the leather changes following EDTA treatment and whether this can be detected using FTIR. This paper presents the results of the drying trial, gives an account of the analytical work and presents preliminary results.

Keywords Leather, archaeological, treatment, drying, air drying, freeze-drying, glycerol, PEG, polyethylene glycol, impregnation, consolidation, Na2EDTA, EDTA, Conservation, Analysis, Fourier-transform infrared spectroscopy (FTIR), Scanning Electron Microscopy (SEM), Environmental Scanning Electron Microscopy (ESEM)

Introduction

The leather drying project was carried out by English Heritage (EH) in conjunction with the Museum of London. The WOAM paper ‘A comparative study of various impregnation and drying methods for waterlogged archaeological leather’ provides a full background to the project (Karsten et al. 2011). In summary, the aim of this project was to provide comparative data on the efficacy of the most commonly used remedial conservation methods for waterlogged leather. The first impetus for the project was the development of new EH guidelines on waterlogged organic artefacts (Karsten et al. 2012) which required research to test the efficacy of treatments. The second impetus was the issue of unconserved back-logs of leather in the London region where the cost of vacuum freeze drying was cited as the main reason for not conserving the leather.

Different impregnation and drying methods were therefore compared using parameters such as shrinkage, flexibility, time, effort and equipment to evaluate each method. The leather was also examined by Fourier- transform infrared spectroscopy (FTIR) and Scanning Electron Microscopy (SEM). The drying trial is summarised and the results are briefly described. The results of the analytical work form the main part of this paper.

The Drying Trial

The sample material was donated by University College London and originated from Novgorod, an urban site in North-West Russia, where it was collected from the spoil heap. A total of 89 bags contained wet archaeological leather off cuts from three different locations, indicated by the codes NE, NF and T. All the leather from region T was much more degraded than the leather from region NE or NF.

The leather was photographed and drawn. Drawings were annotated to establish the shrinkage after conservation. A condition score originally developed by Suenson-Taylor and Sully (1996) was used to record the overall condition of the leather before and after conservation. The conservation treatments involved combinations of different pre-treatment, impregnation and drying techniques (see Table 1):  Half of the samples were pre-treated with the complexing agent disodium ethylene diamine tetra

acetic acid (Na2EDTA) for two hours followed by 48 hours rinsing in running tap water.

ICOM-CC Leather and Related Materials WG Offenbach 2012

 The impregnation medium was either polyethylene glycol (PEG 400) or glycerol at a concentration of 20%. Some samples did not receive any treatment.  The four drying methods were:  Air drying: The leather was laid out on a polyethylene foam (Jiffy Foam®) lined tray with a polythene cover loosely draped over it at ambient conditions, but away from direct sunlight.  Controlled air drying using saturated salt solutions: The leather was placed on perforated shelves inside a humidity chamber with the salt solution at the bottom. The solution was changed every two days: Barium chloride – 90%RH, Potassium iodide – 70%RH, Magnesium nitrate – 55%RH.  Vacuum freeze drying: The leather was frozen in a domestic chest freezer for two days and then placed on an acid free tissue lined trays inside an otherwise empty freeze drying chamber at - 30ºC.  Non-vacuum freeze drying: The leather was placed on polyethylene foam (Plastazote®) lined baskets. The baskets were stacked on top of each other between a fan and silica gel.

The end point for all drying methods was established when two constant weights were measured.

Table 1: Overview of treatment and drying methods. Complexing agent wash Impregnation Drying Air drying 20% polyethylene glycerol (PEG Controlled air drying 400) Non-vacuum freeze drying Vacuum freeze drying 5% Na2EDTA & rinsed Air drying 20% glycerol Controlled air drying Non-vacuum freeze drying Vacuum freeze drying Air drying No impregnation Controlled air drying Non-vacuum freeze drying Vacuum freeze drying Air drying 20% polyethylene glycerol (PEG Controlled air drying 400) Non-vacuum freeze drying Vacuum freeze drying No treatment Air drying 20% glycerol Controlled air drying Non-vacuum freeze drying Vacuum freeze drying Air drying No impregnation Controlled air drying Non-vacuum freeze drying Vacuum freeze drying

Analytical Work: Aims and Methodology

(1) SEM

Scanning electron microscope (SEM) analysis aimed to show differences in the leather fibre structure before and after treatment. The focus was to characterize firstly the influence of the impregnation agent on the leather structure and secondly, how the four different drying methods affected the fibre structure (Table 2). To achieve the first, glycerol and PEG treated samples were directly compared by using the extremes of vacuum freeze dried and air dried samples. For the second focus, the four different drying methods were compared using samples that had received no impregnation.

ICOM-CC Leather and Related Materials WG Offenbach 2012

Table 2: Overview of leather that was analysed by ESEM and SEM. Leather Impregnation Drying method Aim of sample method analysis NF58 20% Glycerol Vacuum Freeze Drying NF19 20% Glycerol Air Drying Influence of T4 20% PEG400 Vacuum Freeze Drying bulking agent NF34 20% PEG400 Air Drying on fibre structure NE9 no impregnation Non-vacuum Freeze Drying Influence of NF3 no impregnation Air Drying drying method NF12 no impregnation Controlled Air Drying on fibre NF13 no impregnation Vacuum Freeze Drying structure

Samples were cut with a fresh scalpel blade. The samples were mounted on the newly cut edge, i.e. standing on a cross section, so that the opposite un-cut cross section could be viewed in the SEM. It was suspected that cutting by scalpel would cause compression within the leather structure, which could jeopardise image results.

SEM Imaging of Wet Leather Samples Attempts to image the wet leather samples were carried out in the environmental SEM (ESEM) at the Institute of Archaeology, UCL but proved generally unsuccessful. Only one wet samples could be imaged. The difficulties consisted in the samples drying during the process of mounting, de-gassing and specimen beam interaction. Normally, dry samples are carbon or gold coated and imaged under vacuum. If biological and wet samples are placed under vacuum the water vapour from the material interferes with the X-ray beam and the samples dry out. Biological and wet samples must therefore be left uncoated and not placed under total vacuum.

Whilst some images were taken, the main difficulty was balancing the required environmental parameters within the chamber (to maintain the samples in a wet state) and image quality. Since UCL staff rarely use the SEM for wet samples, it was decided that there was not sufficient time and expertise with wet material to achieve optimum conditions and image quality. A decision was therefore taken to undertake SEM imaging of the leather after conservation and in the dry state.

SEM Imaging of Dry Leather Samples The dry leather samples were mounted as described above, gold coated and analysed using an FEI Inspect F FEG SEM at Fort Cumberland, Portsmouth.

(2) FTIR

Fourier-transform infrared spectroscopy (FTIR) analysis aimed to show differences in the leather as a result of different treatment and drying methods, with emphasis on the effects of EDTA. 24 samples were analysed with the data collected using a Perkin-Elmer ‘Spectrum 100’ spectrometer fitted with an ATR (attenuated total reflectance) accessory; spectra were recorded over the range 650-4000 cm-1, with a resolution of 4.00 cm-1, and were averaged over 32 accumulations. The ATR technique was chosen as it does not require samples to be taken from the bulk specimen.

The samples represent the complete range of treatment and drying method combinations (Table 3). They were analysed twice; both in the wet state (where applicable, after their treatment with EDTA) and, after drying.

ICOM-CC Leather and Related Materials WG Offenbach 2012

Table 3: Overview of leather that was analysed by FTIR. Leather Treatment Drying Method T6 Vacuum Freeze Drying NF50 No impregnation Controlled Air Drying NF40 Air Drying NE13 Vacuum Freeze Drying NE4 Air Drying NF47 Controlled Air Drying NF2 5%EDTA Air Drying NF32 Vacuum Freeze Drying NF1 Vacuum Freeze Drying NF34 20%PEG Air Drying NF42 Air Drying NF24 Vacuum Freeze Drying NF5 Non Vacuum Freeze Drying NF65 20%Glycerol Vacuum Freeze Drying NF19 Air Drying NE5 Air Drying T7 5%EDTA 20%PEG Vacuum Freeze Drying NF60 Vacuum Freeze Drying NF33 Air Drying NF20 Air Drying NF8 Non Vacuum Freeze Drying NF57 5%EDTA 20% Glycerol Non Vacuum Freeze Drying NF35 Non Vacuum Freeze Drying NF32 Vacuum Freeze Drying

Results

Results of the Drying Trial

Shrinkage

Shrinkage was calculated by measurements of length and breadth when possible. Measurements were made with a vernier calliper with a precision of 0.1mm. The percentage shrinkage was calculated from the difference in measurements before and after treatment. The irregular shape and size of the experimental samples made any attempt to measure area or volume impractical.

The average shrinkage rates for different drying methods are reasonable close to each other and given the relatively high standard deviations, no significance can be detected. Thus, it would seem that drying method has a relatively small effect on shrinkage. The average shrinkage rates for different treatment methods show greater differences, despite the relatively high standard deviations. No treatment or treatment with 5%EDTA alone gave rise to the greatest shrinkage while PEG400 and glycerol gave lowest. With both PEG400 and glycerol, the pre-treatment with 5%EDTA increased the shrinkage of the leather (Table 4).

ICOM-CC Leather and Related Materials WG Offenbach 2012

Average Average per Standard Average Average per impregnation Standard Conservation Method Shrinkage (%) drying method Deviation Conservation Method Shrinkage (%) and/or washing method Deviation On average all air dried 3.7 On average all 20% Glycerol 1.5 20% Glycerol/ AD (4) 2.17 20% Glycerol/ AD (4) 2.17 samples shrunk by impregnated samples shrunk 20% PEG 400/ AD (4) 3.17 5.72%. 20% Glycerol/ CAD (2) 3.55 by 3.92%. 5% EDTA/ AD (4) 10.99 20% Glycerol/ NVFD (5) 4.2

5% EDTA 20% Glycerol/ AD (3) 3.21 20% Glycerol/ VFD (3) 5.76 On average all 20% PEG 400 2.4 5% EDTA 20% PEG 400/ AD (3) 5.26 20% PEG 400/ AD (4) 3.17 impregnated samples shrunk No Impregnation/ AD (4) 9.49 20% PEG 400/ CAD (3) 4.83 by 5.84%. On average all controlled 7.6 20% Glycerol/ CAD (2) 3.55 20% PEG 400/ NVFD (3) 8.86 air dried samples shrunk 20% PEG 400/ CAD (3) 4.83 by 8.78%. 20% PEG 400/ VFD (4) 6.5 On average all 5% EDTA 5.3 5% EDTA/ CAD (3) 22.03 5% EDTA/ AD (4) 10.99 impregnated samples shrunk 5% EDTA 20% Glycerol/ CAD (3) 3.49 5% EDTA/ CAD (3) 22.03 by 14.25%. 5% EDTA 20% PEG 400/ CAD (3) 4.73 5% EDTA/ NVFD (4) 13.5 No Impregnation/ CAD (1) 14.1 5% EDTA/ VFD (4) 10.49 On average all non 3.5 On average all 5% EDTA 20% 0.9 20% Glycerol/ NVFD (5) 4.2 5% EDTA 20% Glycerol/ AD (3) 3.21 vacuum freeze dried Glycerol impregnated samples 20% PEG 400/ NVFD (3) 8.86 samples shrunk by 5% EDTA 20% Glycerol/ CAD (3) 3.49 shrunk by 4.04%. 7.51%. 5% EDTA 20% Glycerol/ NVFD 5% EDTA/ NVFD (4) 13.5 (3) 5.18 5% EDTA 20% Glycerol/ NVFD (3) 5.18 5% EDTA 20% Glycerol/ VFD (4) 4.27 On average all 5% EDTA 20% 2.0 5% EDTA 20% PEG 400/ NVFD (3) 5 5% EDTA 20% PEG 400/ AD (3) 5.26 PEG 400 impregnated 5% EDTA 20% PEG 400/ CAD samples shrunk by 5.99%. No Impregnation/ NVFD (4) 8.32 (3) 4.73 On average all vacuum 2.3 5% EDTA 20% PEG 400/ NVFD 20% Glycerol/ VFD (3) 5.76 freeze dried samples (3) 5 shrunk by 7.48%. 5% EDTA 20% PEG 400/ VFD 20% PEG 400/ VFD (4) 6.5 (5) 8.95 On average all no 2.6 5% EDTA/ VFD (4) 10.49 No Impregnation/ AD (4) 9.49 impregnation samples shrunk 5% EDTA 20% Glycerol/ VFD (4) 4.27 No Impregnation/ CAD (1) 14.1 by 10.2%. 5% EDTA 20% PEG 400/ VFD (5) 8.95 No Impregnation/ NVFD (4) 8.32 No Impregnation/ VFD (5) 8.9 No Impregnation/ VFD (5) 8.9

Table 4: Mean shrinkage and standard deviation per treatment category1.

1 Please note, due to a re-appraisal of the methodology for shrinkage calculations this data presented here differs from the data presented in Karsten et al. 2012. This however does not alter the overall outcome and conclusions. 81

ICOM-CC Leather and Related Materials WG Offenbach 2012

Flexibility

Flexibility was measured in two ways: Method 1 involved suspending the leather over an edge and measuring the degree of change in movement. Method 2 involved a qualitative judgment of flexibility in terms of “acceptable” and “unacceptable”.

According to method 1 all samples decreased in flexibility and method 2 resulted in both increases (change from unacceptable to acceptable) and decreases (change from acceptable to unacceptable) in flexibility.

Method 1: No samples increased in flexibility according to method 1. The lowest losses in flexibility occurred with EDTA alone. The PEG treated leather resulted in the greatest loss in flexibility. A decrease in flexibility resulted from all drying methods and the difference between them was not significant.

Method 2: Of the 89 leathers 11 were classed as unacceptable and 78 were classed as acceptable before conservation. After conservation 31 leathers were classed as unacceptable and 58 as acceptable. Table 5 gives an overview of how the flexibility was rated after treatment.

Table 5: Changes in flexibility before and after conservation according to Method 2. Treatment Method Flexibility Before Conservation After Conservation Method 2 Air Drying acceptable 21 14 unacceptable 3 10 Controlled Air Drying acceptable 14 11 unacceptable 4 7 Non Vacuum freeze acceptable 21 19 Drying unacceptable 2 4 Vacuum Freeze Drying acceptable 22 14 unacceptable 2 10 No impregnation acceptable 14 6 unacceptable 0 8

5% Na2EDTA acceptable 9 6 unacceptable 6 9 20%Glycerol acceptable 14 13 unacceptable 1 2 20%PEG400 acceptable 14 9 unacceptable 1 6

5% Na2EDTA + acceptable 21 14 20%Glycerol unacceptable 3 10

5% Na2EDTA + acceptable 14 10 20%PEG400 unacceptable 1 5

Condition Score

The leather samples were judged in the following categories: cohesivity, friability and flexibility (Suenson- Taylor and Sully 1996). A score of 1 – 4 was assigned, whereas the higher the number, the better the condition per category.

Overall, 43% of the condition scores increased, 25% decreased and 32% remained unchanged. Increases in value occurred mainly from improvements in the cohesivity and friability values. Decreases in value occurred mainly in the flexibility value. Items from site T decreased in condition score across all three categories. Overall, based on the condition score, the best treatment is 5% EDTA 20% glycerol followed by controlled air drying or non vacuum freeze drying. 82

ICOM-CC Leather and Related Materials WG Offenbach 2012

Effect of EDTA

The spectra of EDTA treated leather looked no different to the spectra that had not received this treatment (Figure 4). This shows that EDTA was successfully washed out. But it also raises the question of whether FTIR is the best method to detect the effects of EDTA on leather. Jones and Hovmand (2001) have used XRD in the past and this seems to be a better method.

Figure 4: Comparison of leather that has received no impregnation (blue) and leather that has received EDTA only (black).

All spectra showed some differences when comparing the before and after conservation spectra (Figures 8- 9). Differences were also detected between leather that had not received any treatment, PEG or glycerol treatment (Figures 5-7).

All leather that had received no bulking agent treatment displayed a double peak between 1200-1000cm-1, with a small peak at around 1100 cm-1 and a bigger peak at around 1180 cm-1 (Figure 5).

ICOM-CC Leather and Related Materials WG Offenbach 2012

Figure 5: Average spectra of leather that has received no impregnation treatment.

All PEG treated leather (with or without EDTA) showed a single broad peak between 1200-1000cm-1, with a small shoulder at around 1170 cm-1 (Figure 6).

Figure 6: Average spectra of leather that has received PEG impregnation treatment.

All glycerol treated samples (with or without EDTA) show a triple peak between 1200-1000 cm-1 (Figure 7).

ICOM-CC Leather and Related Materials WG Offenbach 2012

Figure 7: Average spectra of leather that has received glycerol impregnation treatment.

Looking at the spectra of glycerol treated leather after conservation, they display a flatter peak at around 1700-1500cm-1 compared to the before treatment spectra (Figure 8).

Figure 8: Comparison of glycerol treated leather, before conservation (blue) and after conservation (black).

ICOM-CC Leather and Related Materials WG Offenbach 2012

A shift and increase in peak intensity took place in all PEG treated leather, where the rather flat peak of region 1500-1300cm-1, changes to two distinct peaks after conservation (Figure 9).

Figure 9: Comparison of PEG treated leather, before conservation (blue) and after conservation (black).

Discussion

SEM The SEM analysis of the leather in its conserved, dry state was successful in observing the fibre structure and identifying different characteristics. These observations could be related to drying methods and contributed to interpreting data such as flexibility. It should however be borne in mind that we were not able to directly compare the pretreatment fibre structure to the post-treatment fibre structure. It is very likely that some of the described fibre structures were already present before treatment. Additionally, the SEM images only focus on a small section and may not be representative of the overall leather structure.

FTIR The aim of the FTIR analysis was to see if it could be used to show differences in the leather before and after treatment, with an emphasis on the effects of EDTA. All spectra displayed emphatic differences between before and after treatment. The pre-treatment spectra highlight the degraded condition of the leather, with a loss in detail and subtlety of many spectral features, particularly the characteristic amide bands (Amide I centered at approximately 1640 cm-1, and Amide II at approximately 1530 cm-1), as well as a broad, strong water peak (in the region 3000-3600 cm-1).

Of the EDTA treated samples analysed, FTIR did not detect the presence of EDTA indicating that it must have been fully washed out as intended. As a sequestering agent, EDTA removes minerals from the leather. It is not clear, whether these minerals form part of the leather structure (and should therefore remain) or whether they have leached into the leather from the burial environment and form a contaminant that should be removed (Hovmand and Jones 2001 and Ganiaris et al 1982). The leather for EDTA treatment was randomly selected and in normal circumstances would never have been recommended for EDTA treatment as it did not display any iron staining for example that could have caused problems in the future due to 86

ICOM-CC Leather and Related Materials WG Offenbach 2012

oxidation. In general the spectroscopy does not indicate a significant alteration in the chemistry of the leather samples after treatment; this is as expected given the manner in which the treatment agents interact with the material (primarily through physical absorbtion and not through chemical modification).

However, FTIR spectroscopy does bring to light some important aspects: As has been demonstrated elsewhere (Godfrey and Richardson 1990), and shown again in this study, FTIR spectroscopy allows the uptake of consolidants such as PEG and glycerol to be monitored via the development of their characteristic peaks. The ATR sampling method principally results in surface sampling of the specimen, to the order of a few microns in depth. As a result, the strength of the spectral bands arising from PEG or glycerol, combined with the loss of resolution of leather peaks, indicates that a significant quantity of the former material has been absorbed at the surface; however, from this data it is not possible to determine how deeply the consolidant has penetrated into the bulk of the leather. The data also indicate that prior treatment with EDTA influences the ability of the leather to absorb glycerol, although this is not the case with PEG in the case of these samples; if the strength of the particular bands are taken to be representative of the relative presence of the component which generates that band, then by measuring the intensity ratios of bands associated with the leather and with the consolidant, and comparing the results of those specimens which had an intermediate treatment with EDTA and those without, it is possible to gain an indication of the influence that EDTA has on the uptake of the consolidant. In both cases, the Amide I band (1640 cm-1) was used to represent the leather components, whilst bands at 1085 cm-1 and 1105 cm-1 were used for PEG and glycerol respectively (Table 6). When the average ratio for PEG was considered, little difference was observed for the samples treated with EDTA compared to those which had not been treated, indicating that the agent had little influence on uptake. However, the average ratio for the glycerol treated samples was roughly fifteen percent higher for those samples which had also been treated with EDTA, compared to those which had not, suggesting that the agent improved the ability of the leather to absorb the consolidant. However, it must be remembered that the nature of the ATR sampling method means that a surface phenomenon is being observed and that this is, therefore, not necessarily representative of the behavior of the bulk material.

Table 6: Spectral ratios indicating the influence of EDTA treatment on the uptake of consolidants. PEG treated leather Average Ratio Glycerol treated leather Average Ratio (1085 cm-1 / 1640 cm-1) (1105 cm-1 / 1640 cm-1) EDTA treatment 1.91 EDTA treatment 1.05 No EDTA treatment 1.92 No EDTA treatment 0.91

Conclusion

The 2011 WOAM paper (Karsten et al. 2011) drew a number of general conclusions based on the shrinkage, flexibility, time, effort and equipment results. In light of the SEM and FTIR work, it is useful to reiterate these conclusions and outline where the SEM and FTIR have supported these conclusions, made further contributions to them or enable further conclusions and recommendations to be made.

 Pre-treatment with Na2EDTA is not required routinely. FTIR can be used to monitor that EDTA has been rinsed out successfully.

 An impregnation (glycerol or PEG400) is recommended. Samples that had received EDTA treatment only or no impregnation resulted in higher shrinkage values than the ones impregnated with PEG or glycerol. Whereas the SEM did not show any differences between the different impregnation methods, FTIR will detect the presence of a conservation agent (which can be useful when no conservation records are available on old restoration for example).

 In appearance and flexibility, the best results were generally with freeze-dried leather but air-dried leather, carried out as described above, can have good results.

ICOM-CC Leather and Related Materials WG Offenbach 2012

The FTIR did not show any differences between the drying methods. The SEM images of the air-dried leather showed a more compact fibre structure compared to the uniformly open fibre structure of freezing (either with or without vacuum).

 Non-vacuum freeze drying can produce good results but can also take a long time. It is best suited to small quantities of material. SEM supports the statement as the non-vacuum freeze drying produced good results even without impregnation (Figure 2).

Recommendations

Based on these short pieces of analytical work further research venues are proposed for SEM and FTIR, on both their applications and further work.

SEM - We recommend that further SEM work is undertaken  On a larger number of samples to verify the results and interpretations made in this report.  Using the environmental SEM to study the leather in the wet state thereby enabling a before and after comparison to be made.

FTIR - We recommend to  use the FTIR to study and understand degradation processes on archaeological leather more generally.  further study the effects of EDTA on leather and cross reference the results with XRD analysis.  retake FTIR spectra of the leather from this study to see if and how the leather changes over time.

Acknowledgements We would like to thank Dean Sully (UCL) for donating the leather used in this study and Maarit Hirvilammi (Turun Museokeskus, Finland) and Paul Simpson (Isle of Wight Council Museum Service) for sharing their experience of non- vacuum freeze drying. We are grateful to the members of the leather working group for the many fruitful discussions held over the course of this experiment. The members of the working group are: Helen Ganiaris (Museum of London), Liz Goodman (Museum of London and London Archaeology), Jackie Keily (Museum of London), Quita Mould (Barbican Research Associate), Jane Sidell (English Heritage), Kelly Domoney (former EH and MoL intern), Dean Sully (UCL). Thanks are also due to Kevin Reeves (UCL) for assistance with the SEM.

Materials Barium chloride: Fisher Scientific UK, Bishop Meadow Road, Loughborough, Leicester, LE11 5RG, UK Disodium EDTA (Na2EDTA ): Fisher Scientific UK, Bishop Meadow Road, Loughborough, Leicester, LE11 5RG, UK Glycerol: BDH Laboratory Supplies, Poole, Dorset, BH15 1TD, UK Magnesium nitrate: Fisher Scientific UK, Bishop Meadow Road, Loughborough, Leicester, LE11 5RG, UK Polyethylene Glycol (PEG): Brenntag UK and Ireland, Albion House, Rawdon Park, Green Lane, Yeadon, Leeds, LS19 7XX, UK Potassium iodide: Fisher Scientific UK, Bishop Meadow Road, Loughborough, Leicester, LE11 5RG Acid free tissue paper: Preservation by Design Polyethylene Foam (Jiffy foam®): Kewell Converters Ltd, KCL House, Station Road, Edenbridge, Kent, United Kingdom, TN8 6EG, UK Tyvek® label: Conservation Resources, 15 Blacklands Way, Abingdon-on-Thames, Oxfordshire, OX14 1DY, UK Zip Lock Bags: Richards Packaging, Unit 23, Crossgate Road, Park Farm North, Redditch, Worcestershire, B98 7SN, UK Archival Cardboard Boxes: T.G. Nuttall Packaging Limited, Units N1 / N2, Central Park Estate, Mosley Road, Old Trafford, Manchester, M17 1PG, UK

References Ganiaris H., Keene S. and Starling K. 1982. “A comparison of some treatments for excavated leather”, The Conservator, 6, pp. 12-23 Godfrey I. M.; Richardson G. W. 1990. “Analysis of waterlogged leather using Fourier Transform Infrared (FTIR) Spectroscopy - a preliminary study”, AICCM Bulletin 16(3) pp. 73-84

ICOM-CC Leather and Related Materials WG Offenbach 2012

Hovmand I. and Jones J. 2001. “Experimental work on the mineral content of archaeological leather “, Leather Wet and Dry: Current treatments in the Conservation of Waterlogged and Desiccated Archaeological Leather, London, Archetype Publications Ltd., Ed. Wills B., pp 27-36 Karsten A., Graham K., Goodman L., Ganiaris H., Domoney K. 2012. “A comparative study of various impregnation and drying methods for waterlogged archaeological leather”, Proceedings of the 11th ICOM-CC Group on Wet Organic Archaeological Materials Conference, Greenville 2010, Ed. Straetkvern K., Williams E., pp. 595-610 Karsten A., Graham K., Jones J., Mould Q., Walton Rogers P. 2012. “Waterlogged Organic Artefacts. A guideline to their recovery, analysis and conservation”. English Heritage: Swindon Suenson-Taylor K. and Sully D. 1996. << The use of condition score to determine glycerol concentration in the treatment of waterlogged archaeological leather. An experimental solution >>, Proceedings of the 6th ICOM-CC Group on Wet Organic Archaeological Materials Conference, York 1996, Ed. Hoffmann P., Grant T., Spriggs J.A., Daley T., pp. 157-172

Biographies Angela Middleton holds a degree in the conservation of archaeological heritage from the University of Applied Sciences in Berlin and a degree in Maritime Conservation Science from the University of Portsmouth. In the past she has worked for the Newport Medieval Ship Project in Newport/Wales and the Michael Faraday Museum in London. Since 2007 she works as an Archaeological Conservator for English Heritage based in Portsmouth. Here she is responsible to give advice and carry out research into all aspects of archaeological conservation. Her main interests are the conservation of maritime artefacts and the conservation of wet organic artefacts, especially wood and leather. Angela is one of the co- authors for the EH Waterlogged Organic Artefacts guidelines, and a member of ICON and the Archaeological Leather Group. Archaeological Conservator - English Heritage - Fort Cumberland - Fort Cumberland Road - Portsmouth PO4 9LD, United Kingdom - [email protected]

Karla Graham is an accredited member of ICON and gained a B.A. Honours in Archaeology and a M.A. in the Conservation of Historic Objects (Archaeology) from the University of Durham. Previously she has worked as an archaeologist on developer funded projects, an archaeological surveyor for the University of Bournemouth and a project conservator for Oxfordshire County Council. Karla has been at EH since 1999 and in her current role undertakes projects and provides specialist and strategic advice relating to investigative and remedial archaeological conservation and preservation in situ. Her particular interest is preservation in situ and collaborative projects include the study of the effects of re-watering, woodlands and the reburial of wood. She is one of the co-authors for the EH Waterlogged Organic Artefacts guidelines. She is a Project Assurance Officer for the EH National Heritage Protection Commission fund which involves monitoring archaeological organisations commissioned by EH to undertake projects. Senior Archaeological Conservator - English Heritage - Fort Cumberland - Fort Cumberland Road - Portsmouth PO4 9LD, United Kingdom - [email protected]

Paul Garside studied chemistry the University of Southampton, where he remained to carry out research for a PhD investigating the properties of natural polymer fibres, which was awarded in 2002. He subsequently joined the Textile Conservation Centre as Research Fellow in Conservation Science, with a particular interest in plant fibres and weighted silks; he also taught the conservation science component of the Centre’s MA course in textile conservation. In 2009 he joined the British Library as Conservation Scientist. Conservation Scientist - The British Library - St Pancras, 96 Euston Road - London NW1 2DB, United Kingdom - [email protected]

ICOM-CC Leather and Related Materials WG Offenbach 2012

Morphological Assessment of Shrunken Parchment Fibres at Microscopic Level

Kathleen Mühlen Axelsson, René Larsen, Dorte Vestergaard Poulsen Sommer, Steen Kristian Frank

Abstract When subjected to hydrothermal degradation, new and historical parchment undergo morphological transformation of the fibre structure detectable at a microscopic level. By combining shrinkage temperature measurements with a visual fibre assessment performed at microscopic level, conservators have an excellent tool for evaluating the condition of the parchment object. In this study a microscopic visual assessment of fibres in water was performed from newly produced calf parchment, both for unaged samples as well as for samples that were accelerated heat aged (120° C for 96 hours). Furthermore, visual assessment was carried out on accelerated heat aged fibres exposed to a shrinkage process (from 25° C to 85° C) in both regular transmitted light as well as in polarized light. The semi-crystallinity of undeteriorated collagen transforms into amorphous areas during deterioration, as can be observed by the loss of birefringence with increasing deterioration. Furthermore, it is obvious that the oxidation in this case transforms the fibres into thicker fibres with a more flat appearance.

Keywords Parchment, Collagen, Fibres, Degradation, Hydrothermal stability, Assessment, Morphology, Microscopy, Polarization, Birefringence

Introduction

Conservators daily work involves decisions about the best suitable preservation and sometimes restoration treatments for parchment objects in archives, libraries and museums. As for all culture heritage objects, it is of great importance that these decisions are based on scientific approaches. From previous studies we now know that moist treatment and/or exposure to elevated levels of humidity, even at room temperature, may accelerate the transformation process and is likely to cause irreversible changes of the fibre structure and gelatinization (Larsen et al., 2011). Shrinkage temperature measurements, representing the hydrothermal stability of the collagen, resembles the degradation taking place in naturally aged parchment, and is a recommended routine method for analyzing parchments prior to any anticipated treatments. However, shrinkage temperature measurements must not be the sole indicator of the level of deterioration (Larsen et al., 2002, pp. 165 – 180). Together with a microscopic fibre assessment performed in water at room temperature, morphological features and changes of the fibre structure can be detected (Larsen et al., 2002, pp. 69 -72). The small sample amount required (0.1 - 0.3 mg) makes the two methods appropriate for evaluating the impact of restoration methods and, when trained in these methods, conservators have an excellent combined diagnostic tool for assessing the level of degradation of parchment objects.

Parchment and Degradation

Parchment is an animal hide that has been wetted, stretched under tension to dry and scraped thin, and where the hairs have been removed with alkaline solutions. The main composition is the fibrous protein collagen type (I).

With its alternating left-right handed winding throughout the whole structure, collagen fibres may resemble the structure of a rope. In an undamaged fibre the firm structure with helical strands are seen with certain regularity in the microscope. Apart from the undamaged fibre structure, so far eight breakdown characteristics for parchment fibres have been detected at microscopic level within the IDAP project: fraying, splitting, flat fibres, cracks, pearls on a string structure, bundles of fibres/fragments, a gel-like fibre characteristic, and finally dissolved/melted collagen (Larsen et al., 2007, pp. 17 – 21; see also Larsen et al. within this publication). Most often more than one of the fibre characteristics is present in various degrees in the sample. Furthermore, individual fibres often show more than one characteristic along its length.

ICOM-CC Leather and Related Materials WG Offenbach 2012

One of the morphological degradation features that we have focused on recently is the so-called pearls on a string structure. This breakdown characteristic can sometimes be observed on the fibre in water at ambient room temperature at microscopic level. Furthermore, we have noticed that during a hydrothermal increase, fibres may start to gradually wind up into this feature as the degradation proceeds. In between two pearls, a twisting of the fibre is found and two halves of adjacent pearls are known as a butterfly structure.

We have also found that the pearls develop into a more compact nature (i.e. their length (Pl) decreases and their width (Pw) increases ) as the degradation proceeds, with the lowest Pl/Pw ratio of 2.3 reflecting the final stage of shrinkage (the most degraded structure). Additionally, we have measured the relationship between the length of a butterfly (Bl) and the width of the its twist (Tw), whereby Bl/Tw = 17.4 , and which, we believe, is correlated, at the molecular level, with the ratio of 18.0 calculated by dividing the distance between two charged regions in the peptide chain (i.e. sites prone to degradation) with the diameter of a hydrated collagen molecule (Mühlen Axelsson et al., 2012).

The deterioration of collagen is divided into two main pathways: a hydrolytic breakdown, caused by acidic and humid conditions, and an oxidative breakdown typically occurring under dry and heated conditions. Although there are indications that hydrolysis leads to gelatinized and dissolved fibres and that oxidized fibres show sign of fragmented pieces with cracks and brittleness, it is not fully known whether the fibres follow significantly different pathways during the course of breakdown.

Prior studies using the micro-hot table method (MHT) in combination with a polarizing microscope have shown a loss of birefringence of fibres from skin and semi-tanned leather during shrinkage (Young, 1990). In this study we combined the MHT method with a visual assessment of unaged as well as accelerated heat aged parchment fibres at microscopic level both with transmitted light and polarized light.

Method

Fibre Assessment The fibre assessment here performed is a visual and detailed analysis of the fibres morphology, where every single fibre in the viewing field in the microscope is examined for its characteristic along its length. The result is reported as amount of damage in %.

Fibres are scraped from the flesh side of the parchment and placed on a microscope slide with concavity. Fully immersed in demineralized water, the fibres are left to soak in the water for at least 10 minutes to secure full penetration within the structure. The fibres are carefully separated with fine preparation needles, transferred to a flat microscopic slide and in excess water covered with a cover glass. The sample is observed under the microscope with regular transmitted light using a magnification between 50 to 500 times. Around 10 to 15 fibres from four different locations on the microscopic slide have been documented by digital photographs and were used for the visual assessment presented in this paper.

Micro Hot Table (MHT) Method After completing the visual fibre assessment as described above, the fibres may be re-used for shrinkage temperature measurements following the protocol described by Larsen et al. (2002, pp. 55 – 62) by transferring them to a microscope slide with concavity. This method can only be used if the fibres have not been left to dry out. If the choice is to collect new fibres for shrinkage temperature measurement, these must derive from the same area on the parchment as for the microscopic fibre assessment. Due to the inhomogeneous character of the material and to be able to correlate results from different types of analyses, all sampling must be done within a specific area of 2 cm in diameter. The shrinkage process is controlled with a processor and heated at a rate of 2° C/min with a start temperature of 25° C and an end temperature of 100° C. To get statistical accuracy three measurements with a Ts value of ± 2° C is acceptable. However, by connecting a digital camera to a microscope and recording the sequence, the process may be observed many times and two measurements with a Ts value of ± 2° C is enough.

ICOM-CC Leather and Related Materials WG Offenbach 2012

Fibre Assessment in Polarizing Microscope The methods of the MHT and the visual analysis of the fibre morphology can be combined in a polarized microscope by removing the slide from the hot table during shrinkage at a chosen temperature and transferring the fibres to a flat slide. The fibre sample in this case has been observed with crossed polarizers with a magnification between 50 to 500 times. At least three different locations on the microscopic slide have been documented by digital photographs. To get an accurate representation of the birefringence, it is important that the fibre sample is viewed on the stage with crossed polarizers both in a 0° angle and a 45° angle.

Results

A sample of new calf parchment was subjected to accelerated heat ageing in a ventilated oven (120° C, 0% RH) for 96 hours. Previous studies with steric exclusion chromatography (SEC) have shown that 24 hours of artificially heat ageing at this temperature is enough to create cross-linking in parchment. Furthermore, heat ageing at 120° C is recognized as being very close to natural ageing for parchment (Juchauld & Chahine, 2002). Microscopic fibre assessment of the accelerated aged sample as well as of the unaged sample deriving from the same specific area was performed in water at ambient room temperature (25° C). Studies of birefringence in polarized microscope with crossed polarizers were performed on the fibres just mentioned as well as on accelerated heat aged (oxidized) fibres subjected to hydrothermal increase with the MHT method to 45° C, 65° C and 85° C respectively (see Figure 1). By using crossed polarized filters we expected a loss of birefringence with proceeding hydrothermal deterioration (Young, 1990).

Figure 1: Visual assessment of parchment fibres. © Kathleen Mühlen Axelsson, School of Conservation, Copenhagen, July 2012. 92

ICOM-CC Leather and Related Materials WG Offenbach 2012

The main shrinkage interval of the accelerated-aged sample lies between 34.7° C and 47.1° C (s 0.64; ΔT 12.4; n 2). The unaged sample deriving from the same specific area as the aged sample has a shrinkage interval between 58.3° C and 65.7° C (s 0.57; ΔT 7.4; n 2). The first observed shrinkage (Tfirst) for the unaged sample is 46.5° C, while it is 30.9° C for the heat aged sample (see Table 1).

Result of the fibre assessment give an amount of damaged fibres for the accelerated-heat-aged sample of 78.3%. The corresponding figure for the unaged sample is 29.4%. The minor damage in the unaged sample is mainly represented by pearls on a string (around 20%) and some splitting (7%). Although caution is taken when separating the fibres in water, it is difficult to say whether the latter is a consequence of the preparation needles. However, then the same amount of fraying should be the case for the aged sample and here we only find around 1% of frayed fibres. When observing fibres from the heat-aged sample at room temperature, it is significant that around 52% of the fibres consist of flat fibres and around 20% consist of pearls on a string structure. Accordingly, the two samples can be placed in the following categories as seen in Table 1.

Unaged Damage category Heat aged sample Damage category sample Unaged sample (120° C, 96 hrs) Heat aged sample Ts - Tend 58.3° - 65.7° C 1 34.7° - 47.1° C 4 (undamaged) (heavily damaged) Tfirst 46.5° C 1 30.9° C 4 (undamaged) (heavily damaged) Fibre damage 29.4% 1 78.3% 4 (undamaged) (heavily damaged) Average 1 4 damage category (undamaged) (heavily damaged)

Table 1: Damage category for the unaged and the heat aged (oxidized) sample respectively. The average damage category is based on Ts,Tfirst and amount of damaged fibres. Definition of damage category by IDAP (www.idap-parchment.dk).

Discussion

Ts of new limed hide normally lies around 60° C. The low Ts of the heat-aged (oxidized) sample (34.7° C) compared to the Ts of the unaged sample (58.3° C) is a clear indication of a deteriorated sample of the former. Furthermore, the length of the shrinkage interval illustrates a more homogenous fibre mass for the unaged sample (ΔT 7.4) than for the oxidized sample (ΔT 12.4).

As can be seen in Figure 1, the unaged parchment fibres show clear and visible birefringence. From the visual assessment, it is obvious that the oxidation affects the morphology of the fibres to a thicker and more flat characteristic. With an increasing deterioration due to hydrothermal shrinkage this becomes more significant. In previous research we have concluded that a hydrothermal degradation from 25° C to 95° C leads to a reduction of more than half of the total fibre length and, depending on the direction of the skin, and an average of 45% shrinkage for whole parchment pieces (Mühlen Axelsson et al., 2012). It is worth mentioning that the first picture of shrunken fibres from the heat-aged sample is taken at 45° C at which temperature the shrinkage has almost terminated (Ts 34.7° - 47.1° C).

Earlier research has also revealed how fibres curl up into a pearls on a string structure during the course of shrinkage (Mühlen Axelsson et al., 2012). Whether pearls on a string structure is a morphological appearance originating from a certain type of degradation and that transformation into flat broad bands represents a morphology typical for oxidative deterioration needs further research. Although the exposure to dry heat ageing has a clear deteriorating effect on the morphology of the fibres, the heat-aged (oxidized) fibres still show some birefringence at room temperature. However, as the degradation escalates due to hydrothermal increase with the MHT method, the parchment fibres show progressively more loss in birefringence, a phenomena also reported for collagen fibres in skin and semi-tanned leather (Young, 1990). Some regions of the peptide chain are likely to be more disposed to oxidative and hydrolytic degradation, namely those consisting of charged amino acid such as Asp, Glu, Arg and Lys (Larsen, 1995). Although it is 93

ICOM-CC Leather and Related Materials WG Offenbach 2012

not yet known if hydrolysis or oxidation lead to different breakdown characteristics of the fibres, it is our assumption that charged regions in the peptide chain sensitive to degradation may be found in the middle of the pearls and that stable residues are located in the twisted regions (Mühlen Axelsson et al., 2012). To further support our theory, studies of collagen fibres with pearls on a string structure need to be analyzed in polarized microscope.

Conclusion

Measuring the hydrothermal stability (shrinkage temperature) alone does not fully reflect the various degradative conditions of parchment or other collagenous material; a more comprehensive assessment can be obtained by also assessing the microscopic morphologies of the fibres when immersed in water at room temperature. Previous research has proposed this morphological assessment which involved measuring the dimensions of pearls on a string structure, one of eight features found to be characteristic in the degradation of parchment (see Mühlen Axelsson et al., 2012). Determination of 'the main shrinkage interval', 'the first shrinkage event observed', and 'the amount of damaged fibres', respectively, in the sample provide an improved tool for assessing the material's extent of degradation.

In this study, dry heat-aged and non-aged samples of parchment were studied, which provided a range of fibre degradation morphologies. One of the eight characteristic features for degraded parchment fibres, the 'flat and unfolded characteristic', was represented in greater number in the dry-heated sample. Changes of the morphology of the 'flat' fibre characteristic were observed by studying the birefringence at different levels of deterioration due to hydrothermal increase (loss of birefringence is known to be linked to loss in crystallinity and degradation). Dry heat-aged parchment fibres were found to still show some birefringence at room temperature. The methodology adopted in this study successfully combines hydrothermal stability and fibre morphology assessments, by analysing the fibre characteristics before and during hydrothermal shrinkage with regular transmitted light aswell as with polarized light. Loss in birefringence was found both at crossed polarizers at an angle of 0° and when turning the stage 45°.

Materials Axio Scope A.1 with AxioCam camera, Zeiss. www.zeiss.com Calf parchment produced according to our specifications, Pergamena. www.pergamena.net FP90 Central Processor, Mettler Toledo. www.mt.com FP82 Hot Stage, Mettler Toledo. www.mt.com Infinity 1-3 digital camera, Lumenera. www.lumenera.com Studio capture 3.1.1., Studio86Designs. www.studio86designs.co.uk

References Juchauld, F., Chahine, C. 2002. Determination of the Molecular Weight Distribution in Parchment Collagen by Steric Exclusion Chromatography. Microanalysis of Parchment. Larsen, R. (ed.). Archetype Publications Ltd., London, pp. 123 – 131. Larsen, R. 1995. Fundamental Aspects of the Deterioration of Vegetable Tanned Leathers. PhD thesis, University of Copenhagen, The Royal Danish Academy of Fine Arts, School of Conservation (Publ.), Copenhagen. Larsen, R., Poulsen, D.V., Vest, M. 2002. The Hydrothermal Stability (Shrinkage Activity) of Parchment Measured by the Micro Hot Table Method (MHT). Microanalysis of Parchment. Larsen, R. (ed.). Archetype Publications Ltd., London, pp. 55 – 62. Larsen, R., Poulsen, D.V., Odlyha, M., Nielsen, K., Wouters, J., Puchinger, L., Brimblecombe, P., Bowden, D. 2002. The Use of Complementary and Comparative Analysis in Damage Assessment of Parchments. Microanalysis of Parchment. Larsen, R. (ed.). Archetype Publications, London, pp. 165 – 180. Larsen, R. 2007. Introduction to damage and damage assessment of parchment. Improved damage assessment of parchment (IDAP). Asessment, data collection and sharing of knowledge. Larsen, R. (ed.). European Commission, Research Report no. 18, European Communities, Brussels, pp. 17 – 21.

ICOM-CC Leather and Related Materials WG Offenbach 2012

Larsen, R., Poulsen, D.V., Minddal, K., Dahlstrøm, N., Fazlic, N. 2007. Damage of parchment fibres on the microscopic level detected by the micro hot table (MHT) method. Improved damage assessment of parchment (IDAP). Asessment, data collection and sharing of knowledge. Larsen, R. (ed.). European Commission, Research Report no. 18, European Communities, Brussels, pp. 69 – 72. Larsen, R., Sommer D.V.P., Mühlen Axelsson, K. 2011. Scientific Approach in Conservation and Restoration of Leather and Parchment Objects in Archives and Libraries. New Approaches to Book and Paper Conservation-Restoration. Engel, P. et al. (eds.). Verlag Berger Horn, Wien, pp. 239 - 258. Mühlen Axelsson, K., Larsen, R., Sommer, D.V.P. 2012. Dimensional studies of specific microscopic fibre structures in deteriorated parchment before and during shrinkage. Journal of Cultural Heritage 13, pp. 128 – 136. Young, G. S. 1990. Microscopical Hydrothermal Stability Measurements of Skin and Semi-tanned Leather. ICOM Committee for Conservation, 9th Triennal Meeting, Dresden, German Democratic Republic 26 – 31 August 1990, pp. 626 – 631.

Biographies Kathleen Mühlen Axelsson holds a Master of Conservation-Restoration Science (MSc) from the School of Conservation in Copenhagen. Currently working as a Research Assistant and PhD student on leather and parchment in the ongoing EC project MEMORI (Measurement, Effect Assessment and Mitigation of Pollutant Impact on Movable Cultural Assets - Innovative Research for Marker Transfer) at the School of Conservation in Copenhagen. Co-organiser and workshop tutor at several IDAP (Improved Damaged Assessment of Parchment) courses in Europe. Has been working as a Research Assistant on various leather and parchment projects for the School of Conservation in Copenhagen since 2007. Worked as a conservator at the Regional Archives in Lund, Sweden between 2002 - 2007, where she was Head of the Conservation and Bookbinding Department between 2004 - 2007.

René Larsen is associate Professor at the School of Conservation in Copenhagen. Between 1996 – 2012 rector of the School of Conservation in Copenhagen. Has a background as a Professional Bookbinder, a Master in the Science of Conservation (1986) and a PhD in Biochemistry (1995). Has been coordinator of the following EC projects: STEP Leather; Deterioration and Conservation of Vegetable Tanned Leather (Environment Leather project); Micromethods for the Analysis of Parchment (MAP); Improved Damaged Assessment of Parchment (IDAP). Organizer of and tutor in several IDAP assessment courses. 2002 – 2003 member of the Committee for Physical Cultural Heritage for reporting to the Danish Parliament. 2004 – 2005 member of the working group regarding recommendations for handling, storage and exhibition of cultural heritage objects, the Danish National Cultural Heritage Authority. 2004 – 2009 member of the national working group on mass deacidification of paper, The Danish National Library Authority. Leading scientist in leather and parchment and Head representative for the School of Conservation in the ongoing EC project MEMORI.

ICOM-CC Leather and Related Materials WG Offenbach 2012

The Leather Furnishings in Palazzo Chigi in Ariccia: Documentary Sources Mara Nimmo, Mariabianca Paris, Francesco Petrucci

Abstract Though originally all the Chigi family’s country residences in the Baroque period had numerous leather furnishings, only the Palazzo in Ariccia (27 km south of Rome) has retained a significant part of this patrimony, with some wall hangings still in the same rooms for which they were conceived and some taken down or moved to other rooms; the older pieces, though, only survive in fragmentary condition. The examination of ac counts and of inventories permits the dating of most of the leatherworks, and the identification of many of the artisans involved in their execution and installation, such as coramari (leatherworkers), carvers of the wood moulds and painters in charge of specific decoration. The research has also enriched our knowledge of the historical technical terminology dealing with these types of artefacts.

Keywords Gilt leather, damasked leather, Roman Baroque leather furnishings, wall-hanging, furniture cover, door-curtain, leatherworker, wall-covering, flocked leather.

Introduction

Palazzo Chigi in Ariccia represents a rare example of an unaltered Baroque country residence (Figure 1), documenting the prestigious role played by one of the greatest of the Italian Papal dynasties, the Chigi. The building was constructed by the Savelli family in the 16 th century and passed into Chigi hands in 1661 together with all their lands. The new joint owners were the family of Pope Alexander VII: his brother, Prince Mario and his two nephews, Prince Agostino and Cardinal Flavio. After having the village radically rebuilt by Gian Lorenzo Bernini, from 1666 to 1672 the Chigi also significantly enlarged the Ariccia Palace1. This reconstruction was executed by the young architect Carlo Fontana using plans drawn up by Bernini and by 1673 the new Palazzo had been lavishly decorated. Further enlargements and extensive decoration took place during the period 1740 -43.

The palace was ceded by the Chigi family to the town of Ariccia in 1988 and restored for public use. The past twenty-two years have seen the conservation of the architecture, maintenance and/or conservation and restoration of related paintings, sculptures and rich furnishings. Among the latter there are several examples of gilt leather wall hangings and furniture covers, complete or fragmentary, which document the extensive use of leather in decorating the country dwellings during the Baroque period.

In order to gain further insight into this singular type of artefact, a study was conducted in the Chigi Archives held in the Archivio Segreto of the Biblioteca Apostolica Vaticana (BAV).

Figure 1: Palazzo Chigi in Ariccia, south facade (© Palazzo Chigi, Comune di Ariccia).

ICOM-CC Leather and Related Materials WG Offenbach 2012

Leather Furnishings and Their Makers The Chigi’s Wardrobe inventory dated 1672 and 16732 provides an overview of the decorations and leather works within the palace upon completion of the renovation works (Figure 2). It includes quite detailed descriptions of portières, furniture covers, as well as wall hangings in gilt and painted leather, in some cases embossed or flocked, etc. On the piano nobile were mounted leather hangings specially made for the palace (parati di corame nuovi) covering seven rooms in the east wing (including the apartment of the Prince) and the chapel facing the Sala Maestra, some still existing also in their original location.

Old leather hangings (parati di corami vecchi) were readapted in the twelve rooms on the second floor, which held the guest rooms and the women’s apartment, as well as in two rooms in the south wing of the ground floor3. None survive in place today, and they are described so generically they cannot be linked with any certainty to various surviving fragmentary pieces of leather.

Figure 2: Plans of the piano nobile and the mezzanino: location of the leather furnishings according to 1672 Wardrobe Book. Wall- hagings (in yellow), portières (P) and covers for: altar (A); box-bed (B); chair (C); object (O); table (T) (©ISCR).

On the ground floor, the “summer apartment” of Ø 5,8 Cardinal Flavio Chigi in the south-west wing Antonio Mugnaione followed a different decorative scheme, being cm decorated with taffeta according to the French style of Louis XVI’s court. The Inventario of the Cardinal4 recorded only a few leather covers Ø 5 cm Agostino Nespola 5 placed on tables and box-beds , protecting the more highly prized textile covers. Ø 5,5 Giuseppe Montori The Chigi account books from 1656 to 1682 cm permits dating of most of the new leatherworks and the identification of many of the artisans involved in their execution and installation (Table 1). Comparison with the 1672 Wardrobe book and Ø 5 cm Mattia Turchi later Inventories (1705, 1744, 1777) also enables us to identify the original location of numerous artefacts, as well as their provenance and later re- Ø 7 cm Paolo Ridolfi locations or adaptations.

Table 1: Stamp impressions with the initials of the coramari found on the rear of some leather furnishings.

The artisan principally responsible for the new leather furnishings of the Palazzo in these years is Agostino Nespola, a coramaro (leatherworker) from Santa Lucia in Banchi in Rome, who works almost uninterruptedly

ICOM-CC Leather and Related Materials WG Offenbach 2012

in the service of Prince Agostino from 1666 to 1682. He also supplied leatherwork for other Chigi residences such as the palaces in Rome (mainly the palace in Piazza Colonna), the Villa delle Volte near Siena and Villa del Voltone in Farnese6. Payments also mention maintenance, repair and relocation into new rooms of already existing pieces: such work comprised a significant part of his activity7.

The place name associated with his name in the invoices indicates his workshop was in the old via Florida in Rome (corresponding to the present-day via dei Banchi Vecchi and via del Pellegrino), which was the main route for pilgrims and foreigners travelling to St. Peter’s and the Apostolic Palaces and where, from the 15th century onwards, the main trades of the city were located, including the slaughterhouses with all their related activities8. Many other gilt leather workshops were located in this area in the 16th and 17th centuries9.

In November 1670 Agostino Nespola provided the Ariccia Palazzo with corami alla veneziana (Venetian style gilt leather), so called for the presence of multicoloured flowers in their decoration. These adorned three rooms of the Prince’s Apartment, including the bedroom of Don Agostino 10. Made of roughly 900 skins, today only a part survives, reused for a room in the 18th century wing of the Palazzo. The motif, embossed in shallow relief with flowers in vases in oval frames, is treated with a naturalistic style characteristic of the Baroque period, enriched with figures of birds and insects (Figure 3).

Figure 3: Decorative motif of the Venetian style wall-hangings, composed of four skins. (©ISCR, B. Malter).

A fourth hanging employing the same pattern but with different colours – gold on a turquoise background – was made for the room near to the north-east tower in the same apartment. Only a few fragments of this hanging survive, but its design is preserved in painted decoration on the wall s which faithfully copies it.

A commission, paid in May 1671, for “lavori diversi di corami di pelli in tt° n° 731 1/3 fatti p diverse stanze” (“various works in decorated leather totalling 731 1/3 hides made for various rooms”)11 almost certainly relates to this hanging together with the green and gold leather wall-hanging with a brocade style of decoration (a broccatelli), made for the bedroom of Princess Maria Virginia Borghese in the south-east tower. It was moved at the beginning of 20th century to the north-east tower, the previous Libreria, where it is still.

In March 1673 Nespola was paid for “large silver skins” (pelli di argento)12 sent to Ariccia to make table covers and in December of the same year he supplied the Wardrobe with a gilt leather wall hanging for the Chapel described in detail in his invoice 13. Still in its original location it has brocade- like decoration with red flock (now faded to a yellow ochre), with the Chigi coat of arms painted on the corner columns (Figure 4).

ICOM-CC Leather and Related Materials WG Offenbach 2012

Figure 4: Piano nobile, Chapel (© Palazzo Chigi, Comune di Ariccia, D. Petrucci), flock leather wall hanging (©ISCR, B. Malter).

Further payments to the coramaro which are certainly for Ariccia date later to 1678-1680 for the supply of decorated leather and other, undefined, work 14.

Gilt leather wall hangings for the furnishing of the new Palazzo were not only brought from Rome, but also made on-site. Account books document payments to various artisans involved in the manufacture of the coverings for the first and second antechamber (Figure 5) next to the Sala Maestra which remain today in their original rooms (Sala del Trucco and Sala Borghese). Their motive of flower vases within an oval surround made from knotted ribbons has a rigorous style recalling Renaissance patterns, and perhaps replicates a design found on some older hangings in the Palazzo.

Figure 5: Piano nobile, first antechamber (today Sala del Trucco), one of the two gilt leather wall hangings made directly in the Palazzo. (© Palazzo Chigi, Comune di Ariccia, D. Petrucci).

In June 1671, a consignment of 550 skins was bought from a merchant Pietro Passarini15, and in August the engraver Paolo Franceschi made two wooden stamps in walnut for impressing decorative patterns16, preserved in the Palazzo (Figure 6). Father Agostino from Gubbio was working on the hangings with two assistants, Antonio Piacentini and Giovanni Garofalo, and various other helpers from January to March 1672, and perhaps even earlier, since November 167117. What Padre Agostino’s exact role in this work was, and indeed how much he contributed to it is not however clear. The relatively modest monthly payment –

ICOM-CC Leather and Related Materials WG Offenbach 2012

four scudi at month for him and his assistant – and his lack of qualification as coramaro or maestro, would seem to suggest he had a subordinate role under some other leatherworker – perhaps Agostino Nespola18 – although not mentioned in the documents. Various other payments attest to the presence of Father Agostino in Ariccia from May 1671 to April 1672, busy cleaning and mounting old hangings on the mezzanine floor19. Among these is a payment for three jugs of wine (about six litres) provided by the cellarman Filippo Giunti to clean old leather hangings in two rooms in the old tower.

Figure 6: Wooden stamps used for the wall hangings of the first and second antechamber. Decorative motives of the friezes and of the vertical bands are carved on the same wooden stamps respectively on the front and the back side (©ISCR, B. Malter).

The 1705 inventory of the palazzo20, following the death of Prince Agostino, confirms the locations of the furnishings on different floors and also provides an appraisal of their value: it is striking how low these assessments are (about 30 scudi for a hanging made of 350 hides) compared to those of the damask wall coverings which decorated, for example, the west wing of the piano nobile (150 scudi and more). A clearer example is the hanging in the Chapel: Nespola had originally been paid 38.14 scudi for it; now all the Chapel’s furnishings (candlesticks, silver chalices, etc.) including the hanging were valued at only 20 scudi.

The first documented changes in the leather hangings’ arrangement occurred between 1705 and 174421, when construction work finished on the north west “tower wing” (braccio del torrione) and a new inventory was compiled.

Figure 7: Plan of the ground floor: location of the leather furnishings according to 1744 Inventory. The Cardinal’s apartment is on the left side of the entrance. Wall-hagings (in yellow), portières (P) and covers for box-bed (B) and table (T) (©ISCR).

In the six rooms of the cardinal’s apartments on the ground floor (Figure 7), the taffeta wall coverings were replaced by red and gold leather hangings, which remain in two rooms ( Stanza del Toro and Stanza dell’Ariete). These hangings, from their typically Baroque decorative patterns 22, almost certainly came from the SS. Apostoli palace23, the Roman residence of Flavio Chigi, which at his death in 1693 passed out of the family’s hands, but whose furniture and collections were moved to his cousin Agostino’s Palazzo in Piazza Colonna in Rome. The round stamps with the initials GM and MT on the back of the skins most likely indicate the Roman leatherworkers Giuseppe Montori and Mattia 100

ICOM-CC Leather and Related Materials WG Offenbach 2012

Turchi, payments to whom can be found in Cardinal Flavio Chigi’s account book for the years 1687- 9024. Giuseppe Montori, whose workshop was located in the Cesarini area, is also documented some years later as working for the Reverenda Camera Apostolica 25.

Another change concerns the piano nobile where one of the “venetian style” wall hangings (in the room following the second antechamber) was replaced by another hanging coloured gold and silver with flowers and winged putti embossed in low relief, which is still there now (Figure 8). We know nothing about the provenance of this hanging, though it bears Nespola’s seal on the rear.

Finally, on the second floor, three rooms – two in the old tower and the corridor leading to the Princess’ room in the new apartment – were decorated with leather hangings with alternating red, green and gold skin stripes. The description in the inventory means we can establish a link between these hangings and the one which today is in the Summer Dining Room, with brocade patterns identical to those in the Chapel. There is a detailed invoice by Agostino Nespola from November 167426 in the Chigi accounts for a similar piece made for the palace in Piazza Navona, which might be the very same hanging. The last 18th century inventory from 1777 lists no further changes27.

Figure 8: Piano nobile, Sala Mario dei Fiori, wall hanging which replaced, between 1705 and 1744, one of the “venetian style” coverings in the Prince’s apartment. The design with flowers and winged putti is embossed in low relief (©ISCR, B. Malter).

The Chigi accounts books document the activity of two other gilt leatherworkers, Antonio Mugnaione and Paolo Ridolfi. Invoices and payments concerning Mugnaione refer to Mario Chigi from 1656 to 166128 and to his son cardinal Flavio from 1671 to 167329. In the Ariccia palace two table covers which were located in the Cardinal’s apartment on the ground floor are conserved in fragmentary condition. Ridolfi is mentioned in Prince Agostino’s account book in October 1671 for the payment of a red and gilt leather wall hanging of 309 skins 30. It is unlikely that it was originally made for the Ariccia palace, since it is not mentioned in the list of new furnishings of the Wardrobe compiled in 1672. However in the palace there are fragments of a similar artefact with a decorative pattern embossed in low relief, presenting on back side the initials P.R. A recent archival research documents the activity of both coramari also in the Apostolic palaces respectively under the pontificates of Alexander VII and Clemens X31.

Types and Structure of the Artefacts: Historical Technical Terms

The Chigi account books, together with the invoices of the leatherworkers, besides providing important information on the types of leather furnishings and their composition, represent a precious source for the technical terminology used in the Rome area in the second half of the 17 th century.

The central term corame – the only form employed in the documents – is used to indicate both the material, decorated leather (or sometimes simply leather), as well as an artefact made of leather32. This second usage frequently appears in the inventories principally to indicate covers, upholstery and other coverings in leather: sopracorami di tavola, corami di letti a credenza, etc.

101

ICOM-CC Leather and Related Materials WG Offenbach 2012

The leatherworks (corami) are made up of a number of skins (pelli) whose cost is calculated on the basis of their finish and size (Table 2). The accounts examined here distinguish between pelli d’oro and pelli di argento (gold and silver skins) (decorated with silver leaf and gilded with gold varnish), and pelli rosse (red skins), often described as having a checkerboard pattern (lavorate in scacchi) and used for covers and upholstery. The survival of several of these pieces provides significant support for the idea that pelli rosse indicated skins employing the technique of damasked leather33. Unfortunately, the skins were tinted with organic colorants (principally lakes) which are sensitive to light, thus, in most cases, the paint has faded and the leather has returned to its natural colour.

Type of skin Cost in baiocchi* Vegetable tanned skin** (pelle conciata) 12 b.

Red skin (pelle rossa) 15-16 b. Gold skin (pelle d'oro), normal size 22-23 b.

Gold skin, normal size, for more complex works 26 b.

Gold skin, large 40 b. Gold skin, large, flocked 45 b.

Silver skin (pelle di argento), large 60 b.

* The baiocco is a hundredth of the Scudo, the currency in circulation in the Papal State up to 1866. ** The cost of vegetable tanned skin is deduced by the payment to the merchant Pietro Passerini (see note n.15)

Table 2: Invoices of Antonio Mugnaione and Agostino Nespola (1663-1674): unit cost of skins according to their size and type of finish.

The accounts distinguish also between ordinary sized skins (pelli di misura ordinaria) and large or outsize skins (pelli grandi or fuor di misura) which, if we check surviving pieces, measure approximately 54 x 40 cm and 85 x 51 cm respectively.

Included in the cost of the pelli d’oro is the application of the decorative pattern (by hand or using stamps) by the leatherworker. More complicated compositions would entail the commissioning of an outside painter, who was generally paid by the leatherworker 34.

There are essentially three types of leatherwork mentioned in the Chigi account books and inventories: hangings, portières and covers (of various types) for furniture and other objects, though there is a single mention of a canopy.

In the documents examined, the hangings (mentioned as parato di corame or also, by extension stanzia di corame) are described according to the number of skins used in the various components that make them up: strips, columns, inserts, friezes and openwork/lace 35.The pieces which cover the areas above doors, windows and fireplaces, composed of varying numbers of skins, are itemised separately. Frequently in the accounts the costs are calculated for the chiodi (nails) and bullette (tacks), the latter being nails with broad, round, slightly convex heads which were also used as decoration. This suggests that the hangings were intended to remain permanently on the walls and there was no intention of removing them according to the season. Very few remnants of the original anchoring system survive today. Hooks made of twisted leather which are occasionally found loosely sewn onto the back of the upper side were presumably used to hang the artefacts to the wall before nailing them.

102

ICOM-CC Leather and Related Materials WG Offenbach 2012

The account books also mention payments for portières which were also originally located in the palazzo. They were generally decorated in a similar way to the hangings which they we re hung with, though heraldic subjects, framed by friezes, were also popular. In these latter cases the decoration was carried out by painters36. On their backs the portières had a lining (fodera) made of cloth (for example, fine linen canvas) or leather (red, or more rarely, white) and they were hung on a loose iron rod held above the door by two rings fixed into the wall. In the inventories the portières are always described with their accompanying rod and rings (ferri and occhietti).

An important part of the leatherworkers’ output are the covers (coperte or sovracoperte) for tables, chairs, box-beds, billiard tables, harpsichords, chests, clocks, silver vases, etc. made with both pelli rosse and pelli d’oro, often combining cloth with leather. Numerous variations exist. There are covers entirely made of pelli rosse, with strips of gilt leather (orletti di oro) or braids decorating the borders (Figure 9). Covers for tables or box-beds (Figure 10) usually have a panel of pelli rosse bordered with a fringe (frangia) in gilt and painted leather, with at times cascades (cascate) either in gilt leather or cloth. More rarely the covers are made entirely of pelli d’oro or pelli di argento.

Figure 9 (on the left): Cover for a velvet high chair made of ‘red skins’ (now faded), with strips of gilt leather along the border. Detail of the cover (©ISCR). Figure 10 (on the right): Cover for box-bed made by Agostino Nespola: front side. It is composed by an upper panel of ‘red skins’ with a checkerboard pattern, a fringe and three cascades in gilt leather which is the same used in the bedroom of Princess Maria Virginia Borghese (©ISCR, B. Malter).

Conclusions

Upon completion of restructuring in 1666-72, as was the fashion of the time for country residences, Palazzo Chigi in Ariccia was lavishly decorated with rich leather furnishings such as wall hangings, portières and various furniture covers, several examples of which still survive today. Twenty three rooms on different floors were covered with gilt and painted leather which the 1672 Wardrobe b ook separates into old and new.

The examination of the account books of the Chigi family led to the identification o f the artisans who worked on the furnishings made especially for the Palazzo. Agostino Nespola, coramaro in Rome, who established a long-term and well-nigh exclusive working relationship with prince Agostino Chigi, Father Agostino from Gubbio and his assistants who worked directly in the palace for a brief period and Antonio Mugnaione, in the service of Cardinal Flavio, who had a marginal role in providing furnishings for the palace. In Ariccia they were also involved in the maintenance, repair and installation of old leather wall hangings. The documents also mention other leatherworkers whose pieces are now held in the Palazzo, though these were transferred there from other Chigi residences.

The comparison of inventories from 1672 to 1777 also enabled us to identify the original locations for many furnishings, as well as later modifications and relocations.

103

ICOM-CC Leather and Related Materials WG Offenbach 2012

Finally this archival research provided deeper insight on the types, construction techniques and value of leather furnishings in the Baroque period as well as on the historical use of technical terminology.

Acknowledgments The authors would like to thank Mark Gittins for his help in translating this report.

Endnotes and References 1. F. Petrucci, Palazzo Chigi ad Ariccia, Roma 1984. 2. BAV, Archivio Chigi 2898, Libro della Guardarobba dell’Ariccia cominciato il 18 Maggio 1672 cavato dall'Inventario posto nella filza de Conti diversi n°152. The documentary sources on which the present study is based originate mostly from the research work performed by Mara Nimmo in the BAV’s archives from 1989 to 1990. 3. Documents attest to the presence of leather furnishings in the Palazzo in Ariccia prior to the extensive modifications carried out on the ex-Palazzo Savelli. An inventory compiled before 30 September 1664 (BAV, Archivio Chigi 20477, ff. 1-8) recorded the presence of leather wall hangings in at least twenty rooms, including the bedrooms of Prince Agostino and his brother Sigismondo. 4. BAV, Archivio Chigi 702, from c. 136, Inventario de mobili esistenti nell'Ariccia (1666-67), published in A.M. Mignosi Tantillo, I Chigi ad Ariccia nel ‘600, in L’arte per i papi e per i principi nella campagna romana, grande pittura del ’600 e del ‘700, exhibition catalogue (Roma, Museo Nazionale del Palazzo di Venezia, 8 marzo-13 maggio 1990), Quasar, Roma 1990, vol. II, p. 108 (Appendix 1). 5. Box-beds are found in almost all of the rooms of the Palazzo. These are beds concealed within a wooden structure with doors, similar to a cupboard, which can be opened when needed to form a bench. They usually have inside two mattresses, a bolster and a blanket. Cf. M. Anselmi Zondadari, L’arredamento del Palazzo Chigi Zondadari dall’antico inventario del 1687, in M. Eichberg and F. Rotundo (ed.), Il Palazzo Chigi Zondadari a San Quirico d’Orcia. Architettura e decorazione di un palazzo barocco, Editrice Donchisciotte, San Quirico d’Orcia 2009, pp. 177-197 and pp. 331-350 (Appendix 5). 6. Among the numerous documents see: BAV, Archivio Chigi 43, f. 42, for Rome; BAV, Archivio Chigi 1005, f. 7, mandato n° 51, for Farnese; BAV, Archivio Chigi 1005, f. 195, mandato n° 80, for Siena. 7. BAV, Archivio Chigi 1004, f. 46, mandato n° 34 (30 gennaio 1666); BAV, Archivio Chigi 1004, f. 74, mandato n° 268 (19 ottobre 1666); BAV, Archivio Chigi 1004, f. 189, mandato n° 14 (18 gennaio 1670); BAV, Archivio Chigi 2620, c. 93 (20 ottobre 1677). 8. A. Modigliani, Mercati, botteghe e spazi di commercio a Roma tra Medioevo ed età moderna, Roma nel Rinascimento, Roma 1998, pp.200-201. The author publishes an interesting drawing annexed to a notarial act of 1668 (Archivio di Stato di Roma, Ospedale della SS. Trinità dei Pellegrini, 195), which shows a view of the houses in front of the church of Santa Lucia in via Florida. Their ground floors are occupied by workshops with masonry benches, including a coramaro. 9. Cf. A. Rodolfo, Preziose pelli: i corami dei Palazzi Apostolici Vaticani. Fonti e documenti, in C. Volpi (ed.), Vestire i palazzi. Stoffe, tessuti e parati negli arredi e nell’arte del Barocco, (in press and kindly shown by the author), pp. 27-38. 10. BAV, Archivio Chigi 1806, f. 83, uscita 388 (3 novembre 1670). An example of this style is given by the hangings purchased in Venice by Cardinal Flavio for the palace in San Quirico d’Orcia. Cf. M. Bercé, I pannelli in cuoio dorato e dipinto del Palazzo di San Quirico, in M. Eichberg e F. Rotundo (ed.), op.cit., pp. 199-207 and pp. 355-357 (Appendixes 12-16). 11. BAV, Archivio Chigi 1004, f. 236, mandato n° 82; BAV, Archivio Chigi 43, f. 68 and f. 123. 12. BAV, Archivio Chigi 1091, conto n° 51 (16 gennaio 1673); BAV, Archivio Chigi 1005, f. 7, mandato n° 51 (18 marzo 1673). 13. BAV, Archivio Chigi 1092, conto n° 39 (20 dicembre 1673); BAV, Archivio Chigi 1005, f. 41, mandato n° 39 (3 marzo 1674). For the Chapel and the flock leather technique see: M.C. Berardi, N. Nimmo, M. Paris, The Seveteenth century flock-leather wall hangings of the Chigi Chapel in Ariccia: a case study, in Preprints of ICOM-CC 9th Triennial Meeting, Dresda 1990, vol.II, pp. 611-615. 14. BAV, Archivio Chigi 1005: mandato n° 191 (14 ottobre 1678); mandato n° 13 (19 gennaio 1679); mandato n° 102 (24 agosto 1679); mandato n° 8039 (26 marzo 1680). 15. BAV, Archivio Chigi 1004, f. 241, mandato n° 119. 16. BAV, Archivio Chigi 1090, conto n° 193. 17. BAV, Archivio Chigi 2619, ff. 109 -110; BAV, Archivio Chigi 998, f. 95, pagamento 150/155; BAV, Archivio Chigi 2991, uscita n° 336, n° d’ordine 79. This document specifies a daily payment of 10 baiocchi for work days and 15 baiocchi for holidays. This, however, is low compared with the earnings of another leatherworker active in Ariccia in 1663 who received 50 baiocchi a day (BAV, Archivio Chigi 1081, f. 383, adi 19 maggio 1663). 18. On the back of some of the leather detached from the wall there appears to be a heavily worn stamp with the initials A.N.

104

ICOM-CC Leather and Related Materials WG Offenbach 2012

19. BAV, Archivio Chigi 2991: uscita n° 31; uscita n° 336, n° d’ordine 67; uscita n° 336, n° d’ordine 25; uscita n° 336, n° d’ordine 366. 20. BAV, Archivio Chigi 20552, Ariccia 1705 - Stima et inventario del Palazzo e Casini dell'Ariccia. 21. BAV, Archivio Chigi 20640, Inventario di tutta la roba esistente nel Palazzo dell’Ariccia dell’Ecc.mo Principe Don Augustino Chigi fatto sotto li 12 dicembre 1774. Some changes had already been indicated in Ariccia dopo il 1705 /3 inventari di oggetti aumentati e mancanti nel Palazzo e Casino detto del Massaroni, in confronto di altri inventari fatti nel 1679, 1705 e di altro senza data (BAV, Archivio Chigi 20553). 22. M. Fagiolo dell’Arco, F. Petrucci (ed.), L’Ariccia del Bernini, exhibition catalogue (Ariccia, Palazzo Chigi, 10 ottobre - 31 dicembre 1998), Edizioni De Luca, Roma 1998, pp.110-112. M.G. Bernardini, M. Fagiolo dell’Arco (ed.), Gian Lorenzo Bernini. Regista del Barocco, exhibition catalogue (Roma, Palazzo Venezia, 21 maggio - 16 settembre 1999), Skira, Roma 1999, pp. 402-403 (scheda 155). 23. BAV, Archivio Chigi 703, Inventario del Palazzo a SS Apostoli - Card. Flavio Chigi (ca 1670-76), f. 92. 24. Cf. M. Bercé, I pannelli in cuoio dorato e dipinto, p. 206. 25. Cf. A. Rodolfo, Preziose pelli, p. 29. 26. BAV, Archivio Chigi 1092, conto n° 190. 27. BAV, Archivio Chigi 20825, Inventario delli mobili, ed altro esistenti nel Palazzo della Terra dell’Ariccia di S.E. P.ne il Sig.Principe Don Sigismondo Chigi…li 30 ottobre 1777. 28. BAV, Archivio Chigi 446, ff. 29, 36, 54, 95. 29. BAV, Archivio Chigi 488; BAV, Archivio Chigi 458, c. 210 (185/196); BAV, Archivio Chigi 489, envelope of August 1673; BAV, Archivio Chigi 539, mandato n° 774 and mandato n° 859. 30. BAV, Archivio Chigi 1004, f. 288, mandato n° 188 31. Cf. A.Rodolfo, Preziose pelli, pp. 29, 35. 32. In the third edition of the Vocabolario della Crusca (1691), where the term appears for the first time, corame is defined as “Aggregato di cuoi. Paramento di cuoi” (Aggregation of leatherworks. Leather wall hanging). In a later edition the definition is further specified as “Cuoio lavorato e stampato, e spesso fregiato d’oro, per addobbo di camere, sale e sedie e simili” (Leather worked and stamped, often decorated with gold, to adorn rooms, halls and chairs and similar things). 33. This term means a piece of leather worked to imitate damask cloth, employing both a similar decorative repertoire and a gloss and matt textures to differentiate the background from the decorative elements. Damasked leather is made from soft skin which could be draped and wrapped into shapes as naturally as a cloth – presumably a vegetable tanned sheep or ram skin. The technique employed an unusual method of “stamping” to transfer the decoration to the leather, recorded in a 16th century source (Peder Månsson, circa 1520). The leather, coloured on the grain side, is glued flesh- side down onto a wooden block which has the designs carved into its surface and while still damp the leather is rubbed and flattened by a wooden ball operated by a weighted handle attached by a hinge to the wall. The ball tool darkens and shines the areas it works on which become the ground, as if they were polished, while the decorative elements remain lighter and more matt. 34. BAV, Archivio Chigi 1092, conto n° 39; BAV, Archivio Chigi 109, conto n°51. 35. See the glossary illustrating the technical terms in M. Nimmo, M. Paris, L. Rissotto, Cuoio dorato e dipinto. Schedatura di manufatti. Repertorio dei punzoni, Istituto Centrale per il Restauro, Roma, 2008, pp.108-111. 36. For example in 1678 the painter Andrea Alberti received a payment of 30 scudi for painting coats of arms and friezes on six gold portières, with linings of red skins, made by Agostino Nespola for the Villa delle Volte in Siena. See BAV, Archivio Chigi 1005, f.195, mandato n° 80.

Biographies Mara Nimmo has a diploma in Painting Conservation, with four years of post diploma courses in painting conservation and in training of conservators at the Istituto Centrale per il Restauro in Rome (presently ISCR). Degree in Humanities with a three year post-graduate diploma in Medieval and Modern Art History, Università degli studi di Roma “La Sapienza”. From 1975 to 1998 she was first a conservator and then an art historian at the ICR, where in 1987 she established the section for the conservation and restoration of leather artefacts and its conservation studio, which has the responsibility of carrying out studies, research and experiments in the sector. Among other duties, she taught the history of artistic techniques at the Arezzo campus of the Università degli studi di Siena up to 2005. She has published extensively on technical subjects. Via delle Costellazioni 300 - 00141 Roma, Italy - [email protected]

Mariabianca Paris has a diploma in Painting Conservation at the Istituto Centrale per il Restauro in Rome (presently ISCR) in 1980, with a one year post-diploma course. Degree in Medieval and Modern Art History at Università degli studi di Roma “La Sapienza” in 1986. Conservator at ISCR; among other duties, she has ongoing responsibility for research, training, treatments and consultancy in the Leather Conservation Studio 105

ICOM-CC Leather and Related Materials WG Offenbach 2012

that she helped to establish in 1989. Member of the ICOM-CC working group ‘Leather and Related Materials’ since 1990, she coordinated it in the period 2008-2011. She is the author of several technical publications. Istituto Superiore per la Conservazione e il Restauro - Via di San Michele 23 - 00153 Roma, Italy - [email protected]

Francesco Petrucci is an architect, art and architecture historian. Degree in Architecture at Università degli studi di Roma “La Sapienza”, with a post-graduate diploma in conservation of historical buildings. Since 1998 Curator of the Palazzo Chigi in Ariccia. As head of the Technical Office in Ariccia Municipality between 1984 and 1998 he planned and directed the conservation of numerous 17th century building in the area. Editor of the magazine “Castelli Romani” and since 2002 Associate Professor of Art History at College of Human Science, Auburn University (Alabama, USA). His main research interests are sculpture, painting, architecture and decorative arts of 17th and 18th centuries. Palazzo Chigi - Piazza di Corte 14 - 00040 Ariccia (Roma), Italy - [email protected]

© International Council of Museums – Committee for Conservation (ICOM-CC) – Working Group on Leather and related Materials, 2013 © International Council of Museums – Committee for Conservation (ICOM -CC) – Working Group on Leather and Related Materials, 2012

106

ICOM-CC Leather and Related Materials WG Offenbach 2012

Skin Clothing from the North: Research, Documentation and Preventive Conservation

Anne Lisbeth Schmidt

Abstract ‘Skin Clothing from the North’ is part of The National Museum of Denmark’s interdisciplinary research initiative Northern Worlds 2009-2013. In 2010-2012, the project focused on a study of the unique circumpolar skin clothing collection acquired between c. 1850 and 1950, from Greenland, Canada, Alaska, Siberia and Scandinavia representing the various circumpolar peoples. The collection includes all types of garments made from various Arctic mammals, including gut-skins as well as bird and fish skins, and the skin materials appear with and without fur, tanned or cured, dyed or undyed, etc. A number of costumes have been documented (including high-resolution photodocumentation) with regard to: a) pattern, by means of a new non-destructive method, b) identification of animal species, using hair microscopy and DNA analysis, c) identification of the sewing method, partly by means of X-radiography, and d) 3D photos of the costume. One intention of the new documentation is to make research feasible, and one key advantage is that we can now minimize physical handling of the clothing.

Keywords Skin clothing, circumpolar peoples, indigenous, Arctic, north, Denmark, species, identification, DNA, hair, microscopy, pattern, sewing, 3D imaging, database, documentation, non-destructive analyses, minimized handling, research.

Introduction

In 2009 the National Museum of Denmark launched the interdisciplinary research initiative Northern Worlds 2009-2013, where seven of the museum’s ten research units – together with external partners from Denmark and abroad – have taken part in over twenty sub-projects. Within Northern Worlds, the project ‘Skin Clothing from the North’ is associated with other projects under the heading ‘Networks of the North: Communication, Exchange and Cultural Markers’. At the website the common features of these projects are explained as follows: “The Northern societies have always been part of a cultural and economic network in a global context. Together with elements such as the study of costumes and material culture, exchanges of or trading in exotica have always been important markers of ethnic or social affiliation”1. In collaboration with the National Museum’s staff, external partners from Danish universities, from the Danish School of Conservation and from museums in Greenland and Scandinavia - who in November 2009 (Schmidt and Petersen 2010) took part in a workshop - continued to co-operate extensively with the project over the years.

The Project

Skin Clothing from the North studies the museum’s unique historical collection of skin clothing from the Ethnographic Collection. The c. 2100 individual items of skin clothing2 were collected among circumpolar peoples, i.e. the Inuit, Aleut, Siberian and Sami peoples, in the period c. 1850–1950. Most of the clothing was collected or purchased before European influence changed and replaced the original costume traditions and the skin materials used by the various indigenous peoples.

The items of clothing were made from various animal skins and animal guts with a variety of designs and sewing techniques; the skin materials were cured or tanned depending on the origin of the clothing material and the people who had produced it. A substantial number of garments collected in the same regions and in the same time-span – especially from Greenland and North America – are unique examples of parts of full costumes for babies, for boys and girls, and for men and women. The Siberian and Sami collections include only a few complete costumes, some of which were made for shamans.

The purpose of the project is to demonstrate which contexts can be identified among the items of skin clothing in terms of the geographical distribution of skin materials, gender, manufacture, sewing and design, 107

ICOM-CC Leather and Related Materials WG Offenbach 2012

lifestyle, geographical alliances, interactions and trade (Schmidt 2010, p. 10), with inspiration from the theories of the Danish archaeologist and cultural geographer Aage Gudmund Hatt (1884-1960), who in 1914 published his thesis Arktiske Skinddragter i Eurasien og Amerika. En etnografisk Studie. In his research, Hatt worked with two Arctic costume complexes: A, the oldest complex, associated with people living along coasts and rivers (Inuit); and B, the youngest complex, associated with the inland peoples of Eurasia (Hatt 1914, p. 243).

Research Strategy

The following tools were used to identify the contexts in which the circumpolar peoples made and used skin clothing: 1. Description of item (garment, garment part) 2. Digital photography 3. Digital documentation of design 4. Animal species identification

To ensure that the research was conducted in a way that minimized the strain on the old, delicate skin clothing – i.e. which reduced handling of the item in general, since all physical handling could cause major damage to all sorts of items – a systematic work flow was developed and applied. The risks and drawbacks that had to be considered – subject to the general intention of conducting the investigations as irreproa- chably and fruitfully as possible – were as follows: - overall risk of tearing fragile skin - possible loss of loosely attached beads, fringes, hair etc. - danger of pest attack and the effects of repeated freezing treatments - possible exposure of staff to insecticides - more generally, the limited working time available for the project.

Condition and Treatment of the Items A survey of the preceding period of almost thirty years shows how often the circumpolar skin clothing collection had been handled, treated and transported:  In 1985-1986 the collection was cleaned of insecticides (Vingelsgaard and Schmidt 1986; Schmidt 2001). At the same time the state of preservation of all items was assessed. The results of these assessments were given in the author’s master’s dissertation on the preservation of Inuit skin items. Measurements of shrinkage temperature (Schmidt 1991) also indicated that exposure to light degraded Inuit skin artefacts.  In 1987-1988, in connection with the dismantling and renovation of the ethnographic exhibition, the state of preservation of all objects was assessed.  In 1990-1991, selected Inuit items of clothing were preserved, padded with flexible inert washed cotton tricot (jersey knit) and polyester batting material and mounted on wooden constructions for the new exhibition.  In 2004-2006, c. 50% of the items of circumpolar skin clothing were reassessed by conservators because they had been selected as objects of ‘Outstanding National Significance’ (ONS). The ONS project, which has been described in detail by the conservator Eva Salomonsen (2008), embraced a wider range of movable cultural artefacts in the collections of state museums. In the ONS project, about 88% of the circumpolar skin clothing was rated as being in exhibitable or stable condition (category A or B); c. 12% of the items needed conservation treatment (category C or D). The high rate of stability was a result of the new improved exhibition and storage facilities.  In 2010-2012, in the present project, all the items were documented as described in the following.  Over the years there were frequent pest inspections, and items were occasionally lent out to exhibitions elsewhere.

108

ICOM-CC Leather and Related Materials WG Offenbach 2012

Today, the majority of the items of skin clothing are stored flat, wrapped and padded with pH-neutral tissue paper in pH-neutral cardboard boxes, in cold storage (annual average temperature for the last five years 11°C, fluctuating between 9°C and 16°C; average RH 60%, fluctuating between 44% and 73%) adjacent to the museum’s Conservation facilities in Brede north of Copenhagen. Since 1992 c. 300 items have been exhibited in the restored Ethnographic Collections at the National Museum in Copenhagen (annual average temperature for the last ten years 22°C, fluctuating between 19°C and 28°C; average RH 47%, fluctuating between 40% and 60%). A considerable number of these items have been exhibited for over 80 years.

Work Flow in the Project A team of experienced conservators, frequently accompanied by students, conducted all handling of items: dismantling in the exhibition, packing for transport and freezing treatment, mounting for photography, delivery to and returns from documentation studies, and in general removal from and return to storage areas.

The handling of historical organic objects necessarily requires the use of working clothes and gloves, because of potential residue of insecticides. Strict personal hygiene is also necessary as well as high standards in the vacuum-cleaning of horizontal surfaces, followed by regular washing in the relevant areas. The museum’s guidelines for Integrated Pest Management (IPM) state that items of skin material must be frozen (at -30°C for 72 hours) each time they enter or re-enter the storage or exhibition areas. All storage items were frozen at least once before re-entering storage, while items from the exhibition areas were frozen twice, on reception in the Conservation Department and before re-entering the exhibition areas. During the investigations a few cardboard boxes from the stores were found to have detritus from moths and other pests; these items were immediately isolated from the rest and frozen. Before re-entering stores all cardboard boxes were tagged, indicating that the items inside were involved in the project.

In the following, the documentation methods of the project are described.3

1. Documentation of Items For the documentation of items of skin clothing, the database ‘SKINBASE’ was developed in collaboration with the museum’s IT Unit. SKINBASE is sourced by selecting information from the central database of objects ‘GENREG’, along with new high-definition digital photos from the museum’s image database using the commercial database software ‘Cumulus’, along with the general new information generated by the project itself. The original development of GENREG, which was established in 1987, has been described by the archaeologist and IT specialist Lene Rold (1995). Glossaries in ‘GENREG’ were based on (among other sources) the international index Outlines of World Cultures; free text, for example the names of items, was allowed in some fields. The SKINBASE itself was constructed on the basis of a simple hierarchy of mutually affiliated glossaries, and was the main instrument for standardizing the documentation of the skin clothing in terms of the essential identification and naming of items of clothing, identical ways of stating measurements for each item, and the specific identification of the elements of each item of clothing, as well as skin material identification for each element with regard to animal species etc. (see Figures 1 and 2).

Glossaries of sewing techniques, decorations and closing devices were also implemented in the database. Only a few fields in SKINBASE have free text, that is, text prepared for publications/catalogues or exhibition illustrations.

SKINBASE has parallel glossaries of English, Danish and Greenlandic names of items of skin clothing (garments), names of elements of the items of clothing (garment parts), measurements etc., produced after verification in the relevant literature, to which reference is made in the glossary. Guidelines for measurements in each clothing group were given with illustrations in a manual.

109

ICOM-CC Leather and Related Materials WG Offenbach 2012

Figure 1: Initial screen capture of the database SKINBASE.

Parka

1 Back length 2 Chest width 3 Crown length 4 Front length 5 Hood height 6 Hood opening circumference 7 Lower circumference 8 Shoulder height 9 Side height 10 Sleeve length 11 Wrist circumference 12 Wrist-to-wrist length

Figure 2: Guidelines for measurement of parka.

2. Digital Photography In recent years, the photo studio at the Conservation Department has developed an advanced digital recording technique using a Hasselblad camera with a 39 megapixel digital back (Schmidt and Fortuna 2010). The photographic method produces both ‘flat’ photos and 360° photos. The technique was initially used in collaboration with Copenhagen University’s ‘Centre for Textile Research’ to study Iron Age skins and textiles from Denmark (Mannering and Gleba, forthcoming).

In ‘Skin Clothing from the North’, the photographic process was redefined to facilitate systematic naming and storage of files and the establishment of standards for the practical presentation and recording of the items. 110

ICOM-CC Leather and Related Materials WG Offenbach 2012

The naming of photo files as well as other files related to an item, and the storage of files, followed rules established by the IT Unit; the distinguishing name (in this case a number) of the item was generated from the central database GENREG.

Because of the high resolution of the photos, the recordings were limited to one photo of the front and one of the back of the item. To prevent shadows, all ‘flat’ items with outer measurements not exceeding c. 150 x 100 cm were photographed on a glass table, and the item was illuminated from below and from above using studio electronic flash lights. Larger items were photographed on the floor; headgear was photographed mounted on a stand. A meter with standard colours was used and each registration was furnished with the inventory name of the item (see Figure 3). Images are suitable for enlargement and publication.

Figure 3 (on the left): Front of woman’s parka from East Greenland, inventory number L.5064. Photo Roberto Fortuna Figure 4 (on the right): Woman’s full costume from East Greenland. Photo Roberto Fortuna

Together with the ‘flat’ photos of all items, a series of 3D photos of ensembles of related items that made up a full costume was produced. The costume was mounted (see Figure 4) on a specially designed mannequin of stainless steel. A total of 72 costumes were digitally photographed in 360° in a series of 24 registrations (see Figure 5). ‘Flat’ and 3D photos were both stored in the database program Cumulus. In this program an automated zoom-and-move function makes it possible to get close to the surface so that even minute details can be studied. A standard guide to the photographic method was published and the photo registration method is now being implemented in new work processes at the museum.

Figure 5: Preparing to take 360° photos. Photo Roberto Fortuna 3. Digital Documentation of Design A new non-destructive documentation method made it possible to convert measurements of three- dimensional items of skin clothing into accurate two-dimensional patterns including with area measurements. The method, which was developed at the University of Aalborg by the associate professor Karsten Jensen (2010), was used on 100 items of skin clothing chosen from among the selected representative circumpolar costumes for men and women (which were also photographed in 3D). This method, which is described in (Jensen et al. 2012), is non-destructive, and more accurate 111

ICOM-CC Leather and Related Materials WG Offenbach 2012

and fast than older conventional methods, and involves the use of a flexible measuring arm 4 that does not touch the material during the measurement. Optimum measurements were obtained when the skin material tolerated gentle straightening and support that made the seams available for three - dimensional mapping of the contours of the garment elements (e.g. for a parka: front, back, sleeve and hood). Inflexible, buckled materials like the soles of footwear were impossible to measure, but shafts and uppers could easily be measured.

The result of the three-dimensional measurements is reliable two-dimensional patterns (see Figure 6), which, given the links in the field recording sheet, documents the individual d esign of the item of clothing. The field recording sheet describes in detail the individual joining of pattern elements, with their shared seam types, stating the number of stitches per 10 cm. Seam types and thread were normally identified by the naked eye (Petersen 2010) or by magnification of the digital photos; in the case of footwear and other items of clothing with hidden seams, X-ray images were used (Gottlieb 2010). All information on sewing was concurrently registered in the database.

Figure 6: Pattern for a woman’s parka from East Greenland, inventory number L.5064.

4. Animal Species Identification In collaboration with the Centre for Textile Research and the Natural History Museum, animal species used for skin clothing have recently been identified through comparisons with prehistoric European domestic animals using hair microscopy, genetic profiling of skin and mass spectrometry-based protein sequencing of sub-samples of hair and skin. The study showed discrepancies in the identification methods, and in some cases the hair microscopy of prehistoric domestic animals was problematic, probably because hair morphology has changed substantially during domestication (Schmidt et al. 2011). In the present project, Arctic wild and domestic animal species have been identified macroscopically by the naked eye and by reference to the archives. For verification purposes, minute hair samples for microscopy and skin samples for genetic profiling were sampled from the above-mentioned items that made up representative costume ensembles for men and women from the circumpolar area. Sampling was documented by means of digital

112

ICOM-CC Leather and Related Materials WG Offenbach 2012

photos stored in the database. Samples were selected to permit identification of as many different species as possible.

The aim for the future with regard to the identification of Arctic animals is to compile a registry of the morphology of hair cross-sections supported by DNA analyses, and to describe a feasible method of preparing samples for light and electron microscopy.

Conclusion and Perspectives

In the project ‘Skin Clothing from the North’, systematic, non-destructive methods were developed in order to make research accessible for anthropological studies of the contexts for circumpolar skin clothing. The new documentation should provide future research with ‘hands-on’ information, since it is possible to examine all parts of the items at close quarters without access to the physical object by using high-definition photos. Detailed information on measurements, materials, design, sewing, species etc. was compiled for easy access along with information from the original inventories in the database SKINBASE. The great advantage of the project is that it minimizes further handling of the fragile skin clothing collection.

Great efforts are being made at the National Museum to share knowledge. The project paves the way for unlimited research with its future publication on the Internet where everyone is given access to unique treasures. The museum also wishes to offer the database to other museums so they can add their items of skin clothing. Documentation of archaeological items of skin clothing and textile is also a future option. The results of the research completed in the project will also be released in a forthcoming publication.

Acknowledgments Northern Worlds was made possible by a substantial contribution from the Augustinus Foundation. Hans Christian Gulløv, Mads C. Christensen, Steen Weidemann and Morten Ryhl Svendsen are thanked for their comments and revision of this paper. James Manley is thanked for the English revision of the text.

Endnotes 1. See: http://nordligeverdener.natmus.dk/en/home 2. Pairs of footwear, mittens etc. were counted as two items. 3. At the website, films of the processes can be seen. 4. A FaroArm was used for the 3D measurements

References Gottlieb, B., 2010. « X-rays of organic material », Skin Clothing from the North, Abstracts from the seminar at the National Museum of Denmark, November 26-27 2009 (eds. A.L. Schmidt and K.B. Petersen), p. 15. Hatt, A.G., 1914, Arktiske Skinddragter i Eurasien og Amerika. En etnografisk Studie. Copenhagen, J.H. Schultz Forlagsboghandel, 253 p. Jensen, K., 2010. « Automated construction of sewing patterns », Skin Clothing from the North, Abstracts from the seminar at the National Museum of Denmark, November 26-27 2009, (eds. A.L. Schmidt and K.B. Petersen), p. 16. Schmidt, A.L. and Petersen, A.H., 2012. « Analysis of traditional historic clothing. Automated production of a two-dimensional pattern », Archaeometry, published online 22 OCT 2012 Mannering, U., and Gleba, M., forthcoming, Designed for Life and Death, Copenhagen: The National Museum of Denmark. Northern World’s website, 2012. « http://nordligeverdener.natmus.dk/en/home » Petersen. A.H., 2010. « Analysis of sewing techniques in Inuit skin clothing », Skin Clothing from the North, Abstracts from the seminar at the National Museum of Denmark, November 26-27 2009, (eds. A.L. Schmidt and K.B. Petersen), p. 17. Rold, L. 1995. « GENREG – A Simple and Flexible System for Object Registration at The National Museum of Denmark » Selected Papers from the Third International Conference on Hypermedia and Interactivity in Museums (ICHIM ‘95 / MCN ‘95), Volume 1, pp. 19-38. 113

ICOM-CC Leather and Related Materials WG Offenbach 2012

Salomonsen, E., 2008. « An Evaluation and Preservation Project at the National Museum of Denmark – management and presentation », Studies in Conservation, Supplement 1, pp. 200-204. Schmidt, A.L., 1991. Eskimoiske skinds bevaringstilstand, Konservatorskolen, Det Kongelige Danske Kunstakademi, unpublished dissertation. Schmidt, A.L., 2010. « Skin Clothing from the North, Skin Clothing from the North », Abstracts from the seminar at the National Museum of Denmark, November 26-27 2009, (eds. A.L. Schmidt and K.B. Petersen), 10-11. Schmidt, A.L. and Fortuna R., 2010. « Polardesign – skinddragter fra nord », Nationalmuseet marts 2010, www.natmus.dk, pp. 1-6. Schmidt, A.L. and Petersen, K.B., 2010. « Skin Clothing from the North », Abstracts from the seminar at the National Museum of Denmark, November 26-27 2009, 36 p. Schmidt, A. L., Gilbert, M. T. P., Cappellini, E. and Olsen, J. V. 2011. « Identification of animal species in skin clothing from museum collections », ICOM-CC 16th Triennial Conference Lisbon, Portugal, September 19-23, 2011. 5 p. Schmidt, O., 2001. « Insecticide Contamination at The National Museum of Denmark: A Case Study », Collection Forum, Society for the Preservation of Natural History Collections, Volume 16, Numbers 1–2, pp. 92-95. Vingelsgaard, V. and Schmidt, A.L., 1986. « Removal of insecticides from furs and skins: Registration of conservation condition », ICOM Symposium on Ethnographic and Waterlogged Leather. Central Research Laboratory for Objects of Art and Science, Amsterdam, pp. 51-60.

Biography Anne Lisbeth Schmidt is a conservator M.Sc., project coordinator. Since 1983, she is employed at the National Museum (Conservation of organic materials; pest management and exhibition; Research on Danish prehistoric skin collection and Inuit collection). National Museum of Denmark, Conservation, I.C. Modewegsvej, Brede, DK-2800 Kgs. Lyngby, Denmark [email protected]

114

ICOM-CC Leather and Related Materials WG Offenbach 2012

Some Reflections on the Past Treatment of Gilt Leather, and Recent Remedial Work Where This Has Failed

Theo Sturge

Abstract. The author has been working on gilt leather in the UK and the Netherlands which has been treated in the past. Much of this work has been carried out by Henk van Soest. Some of the work is satisfactory, but structural failures were found where adhesives and repair materials had failed. In addition, over oiling of the leather had led to irreversible darkening of the decorative surfaces. Cleaning with micro-fibre cloths and solvents are considered, along with the re-repair of failed areas. Mould, and possibly spew, were found on some of the leather.

Keywords. Gilt leather, old repairs, oil, fat, discolouration, spew, red rot.

Introduction

Gilt leather has received treatment over many years going back to the 19th Century. Some of these early treatments are now failing. Work was also carried out in the latter part of the 20th Century which is also giving rise to concern. Some of the failures simply relate to structural problems where adhesives and repair materials have given way, but the problems in other areas relate to inappropriate materials which are causing serious difficulties. Some of these can be corrected, but others have caused irreversible damage. The principal area of concern relates to the application of oils and fats to the leather. These have darkened the decorative layers by staining both the yellow varnish over the silver and the paint layers, and, even more catastrophically, appear to have destroyed the silver layer on leather in Schiedam, the Netherlands. As a result, even if a way could be found to remove the oil and fat and return the varnish and paint to its original state the golden colour will be lost forever. The following case studies are not intended to give a full conservation report on the leather treated, instead they are used to highlight the specific issues relating to the previous treatments.

A 19th Century Failure : A Screen from Charlecote Park, UK

At Charlecote Park, the National Trust, UK, there is a very large 18th Century six fold gilt leather screen, the folds are 3.0 x 0.55 m. The leather is hand punched with a repeating design. Treatment is complicated by the presence of Chinoiserie paper on the other side making handling difficult and restricting access to the back.

Condition The extensive splits in the leather had been repaired with leather and animal glue. The repair leather was typical of late 19th Century leather. It had severe red rot and had lost virtually all of its strength. As a result, the repairs had failed. In some areas the animal glue had been applied too hot and the leather had shrunk and distorted (Figure 1). The acid in the red rotted leather had not transferred to the original gilt leather which, apart from the previous tears and some distortion remained in excellent condition.

Figure 1: Charlecote Park screen. The red colour of the 19th Century leather with red rot can be seen in the gap. The edges of the leather have curled where animal glue has been applied too hot and the leather has shrunk.

115

ICOM-CC Leather and Related Materials WG Offenbach 2012

Treatment Because of the problems of working adjacent to the Chinoiserie paper the original plan was to take the leather off the frames, repair it, and put it back. This proved impossible because during the previous work much of the leather had been glued to the wooden frame, and it was impossible to get it off without serious damage.

The leather used for the old repairs was very easily removed as it had no strength, and the bond was very weak. The animal glue was more difficult as little moisture could be used. Instead it was removed mechanically with a cut riffler from Alec Tiranti Ltd. This did not remove all of the glue, but the remaining parts were scratched well into the surface and when the leather was gently manipulated it cracked and the leather relaxed so it could be repaired again.

The new repairs were carried out with archival calf from J. Hewit and Sons Ltd and a mixture of Lascaux acrylic resins containing 3 parts of the harder 498HV and one part of the softer 360HV added to give more flexibility.

20th Century Treatments

Treatments by Henk van Soest (Deceased) Henk van Soest was a pioneer in the conservation of leather wall hangings, and treated a large number of rooms around Europe during the latter part of the 20th Century. Much of the leather he treated is in a satisfactory stable condition, but, as often happens with new treatments, some of the results are less than ideal. Inevitably, this paper tends to dwell on the problem areas. Sadly, there is very little documentation for van Soest’s work so most of the treatments have to be inferred from what can be seen. Most of the treated leather has had new edges / strip linings added. These are made with a distinctive gold painted leather. This tends to give each section a noticeable outline and is often seen as a van Soest “trade mark”. In many cases the leather has been edged with a black Lycra textile. This is sewn to a fine polyester (?) net which is attached to the back of the leather with adhesive. The idea is that the Lycra is attached to frames etc. and it allows the leather to expand and contract with changes in humidity. The Lycra was still sound on all the examples worked on by the author, but it is understood that the early examples are now starting to give way as the Lycra perishes and it looses its elasticity. The repairs are often carried out with a combination of the golden coloured leather and the net. The leather is applied first and then the net is applied over the top and onto the surrounding original leather.

The leather was almost invariably treated with oils and fats applied from the back. The idea was to bring the oil and fat content back to an ideal theoretical figure. Unfortunately, over time this sometimes migrates through into the decorative layers to give a sticky surface and severe darkening. It was also treated with imidazole. This is a buffer added to correct the acidity of the leather. This may be coming out onto the surface as fine bright crystals. The application of both fats/oils and imidazole is outlined in Pieter Hallebeek’s (2007) paper.

Examples of Work

Dunster Castle Hangings, UK Dunster Castle, the National Trust, UK, has a very important series of wall hangings depicting scenes from the life of Cleopatra. The gilt leather is in large sheets and is punched and painted. The leather was treated in 1982. The back was lined with net and the edges had Lycra attached. The leather is sound and holding together well. Repairs to the roof meant that two of the leather panels had to be taken down, and when it was put back up in 2009 the opportunity was taken to replace the Lycra with Velcro. This was for two reasons. It was likely that the Lycra would, in time, perish and need replacing, and the Velcro would allow it to be taken down quickly in an emergency such as a fire.

116

ICOM-CC Leather and Related Materials WG Offenbach 2012

Dokkum Town Hall, the Netherlands There are two rooms of leather in the Town Hall in Dokkum. Both have been conserved by van Soest. One was successful, and the other had failed. They were treated in the 1980’s. Room one. This room has moulded leather in individual sheets which are sewn together around all sides. The leather has been strip lined with the gold coloured leather and then re-sewn. The leather is attached to the wall with Lycra. In 2011 the leather was in good condition and no work was needed beyond a light clean. Room two. The second room has leather in vertical strips with horizontal skived joints. It is likely that the vertical seams were once sewn. The leather has been strip lined and repaired with gold painted leather, and a layer of net has been applied over the new leather and onto the original, but not over the whole back. The edges are attached to Lycra which is stapled to light weight lift out timber boards. The lift out boards, set in a channel top and bottom, allow the leather to be easily taken off the walls.

The strip linings and repairs were failing. Both the leather and the net were pulling away leaving gaps big enough to put a hand behind the leather (Figures 2 and 3). The gold painted strip linings had been attached with the smooth painted surface against the back of the original. This did not give a very good surface for the adhesive to bond to, giving a relatively weak join. It was thought that the failure had been triggered by the room becoming excessively hot and dry for a relatively short period. The good aspect was that the original leather had not been damaged, instead the repairs had failed. However, the Lycra did not appear to have fulfilled its function of allowing variations in size with changes in relative humidity. There was some darkening to the leather from the oil and fat added, and the surface was slightly sticky.

Figure 2 (on the left): Dokkum Town Hall. Before treatment. The strip lining of gold painted leather is becoming detached.

Figure 3 (on the right): Dokkum Town Hall. Before treatment. The net to reinforce the strip lining is coming away. The black Lycra sewn to the edge of the netting can also be seen.

Treatment The boards were taken down, and the staples holding the Lycra were removed. Rather than re-conserve all the leather, those areas which had failed were re-treated. The surplus adhesive was removed with a riffler, and the painted surface on the strip lining to which adhesive was to be applied was roughened with glass paper to give a key for the adhesive. The leather of the strip lining was re-attached with Evacon-R, an ethylene vinyl acetate co-polymer dispersion based adhesive. It was chosen for its high tack and comparatively high strength. The high tack made it easy to position the strips accurately, and reduced the need for extended drying times. The net was also coming away. In some cases it was completely detached, whilst in others it was blistering away. The existing net had adhesive injected underneath in the blistered areas, and it was applied with a brush where it was detached. It was then rolled down onto the surface with an artists’ ink roller. The surplus was removed by rolling the surface with kitchen paper which was removed immediately. This gave a neat surface finish. The adhesive for this was a mixture of Lascaux acrylic resins 117

ICOM-CC Leather and Related Materials WG Offenbach 2012

containing 3 parts of the harder 498HV and one part of the softer 360HV. This was chosen as it was less tacky than the Evacon-R so it did not stick to the roller, it also allows for a slightly longer working time. It is also a little more flexible. Many of the failures had started at the corners. The net and leather on the two sides did not always meet here. Where there was considered to be a potential weakness, a patch of 34 gsm Reemay, a non-woven polyester textile, was applied over the surface and overlapping the net (Figure 4). The adhesive was the Lascaux mixture.

The leather was cleaned with micro-fibre cloths dampened with a water spray. These are good at absorbing fats and oils. Considerable amounts of fat/oil were removed. The cloths turned an oily colour and went stiff on drying. This improved the feel of the surface and it was no longer sticky. However, the improvement to the appearance was quite modest. The leather was reinstated on its boards by fixing the Lycra back into place with stainless steel staples. The staff of the Town Hall is endeavouring to ensure that the environmental conditions are better controlled in the future.

Schiedam Weeshuis, the Netherlands The Weeshuis (Orphanage) in Schiedam is owned by Vereniging Hendrick de Keyser and was treated in 1988. One room has gilt leather on the walls. The leather is skived jointed to form large sheets. These have a punched and painted decoration. The leather had been repaired in the usual way with gold painted leather. This is intended to look like gilt leather but it does not involve silver or a yellow varnish. This was used to repair the edges and splits, it had also been used to fill the gaps. The leather is mounted on the walls and door with a mixture of Lycra, and where Lycra was not viable, staples.

Condition Apart from some very minor damage, the repairs were sound. However, the leather had been very heavily dressed / oiled and large areas were very dark in colour and surface, particularly in the dark areas, was sticky (Figure 5). In addition, the room had been damp (now corrected) and there was a white material on the leather. Some of this was very obviously mould (Figure 6). In other areas there was a white material on the surface which was rather different. This had the appearance of spew, but this has not yet been confirmed by analysis. Spew is a white waxy material that comes from low grade neatsfoot oil. It has a similar appearance to mould. If a “cold test” is carried out on neatsfoot oil by putting a bottle of it in a fridge, low grade oil goes cloudy as this material crystallizes out. Spew is commonly found on leather items which have been treated with saddle soap. This contains a lot of neatsfoot oil. When the white deposit is found, on enquiry it is often found that saddle soap has been applied. It is totally unsuitable for static museum objects. The possible spew followed the lines of cracks in the surface of the leather suggesting that oil and fat, and perhaps spew, had migrated through along these weak lines of least resistance (Figure 7). It is also possible that it was a fine mould, rather than spew, growing on the excess fat. There were fine clear crystals embedded in the surface. It seems likely that these are imidazole.

Figure 5: Schiedam Weeshuis. Some areas have retained their golden colour, whilst others have been severely darkened by the oils and fat. 118

ICOM-CC Leather and Related Materials WG Offenbach 2012

Figure 6: Schiedam Weeshuis. Mould on a damp (now treated) wall.

Figure 7: Schiedam Weeshuis. Lines of white along cracks in the surface. It is suggested this is spew rather than mould.

Investigation There was no obvious reason for some areas to be dark and others still relatively bright. They were said to be all bright at the time the leather was treated. One possibility is that some areas of the leather naturally absorbed more of the oil and fat than others because of variation in the skin quality. However, it seems more likely that it was caused by uneven application. One section had distinct dark strips (Figure 8). It is suggested that the oil and fat was methodically applied in sections, but each section overlapped giving a double application in the dark areas. It appears that it was applied in sections down one side and then the other. This could explain the dark bands which run horizontally between the sections as they worked their way along the leather, and the vertical strip up the middle as they went up the other side.

Figure 8 (on the right): Schiedam Weeshuis. Bands of dark leather between sections of bright leather. Probably caused by applying the dressing in overlapping sections.

Various people have looked at the leather. Martine Posthuma de Boer (Universiteit van Amsterdam) found some layers of over varnish on the surface and a layer of wax. The rest of her results are awaited with interest. It is thought that the wax was applied by van Soest. The darkest areas had the stickiest surface. Assuming that the darkening is caused by the migration of the oil and fat through the leather it would seem that these have gone into the wax and softened it leaving it sticky.

Eloy Koldeweij (Ministry of Education, Culture and Science, Cultural Heritage Agency, The Netherlands) visited the site with Han Boersma, a free-lance picture restorer. It had been assumed that the darkening was caused simply by the discolouration of the varnish and paint. At Eloy’s suggestion, Han returned on a further 119

ICOM-CC Leather and Related Materials WG Offenbach 2012

occasion to carry out some investigation as Eloy wished to check the condition of the silver. A small strip of varnish was removed high on the wall in a section where the colour went from gold to degraded dark brown. The varnish was removed with acetone (Figure 9). Under the golden area there was sound, bright silver. In the dark area no silver could be found and the cleaning went through to the underlying dark coloured leather. This suggests that some aspect of the oil and fat present has led to the destruction of the silver layer. If this is really the case it suggests that attempts to degrease gilt leather to retrieve the colour may be futile. It may be that further deterioration can be halted, but the damage already done is, tragically, permanent. Figure 9: Schiedam Weeshuis. Cleaning test. The varnish has been removed. In the bright area there is still silver under the yellow varnish. In the dark area no trace of the silver remains and the leather is exposed. Treatment The brief was to improve the overall appearance, and to reduce the stickiness of the surface, particularly in the dark areas. In addition, the removal of any oil and fat from the surface was considered desirable.

As with Dokkum, the initial cleaning was with damp micro-fibre cloths. This removed surface soiling, the mould and the possible spew, but it did not take very much of the sticky surface away, there was much less oil on the cloths than in Dokkum. Two options for further cleaning were considered. One was to clean with white spirit, the other was to clean with a 50/50 mix of water and white spirit to which a little detergent, Synperonic A7, had been added to form an emulsion. Both of these worked, although the water mix was a little slower. It was decided to use white spirit on its own. The mixture containing detergent would inevitably leave a little detergent in the surface of the leather however well it was swabbed away, and this addition was considered undesirable.

This secondary cleaning went in three stages. First, white towelling, folded into pads, was used with the white spirit. Whilst the surface was still wet it was further cleaned with pads of white cotton sheeting. Even at this final stage considerable amounts of oil and fat, plus wax, were coming away (Figure 10). The cleaning was stopped when the stickiness was significantly reduced but whilst there was still a little wax on the surface. Once dry, the surface was given a final clean with a dry micro- fibre cloth. This polished the wax to give an even finish and reduced the last of the stickiness. Pads were used rather than cotton wool swabs because the area was very large and it was easier to get an even result.

Figure 10: Schiedam Weeshuis. Cloth used to remove oil, fat and wax from the surface. This was the third stage of cleaning and considerable amounts of soiling were still being removed. Discussion

From the limited number of sites where work has been carried out post van Soest it is clear that the quality of the long term end result is very variable. Although the methods used are very recognisably van Soest’s, some have given satisfactory results. It is not clear why some have been more successful than others. Without cleaning more test strips on other heavily oiled and discoloured leather we cannot be sure if the silver is always destroyed when the leather turns dark in colour. 120

ICOM-CC Leather and Related Materials WG Offenbach 2012

Conclusion

It is clear that the application of fats and oils to gilt leather in the past have not, in general, been beneficial. Future work on gilt leather must concentrate on the fact that it is a decorative surface which happens to be applied to leather. The decorative surface must be given priority over the state of the leather. It is very easy to treat the leather as the priority and to try and, for example, change it so that it has a correct theoretical oil and fat content. However, if this is going to lead to the destruction of the decorative surface it is essential that the emphasis be changed. The decorative surface in Schiedam has been destroyed and there is no possibility of retrieving it. The surface of untreated gilt leather is often in good condition and it is only the application of inappropriate treatments which damages it. In addition to the damage done by oils and fats, the application of varnishes in the 19th Century is causing problems. These have now discoloured and are disfiguring the surfaces and they are very difficult to remove.

Acknowledgments The author is very grateful for the permission of the National Trust, Dokkum Gemeente Dongeradeel, and Vereniging Hendrick de Keyser, Society for the preservation of historically and architecturally important houses in the Netherlands, www.hendrickdekeyser.nl for permission to publish the case studies. The staff of all three have given considerable assistance. Discussions with Martine Posthuma de Boer and Han Boersma in Schiedam were very useful. Eloy Koldeweij, as always, provided helpful advice.

Materials - Archival leather. Vegetable tanned leather with a secondary 2% aluminium tannage. J. Hewit & Sons Ltd., 12 Nettlehill Road, Houstoun Industrial Estate, Livingston, West Lothian, EH54 5DL, Scotland, UK. Tel: +44 (0)1506 444160 Fax: +44 (0)1506 437749 www.hewit.com - Cut riffler. Address: 3 Pipers Court, Berkshire Drive, Thatcham, Berkshire, RG19 4ER, UK Tel: +44 (0)845 123 2100 Fax: +44 (0)845 123 2101 www.tiranti.co.uk - Evacon-R. Water reversible ethylene vinyl acetate dispersion based adhesive. Conservation By Design, Timecare Works, 5 Singer Way, Kempston, Bedford, MK42 7AW, UK. Tel: +44(0)1234 846300 Fax: +44(0)1234 852334 www.conservation-by-design.co.uk - Lascaux acrylic dispersions 498HV and 360HV. Lascaux Colours & Restauro, Barbara Diethelm AG, Zürichstrasse 42, CH-8306 Brüttisellen, Switzerland. Tel. +41 44 807 41 41, Fax +41 44 807 41 40 www.lascaux.ch - Micro-fibre cleaning cloths. Available from good hardware shops. - Reemay. Non-woven polyester textile. Conservations Resources UK Ltd, 15 Blacklands Way, Abingdon-on-Thames, Oxon, OX14 1DY, UK. Tel: +44 (0)1235553166 Fax: +44 (0)1235 534865. www.conservation-resources.co.uk - Synperonic A7 detergent. Alcohol ethoxylate. Conservation Resources, see above.

References Hallebeek Pieter B. 2007. ‘Analytical assessment of the condition of gilt leather wall hangings and consequences for treatment / Estimation par l‘analyse de l‘état du cuir doré de tentures murales et les consequences pour le traitment.’ Interim Meeting Working Group Leather and Related Materials, ICOM-CC, Brussels, Belgium, 1998. Sturge/Lulu. pp. 16-20.

Biography Theo Sturge trained in archaeological conservation at the Institute of Archaeology, London, in the early 1970’s. In his third year he specialised in waterlogged archaeological leather. On leaving university he worked in museums in Leicester and Coventry. Prior to setting up his own business in 2000, specialising in leather conservation, he worked at the Leather Conservation Centre for 6 years where he was senior conservator. He has published widely on leather conservation. Sturge Conservation Studio - 6 Woodland Avenue – Northampton NN3 2BY, UK - www.leatherconservation.co.uk - [email protected]

121

ICOM-CC Leather and Related Materials WG Offenbach 2012

The Role of Leather Science and Technology in Heritage Conservation

Roy Thomson

Abstract This paper aims to discuss the major concepts of leather science and technology which need to be understood if conservators are to achieve the essential holistic view of the skin-based substrates they are required to treat. It will give examples where this synthesis of disciplines has been achieved. The paper also considers why there is a lack of knowledge among so many non-specialist conservators about these materials and little appreciation of their widely different properties.

Keywords Leather science, materials science, historical technology, deterioration mechanisms, conservation, education

Introduction

This paper, based on my Doctoral thesis (Thomson, 2011), has been written from the unusual point of view of someone who is an experienced Leather Chemist, tanner and Accredited Conservator. Over recent years, I have been asked to advise on an eighteenth century painted parchment fan, destroyed by treating it as if it were paper; puff ball fungi excavated from a Romano-British site, conserved and exhibited as leather purses; early medieval felt insoles, mistaken for buff leather and a seventeenth century gilt leather altar frontal, ruined by the application of a commercial leather ‘cleaning and feeding’ product. These and other similar incidents led me to question how and why these mistakes occurred, and whether an understanding of basic concepts of leather science could have prevented these. These include:  Materials Science (The nature and properties of the material from which the object was made.)  Historical Technology (How the material was manufactured.)  Deterioration Mechanisms (The causes and mechanisms of its decay.)  Conservation Treatment Methods (How this deterioration can be mitigated.)

It must be appreciated that these categories are not discrete but interact fundamentally with each other. For instance, appropriate conservation treatment methods cannot be devised unless there is a knowledge of how the deterioration has occurred, which itself is based on an understanding of the Materials Science of the substrate.

Materials Science

Those without any exposure to leather science are unlikely to be aware that animal skin can be transformed into a variety of disparate products such as rawhide, parchment, pseudo-leathers, alum tawed skins and the many and varied types of leather prepared using a wide range of tanning materials. They fail to recognise that these products have very different characteristics and can react differently towards agents of decay and conservation procedures. Misidentification can therefore lead to inappropriate and damaging treatment. Thus it is essential that conservators become aware of these facts, together with the scientific and technological factors leading to these variations. In particular, they should have an understanding of how a skin is converted into leather and what the tanning process actually involves (Thomson, 2006a).

Leather has been defined as a material made from the outer skin of any vertebrate animal by any process or series of processes which render it non-putrescible under warm moist conditions. A true leather should retain this fundamental property after repeated wetting and drying.

This definition, however, only describes what leather is and not how this biochemical stability is achieved. To appreciate this, it is necessary for the conservator to have an understanding of both the structure of the skin and the chemical basis of the tanning process.

122

ICOM-CC Leather and Related Materials WG Offenbach 2012

From a leather science point of view, the most important component of all vertebrate skins is collagen. If a conservator is to treat a skin-based product safely, an understanding of the chemical make up of this protein is necessary, together with an appreciation of how molecules of this material build up into the intricate, three- dimensional, interwoven collagen fibre structure that gives leather and the other skin products their unique physical properties. There should also be an appreciation that it is the differences in detail between the structures of skins from different species which impart their specific physical characteristics. Leathers made from sheepskins, for instance, are generally softer and more likely to delaminate that those made from goat skins.

Conservators need to have information about the basic mechanisms of the tanning process and an understanding that these all involve the formation of biochemically stable chemical crosslinks between adjacent polypeptide chains. Without the introduction of these crosslinks into the protein structure, as in the case of parchment, or if they are unstable, as in the case of alum tawed pelts, the products are not leather and require their own treatments.

Above all, it should be emphasised that the term ‘leather’, like ‘wood’ or ‘metal’, covers not one but a range of related materials with many characteristics in common but each varying in its properties, modes of deterioration and reaction to conservation treatments. If this is not fully appreciated, incorrect treatment decisions will be made, resulting in damage or a complete loss of skin-based artefacts.

Historical Technology

The leather producing and leather working industries formed a vital part of European economies from the medieval period until the middle of the last century (Thomson, 1981). Despite this, relatively little has been published on the industrial history of the leather industries. From the point of view of the conservator, it can be argued that, without a basic knowledge of how leathers were manufactured in the past, the effects of alterations in tanning technologies cannot be properly understood and taken into account, resulting in inappropriate treatment.

Vegetable Tanning The great majority of the leather used to make the artefacts requiring conservation today will have been vegetable tanned and a significant percentage of these will be suffering from ‘red rot’. Investigations into this form of deterioration, resulting from the effects of strong acids on leathers produced using vegetable tannins, have concentrated on bookbinding leathers but it should be noted that red rot is found extensively in a wide range of other leather objects (Thomson, 2001). Many bookbinders and book conservators working today attribute the rapid deterioration of late nineteenth century leathers to the speeding up of the tanning process. A detailed evaluation of the historical technology over this period, however, shows this not to be the case. The changes in manufacturing techniques were not an attempt to cheapen the process but a response by the tanners to the demand for ever-increasing quantities of uniform, brightly coloured, thin leathers. These innovations included the introduction of synthetic dyestuffs, the importation of crust leathers and the development of shaving machines. All of these led to the necessity for the addition of acids to the processing liquors with the unforeseen consequences we are dealing with today. It is with an awareness of these historic technological changes that methods to counter the effects of red rot have been developed over the last century.

Alum Tawing Alum tawing was widespread throughout Europe from Classical Antiquity until it was superseded by the chrome tanning process at the beginning of the twentieth century (Thomson, 2009). The soft, white, leather- like product is not resistant to water so cannot be considered a true leather. If it remains dry, however, it is remarkably stable.

Since the middle of the last century, it has been thought that this process developed in the Middle East in the second millennium BC or even earlier. This hypothesis, based on translations of recipes for the preparation 123

ICOM-CC Leather and Related Materials WG Offenbach 2012

of skins inscribed on Sumerian tablets, has been widely disseminated by a number of authors, including myself. This theory is now thought to be incorrect and the recipes are considered to be descriptions of dyeing, incorporating alum as a mordant, not alum tawing (Driel-Murray, 2000, 2002). The earliest direct evidence for the use of alum for tawing is now believed to come from a second century AD whittawyer’s workshop from Roman Gaul (Bogard et alli, 2002). This somewhat esoteric historical information could be significant when the treatment of an archaeological skin-based artefact from, for example, early Egypt, is being considered. Similarly, evidence for the presence of aluminium salts in the surface of archaeological leather objects cannot be taken to indicate that it was tawed throughout its thickness. They could well have arisen from the application of a mordant on the surface, with the skin itself having been pseudo tanned.

Chrome Tanning Chrome tanned leathers are significantly more stable than those prepared by other methods. Nevertheless, artefacts made from chrome tanned leathers are increasingly likely to require preservation treatment. Conservators, therefore, need to be informed about the properties of this material and how these differ from those of vegetable tanned leathers.

Despite published evidence to the contrary (Thomson, 1985), many people, both leather chemists and conservators, still consider the chrome tanning process to date back only to the patents of Augustus Schultz of 1884. The leather from any artefact dating from earlier than this is, therefore, assumed to be vegetable tanned. This could prove to be a dangerous assumption. As an example, if an object known to date from 1880 is being examined prior to conservation and is assumed to be vegetable tanned, it may have its pH measured. If it had, in fact, been processed using one of the early, pre-Schultz, chrome tanning methods, it would probably have a pH of 2.8-3.2. A vegetable tanned leather with this value could be diagnosed as suffering from red rot. Treatments for red rot have included using buffer salts, particularly potassium lactate. Leather chemists are aware that salts of organic acids, such as lactates, react readily with chrome leather, breaking the tanning bonds between the chromium complexes and the collagen. This treatment would, therefore, damage chrome tanned leathers. It is precisely such knowledge of leather technology which has such a vital role in Heritage Conservation.

Deterioration Mechanisms and Conservation Treatments

As with objects produced from other materials, particularly organic materials, artefacts made from leather and other skin-based products are subject to agents of decay (Thomson, 2006b). These can be categorised as biological, physical or chemical and information from the field of leather science can be applied both to gain an insight into how the deterioration occurs and to aid the development of treatments to mitigate its effects.

Biological Deterioration Tanners and leather chemists are only too aware that leather and other skin products are subject to biological attack. While methods developed within the leather producing industries for preventing and treating such attack are not necessarily directly transferable for use with Heritage collections, it has proved useful for conservators to give consideration to some of the latest techniques employed. For instance, where it is not possible to prevent mould growth with environmental control, modern fungicides, which have been approved for industrial use by health and safety authorities worldwide, have been applied successfully.

The application of basic leather scientific knowledge also enabled a novel method for the eradication of insect pests to be employed on artefacts made from skin-based products. In the early 1990s, a method was developed by Thermo Lignum GmbH for the eradication of insect pests in wooden artefacts, using elevated temperatures. It was then suggested that this procedure could be applied to a wider range of Cultural Heritage objects, including those made from leather. The response of most conservators dealing with leather was negative. They feared that the increases in temperature experienced during this process were likely to cause damage, particularly if they were in a deteriorated condition. A project was undertaken by the author to investigate the potential hazards (Thomson, 1995). The results showed that, even with samples 124

ICOM-CC Leather and Related Materials WG Offenbach 2012

whose appearance and shrinkage temperatures indicated significant deterioration, the process had no deleterious effects. Although these findings were unambiguous, many conservators remained unconvinced. They did not appreciate that shrinkage is not just a factor of increased temperatures but is a result of hydrothermal activity, requiring the presence of water as well as heat. Leather scientists know that the temperatures at which skin-based products shrink are critically dependent on their moisture content and that, under the conditions prevailing during the Thermo Lignum process, even deteriorated leathers should be unaffected.

The author has employed this procedure to eradicate insect pests in objects where other treatment methods would have been problematic, including a bulky, post-medieval saddle with a complex structure made up of different types of leather, textiles, metals, wood and miscellaneous organic stuffing materials.

Physical Deterioration The fact that skin-based products undergo considerable expansion and contraction as the humidity of the surrounding atmosphere changes will be known by leather scientists. On the other hand, many non- specialist conservators find it hard to believe that leather can decrease in area by more than 10% if the relative humidity is reduced from 80% to 20%. This could occur if an object is transferred from an open, unheated store on a rainy winter’s day into a centrally heated display area, with disastrous results. It is just such basic information, readily available in the scientific and technological literature associated with the manufacture and properties of leather that one would wish to be more widely disseminated.

Leather scientists have also shown that it is the repeated absorption and desorption of moisture from and to the atmosphere which is primarily responsible for the type of deterioration, known as age hardening. This type of damage is observed mainly with thicker leathers found, for example, in military, transport and rural life collections emphasising the need for environmental control in these contexts.

Chemical Deterioration The deleterious effects of acidic pollutants on a range of skin products have been studied by chemists allied to the leather industries for over one hundred and fifty years. More recently, such investigations have been taken forward by conservation scientists. This important, fundamental work has, in the main, concentrated on determining the mechanism of this chemical deterioration (Larsen, 1994, 1996, 2002). The EU funded CRAFT Leather Project, on the other hand, had the practical, technological aim to develop a range of bookbinding leathers which combined the chemical properties which resisted acidic polluted atmospheres with the stringent physical characteristics demanded by the bookbinder (Thomson, 2003). By combining the knowledge and experience of bookbinders, tanners and leather chemists, a major breakthrough was achieved. One of the tanner partners, applying information gained from the scientific studies, developed a new range of semi-alum leathers with all the desired characteristics. These were being produced and sold on a commercial scale before the Project was fully completed. These leathers continue to be available.

Conservation Education

The first half of the twentieth century saw the development of the modern conservation profession. It was during this period that Francis Rawlings, Scientific Advisor to London’s National Gallery, wrote to George Stout, Head of Conservation at the Fogg Art Gallery at Harvard, about the training of professional conservators. He stated that: “Conservators should not only be good practitioners but scholars as well, knowing not only what they do but why they do it and prepared to discuss fundamental questions with their opposite numbers [i.e. curators] in aesthetics, art history and so forth.” He went on to suggest that: “Embedded in this [training] matrix should be an intensive ad hoc course in physics and chemistry – brief and circumscribed, but entirely scientific and objective in nature, given by scientists sensitive to art of course” (Boothroyd Brooks, 2000).

125

ICOM-CC Leather and Related Materials WG Offenbach 2012

It is not necessary for a conservator to become a fully trained scientist so a distinction must be made between which parts of the sciences are indispensable for the conservator, which parts are useful and which superfluous. It is also necessary to distinguish between which scientific knowledge is required by all conservators and which is necessary or useful for those who wish to specialise in a particular discipline.

The majority of conservators are Arts specialists with many having studied such subjects as Fine Arts, Art History or Archaeology to a first degree level. When enrolling on specialist conservation courses, their training in the sciences may be limited. There is no doubt that attempts are made to overcome this lack of scientific background by the inclusion of Conservation Science modules in all graduate and post-graduate conservation courses. Despite this, the writer has experienced what he has termed ‘chemophobia’ among a significant minority of conservators. Nevertheless, leather making is fundamentally a chemical process, as is its deterioration, so an appropriate knowledge of chemistry is a prerequisite for the understanding of the material. It is in this context that an understanding of leather science must be achieved.

With many materials, such as metals, ceramics, textiles and paper, this understanding is provided by specific education programmes. Despite their overwhelming presence in Heritage collections, this does not appear to be the case with leather and other skin-based products. In Britain, at least, formal teaching about these products appears to be restricted to one day at most throughout the whole course with some further, limited, practical exercises. Judging by my international correspondence, the problem is widespread.

Taking all these factors into consideration, it is not surprising that there is a distinct lack of knowledge among non-specialist conservators of these different materials and little appreciation of their widely differing properties.

This problem is exacerbated by the lack of literature relating the science and technology of leather to its conservation. The majority publications which exist are concerned mainly with descriptions of practical conservation projects, without addressing this fundamental aspect. The problem is compounded by the fact that what is written is often characterised by a lack precision in the nomenclature employed. Such inaccuracies would not be tolerated when dealing with other specialities, such as textile or ceramic production. This can be attributed to the authors of these publications having only a superficial knowledge of the subject. They sometimes give an impression of having read what the different processing stages consist of without understanding why they are carried out or how they are affecting the properties of the final product. In some cases, however, the misuse of widely accepted technical terms can only be described as dismissive.

All of these points reinforce a conviction that there is a general lack of knowledge about the subjects considered in this paper and, in particular, the fundamental nature of skin- based materials. There is a danger that this could lead to a lack of confidence among conservators, or worse, uninformed overconfidence, with resulting damage to artefacts. Further, this can lead to an unwillingness to move beyond familiar traditional practices which in some cases, such as the ‘feeding’ of leather, are positively harmful. An improvement in this situation can only come about if there is closer collaboration between leather scientists and technologists with conservators, archaeologists and curators. In this way, information such as that analysed in this paper could be disseminated. The author’s experience suggests that this can be mutually beneficial.

I realise that some of these comments may have given the impression that all Heritage professionals lack any knowledge about leather. This is not the case. There exist specialist groups and individuals whose aim is to raise the profile of all aspects of leather within the Heritage community. The Leatherworking Group of ICOM-CC is the leading example.

What is beyond dispute is the importance of the many different skin-based materials which are found in Heritage collections everywhere. We all need to emphasise that they are no less important than objects made from textiles and metals, for example, and merit equal status. 126

ICOM-CC Leather and Related Materials WG Offenbach 2012

References Bogard, P. et alii. 2002. « Passer les peaux en blanc »: une pratique Gallo-Romaine.” In: F. Adouin – Rouzeau and S. Beyries (eds.) Le Travail du Cuir de la Préhistoire à Nos Jours. Antibes: APDCA. pp. 231-250. Boothroyd Brooks, H. 2000. . A Short History of IIC. London: International Institute of Conservation, 2000. p.7. Driel-Murray, C. van 2000. “Leatherwork and skin products.” In: I. Shaw and P. Nicholson, (eds.) Ancient Egyptian Materials and Technology. Cambridge: Cambridge University Press, pp. 299-319. Driel-Murray, C. van 2002. “Ancient skin technology and the impact of Rome on tanning technology.” In: F. Adouin – Rouzeau and S. Beyries (eds.) Le Travail du Cuir de la Préhistoire à Nos Jours. Antibes: APDCA. pp. 251-265. Larsen, R. (ed.) 1994. STEP Leather Project Evaluation of the Correlation between Natural and Artificial aging of Vegetable Tanned Leather. Copenhagen: Royal Danish Academy of Fine Arts. Larsen, R., (ed.) 1997. ENVIRONMENT Leather Project, Deterioration and Conservation of Vegetable Tanned Leathers. Final Report. Copenhagen: Royal Danish Academy of Fine Arts. Larsen, R. (ed.) 2002. Microanalysis of Parchment. London: Archetype. Thomson, R. 1981. “Tanning - Man’s first manufacturing process?” Trans. Newcomen Society, 53. pp. 139-156. Thomson, R. 1985. “Chrome tanning in the nineteenth century. Atkin Memorial Lecture 1984.” J. Soc. Leather Technol. Chem., 69. pp. 93-98. Thomson, R. 1995. “The effect of the Thermo Lignum pest eradication treatment on leather and other skin products.” In: The Treatment of and Research into Leather. Postprints of ICOM-CC Leathercraft Group Interim Meeting. Amsterdam, pp. 67-76. Thomson, R. 2001. “Bookbinding leather: yesterday, today and perhaps tomorrow.” J. Soc. Leather Technol. Chem., 85. pp. 66-71. Thomson, R. 2003. “Towards a longer lasting leather: a summary of the CRAFT Leather Project.” The Bookbinder, 17. pp. 65-70. Thomson, R. 2006a. “The nature and properties of leather”. In: Marion Kite and Roy Thomson, (eds.) The Conservation of Leather and Related Materials. Oxford: Butterworth – Heinemann. pp. 1-3. Thomson, R. 2006b. “The deterioration of leather. Procter Memorial Lecture 2005.” J. Soc. Leather Technol. Chem., 90. pp. 137-145. Thomson, R. 2009. “Alum in the leather industry.” J. Soc. Leather Technol. Chem., 2009, 93. pp. 125-129. Thomson, R. 2011. The Role of Leather Science and Technology in Heritage Conservation. Unpublished PhD Thesis, University of Northampton.

Biography Roy Thomson’s background combines experience as a Leather Chemist, a practical tanner and a Conservator. He retired from the post of Chief Executive at the Leather Conservation Centre in 2004. He is an Accredited Conservator, Chartered Chemist, Fellow of the Royal Society of Chemistry, Fellow and Past President of the Society of Leather Technologists and Chemists and Fellow of the International Institute for Conservation. He has lectured and published widely both in the UK and throughout the world. His publication, The Conservation of Leather and Related Materials, co- edited with Marion Kite of the Victoria & Albert Museum, has been described as the definitive work on the subject. His most recent book is Leather tanneries: the archaeological evidence, co-edited with Quita Mould, is the first book to critically examine the remains of this historically vital industry. He has recently completed a PhD investigating the interaction between leather science and technology and Heritage conservation.

The Long House, behind 43 West Street, Oundle, Peterborough, PE8 4EJ, UK - [email protected]

127

ICOM-CC Leather and Related Materials WG Offenbach 2012

Lubrication of Ancient Leather – Imperative or Impossibility? Bernhard Trommer, Andreas Schulze, Heinrich Francke

Abstract The simplest definition of leather is a skin prepared in such a way that it dries without hardening and wets without rottening. Greasing and degreasing are technological important steps of leather manufacture to assure a number of properties like a appropriate softness, fullness, mellowness and flexibility of fibres, water and wear resistance, strength and ability to stretch without cracking etc according to its future purpose. Centuries ago, fatting consisted chiefly of smearing natural oils and greases over the surface of the wet leather. Natural and oxidized oils, saponificated oils and its fatty acids were common components of tannery auxiliaries of this period. The operation of fatliquoring, which originated in the nineteenth century, became the modern fatting technique until today. The first half of the twentieth century was the heyday of the auxiliary chemistry. The fatting technique was upgraded by new substances like sulphated and sulfonated oils and fatty acids, chlorosulfonated hydrocarbons from Fischer-Tropsch process and others which caused a vast number of products. Ancient leather as part of cultural heritage is following its own guidelines of treatment. Myths emerged around the possibility of re-fatting and impregnation causing common misconceptions about the usefulness for ancient and archaeological leather of periodical polishing and stuffing (or ‘feeding’) with special mixtures of lubricants. Painful experiences with lubrication of ancient leather have brought to awareness that these types of actions can ruin leather more than inaction.

Keywords Leather, archaeology, restoration, conservation, lubrication, fat, grease, liquor

Introduction

Native skin contains 60 to 70 % water which ensures its softness and flexibility. The humidity of leather is lowered at a level ranging from 8 to 18% under normal indoor climate. During the process of leather manufacture, unstructured and structured proteins, fat etc. are removed and replaced by different auxiliaries. The essence of leather fatting is leading to the definition of leather in general formulated by Reich (2003) (authors’ translation):  Tannage is a step of manufacture whereby collagen tissue of hides and skin is stabilized permanently as leather, that is, without a collapse or bonding of the fibres through dehydration. This kind of intermediary product does not swell when wetted and has become more stable with regards to hydrolysis, enzymes and temperature. All kinds of auxiliaries, regardless of their composition and type of bonding with collagen, are called tanning agents.

Application of Lubricants on Leather

Chemical and Physical Principles

The simplest definition of leather is a skin prepared in a way that it dries without hardening and wets without rottening. Fatting, in combination with tannage, guarantees flexibility of the fibres and fibre bundles of the dry leather matrix. Lubrication means moisture-independent fibre insulation. The aims of fatting can be summarised as follows (Trommer 2008):

- Tear resistance and flexibility under static and dynamic load; - Elasticity and abrasion resistance under dynamic load; - Waterproofness of leather in wet environment; - Component of finish formula (e.g. plasticiser); - Aesthetic improvements of leather (e.g. lustre, grip); - Protective film to slow down oxidative ageing.

The first two points are derived from fibre insulation. Tannage occurs inside the collagen fibrils; fatting at the surface of collagen fibrils. Both are basic mechanisms to produce leather. A model can be devised whereby,

128

ICOM-CC Leather and Related Materials WG Offenbach 2012

assuming a fixed leather volume with a defined density and degree of tannage, it is possible to calculate how much fibrillar surfaces are available for fatting. A calculation of the theoretical amount of a common fat (referred to a monolayer coating) becomes possible. According to a model by Reich and Oertel (1987), between 0.8 % and 5.1 % of a common anionic fatty alcohol sulphate (alkyl sulphate) is necessary to generate a mono-molecular layer at all fibrils of a chrome tanned cow hide. Both values consider border line cases of uniform steric alignment of the fat molecules. Additional quantities of fat cause additional effects (Raphael and McGrady, 1984) like waterproofness or lower the wash-out of vegetable tanning material. A very high fat content degrades the wash-out again. Impregnation of leather means filling of all interfibrous spaces with fat accompanied by high waterproofness, enhanced thermal conductivity and high wear resistance under permanent swelling or cyclic load e.g. for belting leather.

Lubrication of leather as a technological step of the leather manufacture can be classified into 4 techniques, as shown in Table 1.

Dry fatting process Hot stuffing (stuffing grease) Oil-in-water emulsion (O/W) Fatliquoring (e.g. anionic-charged liquors) Water-in-Oil emulsion (W/O) Table stuffing (Dégras, Moellon) Solvent-based liquoring / treatment Impregnation (hydrocarbon solvent)

Table 1: Basic processes of leather lubrication.

The two basic preconditions for uniform fatted leather fibres are: (1) a comprehensive absorption capacity of the leather matrix; and (2) an adequate conditioning of the fatting matter at the moment of dosage. The bonding forces vary widely. Leather fats can improve weakly tanned leather, or even substitute the mineral or the vegetable tanning matter.

Water is the native lubricant of fresh skin. The water-induced fibre insulation is replaced by fats as a result of leather manufacturing process. Fats are applied on wet material e.g. a leather that has a humidity of approximately ≥ 30 %. Mostly all auxiliaries are dosaged in the form of water-in-oil-emulsion or oil-in-water- emulsion. Following the application of the emulsions, there is diffusion, break of the emulsion, and the fixing of fat. Today's knowledges about steric and bonding effects of lubricants with leather are still incomplete and inconsistent (Gutterres Soares 2001).

Historical Evolution of Fatting

Moulding and dry hardening of a wet hide was an initial kind of manufacture in the earliest history of humanity. The search for a soft and preserved skin post mortem marked the actual beginnings of the leather craftsmanship. Fat tannage and oil dressing dominated leather manufacture for a very long period because of the availability of natural resources and the simplicity of these operations. Careful fleshing, soaking and an intensive mechanical action combined with an adequate fulling with fatty matter like fat, tallow, fish oil produced a soft material. Natural phospholipids became a highly effective emulsifier. Fat tannage was suited to a large number of applications. In the further course of history the preserving effect of volatile oil fission products and smoke has been recognised (Klokkernes 2007; Trommer 2008).

One of the earliest handcrafted leather auxiliary was a by-product of the chamois leather production. The so called Dégras or Moellon was a water-in-oil-emulsion with around 25 % water and a high part of different hydroxy fatty acids. Dégras was the most important fatty matter for vegetable tanned leather until the beginning of the 20th century. It was applied in mixture with other animal and vegetable oils, fats and waxes on wet leather. Table 2 shows historical techniques and auxiliaries.

129

ICOM-CC Leather and Related Materials WG Offenbach 2012

Technique Resoures Fat Content Type of Leather oil-off neatsfoot oil, marine oils, rarely linseed or ≤ 0.4 % sole leather mineral oils 1-3 % bottom leather fulling marine oils (train, sperm, cod liver, fish, seal oil) 0,5-3 % chamois leather nourishing paste (glacé) 14-18 % glace leather, glove kid table grease, mixtures of tallow, train oil, dégras/ moellon with 3-8 % lining leather hand stuffing small dosages of lanoline, horse-grease, soap 6-12 % sleek leather stuff, waxes, mineral oils 17-23 % upper leather, russet upper 10-25 % Zeugleder2, case hide ≤ 30 % Russian leather hot stuffing partly hydrogenated train oil, paraffin, stearin 10-27 % belt leather 10-25 % harness leather ≤ 40 % laces and whip leather

Table 2: Historical techniques and auxiliaries of leather fatting (Gnamm 1951; John 1996; Trommer 2008).

Chamois tannage, glacé tannage and hot stuffing of vegetable-tanned leather became special cases of lubrication. Theses leather types are niche products today. Glacé tanned leather guarantees maximum durability and is utilized by organ builders (Trommer 2006).

Modern Leather Fatting

Formerly used for hand stuffing, commercially today fats have almost completely come out of use in pure form. The industrial technology of lubrication is a treatment with an oil-in-water emulsion. The dispersed phase allows a wet drum processing, known as fatliquoring. The essential components of liquor are an appropriate mixture of oils and an emulsifying agent. Emulsifiers may be divided into 3 types: anionic, cationic and non-ionic. Anionic fatliquors are those most commonly used and include mostly the long- established sulphated and sulphonated oils, and more or less of the unchanged or partly-changed raw oils and additional oils from natural or synthetic sources, which act as the lubricating components.

Leather benefited most from the enormous progress of the oleo- and surfactant chemistry especially in Germany in the first half of the 20th century. Fischer-Tropsch process, catalytic high-pressure hydrogenation, sulfonation, sulfochlorination, etc, provided new classes of leather auxiliaries from non-food sources. Synthetic fat replacement materials (Austauschfette) became important by-products of synthetic fuel production (Kogasin) made by the hydrogenation plants in Central Germany from the year 1936 (Leithardt et al. 1951). The US military estimated highly the quality of German footwear leather despite the strained economic situation of their wartime enemy (Brown et al. 1951). Auxiliaries with short- and medium-chain chlorinated paraffins (SCCP, MCCP) have now disappeared from the market because these products do not meet current ecological and health requirements3.

The very last innovation of leather fatting and impregnation are anti-soiling finishing products for leather based on silicones such as polydimethylsiloxane (PDMS) and amino-silicones (aminoalkyl silicone fluids), and waxes such as fluorinated resins and fluoro surfactants (Hollstein 1987; van Hulten et. al. 2012).

The degree of penetration and lubrication are governed by dosage, temperature, humidity, pH, charge of leather and nature of fats and liquors. Admixtures of mineral oils for example have good lubricating properties and do not spew. They penetrate into the outer layers, and have no affinity for the leather and tend to migrate. Crude or sulphated neatsfoot oil has a low penetrating power, but it has good lubricating properties - better than most other oils. It is stable and does not spew if it is filtered from solid stearines. Sulphated castor oil (Turkey red oil) has also good lubricating characteristics without being greasy. It is light-

2 German type of leather for military purposes, similar to sleek leather 3 SVHC-substances (substances of very high concern) declared by REACH 130

ICOM-CC Leather and Related Materials WG Offenbach 2012

stable and therefore appropriate for pale-coloured leather. Marine animal oils are useful lubricants, but pure- grade raw materials are not available any more for the supplying industry today because of the precarious state of resources. Final distribution of the fat in leather is furthermore subject to histological, stratigraphic and topographical circumstances (Figures 1, 2).

Figure 2: Sheep leather with grease stain and fatty spew over Figure 1: Scheme of stratigraphic distribution the backbone region caused by high residues of native fat and total fat content of a leather cross section. content and storage under changing climate conditions.

Leather Fats for Restoration and Conservation

Ageing and Degradation of Lubricants

The processing of a collagen matrix into leather is always more or less accompanied by significant surrender of its initial state and its native stability. Therefore leather in heritage and archaeological conservation means dealing with an organic material which has a limited lifetime. The main leather substances are corium, tanning matter and fat. Fat is the most unstable component. The effect of changes in native fat distribution on the quality of leather quality has to be taken also into account (Figure 2). The fat content of raw hides ranges approximately between 3 % (cattle hide) and 30 % (sheep, pig) and cannot be removed completely by technical degreasing.

Apart from industrial modification of oils and fats, autoxidation or spontaneous decomposition, either inside or outside the leather, is always possible with different reaction time. Rancidity triggered by oxidative fat cleavage is already noticeable in aerial oxygen and causes formation of hydroxy fatty acids and subsequent aldehydes, ketones and carboxylic acids. The presence of substantial quantities of free fatty acids in leather and its negative consequences can have far-reaching effects. This applies in particular to the migration and crystallisation of saturated fatty acids which are some of the Figure 3: Grain side of vegetable-tanned calf leather (oak and spruce bark extract) with greasy spew after hydrolytic, oxidative ageing (fast most common cause for spews and deposits testing, SEM 250 x magnified). (Trommer 2008). (Hollstein 1979) (Figures 3, 4). 131

ICOM-CC Leather and Related Materials WG Offenbach 2012

Saponification of fatty acids of native and technical fats during manufacture may result in deposits leading, for example, to split leather soles (Tancous 1965).

Acid cleavage of fats may occur at a pH significantly below 3.5. The acidification can be triggered e.g. by absorption and adsorption of airborne pollutants. The degradation of organ leather at sites with high air pollution is well known (Trommer 2006). UV radiation also promotes degradation. Fungal exo-enzymes are able to cause lipolytic cleavage of leather fats. Dust deposits and non-purified auxiliaries are co-substrates for microbiological degradation.

Figure 4: Swelling and greasy deposits, gilt leather, sample Moritzburg Castle, Saxony (Schulze 2011).

Unsaturated fats have a distinctive tendency to autoxidation. Reaction products of the spontaneous process are monomeric and polymeric peroxy, epoxy and aldehyde compounds, alcohols and acids which can interact with other leather components (Elsinger 1982). Metal and metal ions e.g. iron or copper are activators for autoxidation. The released fatty acids may also take effect on metals due to corrosion. A number of tannins can inhibit the process by the way of complexation. Polyunsaturated fats (e.g. marine oils) are able to trigger free-radical chain reaction accompanied with undesirable side effects like “secondary tannage” and fast leather ageing (Trommer 2008). The process is accelerated by increasing pH and moisture of the leather (Elsinger 1982). The manufacture of chamois leather involves the selective technological application of this phenomenon.

Comprehensive Strategy for Preventing Lubricants Defects

Two questions should be clarified prior to starting a conservation treatment of ancient leather involving the addition of lubricants: 1) Is there any information about the original fat content and leather type? 2) What kind of added value and sustainable benefit are expected from adding a lubricant? Museums objects or preserved archaeological findings are not exposed to static, or even less so, dynamic loads, or to any wetting. A greasy film or wax coating can provide aesthetic improvements for a given period, but dust and grime can then stick to it and this combination of a fatty, dirty layer represents a perfect target for microbes afterwards.

The chief argument made in favor of relubrication is to provide a protective film to slow down the oxidative degradation of leather. However, on the other hand, lubrication means fibre insulation and enlargement of the specific surface area of leather, which increases the possibility of absorption and adsorption of atmospheric pollutants. An untanned material like parchment without additional lubrication has a significant smaller inner surface than leather (Figure 5). Nevertheless parchment shows a significant higher ageing resistance than, for example, numerous vegetable-tanned bookbinding leathers.

The link between lubrication and softness is not necessarily connected functionally, because of some types of leather are flexible and soft without a high fat content, e.g. chamois leather or saffian.

Waterlogged archaeological leather or the finish of gilt leather contain quantities of alkaline earth salts (Mg, Ca), metal, metal salts or oxides (Ag, Fe, Mn, Pb, Sn, Zn) which cannot be removed. Pigments and 132

ICOM-CC Leather and Related Materials WG Offenbach 2012

siccatives are “interesting” reactants for degradation processes. These substances offer potential for damaging consequences e.g. hydrophobia, depolymerisation and detanning of the leather, polymerisation (factise deposits) and saponification of the leather fats, swelling or delamination of the leather finish or coating (Figures 6, 7).

Figure 5: Specific surface of leather or processed skin materials. Left: opanke sole, parchment-like dried pig skin. Right: sumac tanned saffian leather (both 2,000 x magnified).

Figure 6: Cross section of Russian leather after catalytic induced oxidative ageing (fast testing): one can see burst and unimpaired fibre bundles and depolymerised (amorphic instead of fibrillar) collagen and fat residues (SEM 5,000 x magnified) (Trommer 2008).

Gilt leather is a prominent victim of fatting treatment or relubrication (Jäger s. d.; Trommer et al. 2010). An impregnation of the flesh side, e.g. with neatsfoot oil. does not improve the tensile strength. The higher resulting weight of the vertically hanging leather panels also results in a higher tensile load for the weak leather fibres. The impregnation of the dry leather is impossible to do in a way that would achieve a uniform uptake, distribution and fixing of the lubricants. Fats und surfactants migrate to the grain side of the gilt leather and get in contact with the parchment glue, silver leaf and painted layer. Staining, swelling and delamination of the gilt leather finish follows (Figure 7). Figure 7: Cracks and delaminated coating, gilt leather panel, sample Moritzburg Castle, Saxony (Schulze 2011). 133

ICOM-CC Leather and Related Materials WG Offenbach 2012

Tests carried out within the ENVIRONMENT project to study common protection and care products for leather conservation did not show any improvements regarding leather structure or tensile strength for a trial period of 10 years (Larsen 1997). Ancient leather objects which suffer from brittle grain are oiled-off periodically (Noehles 1984). The impregnation with fluid paraffin (for example) does not increase grain elasticity (Mariott 1933). The ageing process of collagen increases the ammonia content in leather. An additional treatment of vegetable-tanned leather with neutralising ammonia salts, or a fumigation with ammonia, is often recommended (Noehles 1984) but can promote other secondary effects and even lubrication defects. An increased pH value (which some treatments achieve intentionally or unintentionally) means risk of detannage, increased occurrence of factise (polymerisation) (Stather and Laufmann 1933), accelerated oxidative ageing, saponification or microbial growth.

Fluctuating indoor climate with low temperature and high humidity favours spew of glycerides. Deposits of free fatty acids generally occur at higher room temperatures and at a low relative humidity (Francke 2012). Specific melting and setting point of a number of substances are shown in Table 3. Low melting points favour the sticking and accumulation of dust and grime deposits on the leather surface. This can be exacerbated in the case of leathers exposed to changing indoor climatic conditions, or of changing microclimates close to the leather surface, especially if the temperature fluctuates around the melting or solidification (crystallising) point.

Substance Category Melting point (range) glycerin tristerate triglyceride (lard oil) 72 °C stearic acid cleavage product 70 °C glycerine disterate triglyceride (vegetable) 68 °C palmitate palmitic acid cleavage product 63 °C glycerine disterate cleavage product 60 °C glycerine monosterate cleavage product 55 °C beef tallow triglycerides (animal) 40...50 °C mutton tallow native triglycerides (animal) 38…48 °C wool fat (lanolin) cholesterol fatty acid esters / waxes (animal) 31…43 °C glycerine distearyl oleate triglyceride 42 °C lard, pig fat triglyceride (animal) 36…42 °C glycerine dipalmityl oleate triglyceride 35 °C egg yolk phospholipides (animal) 22…25 °C oleic acid cleavage product 13 °C glycerine trioleate triglyceride 6 °C linoleic acid cleavage product - 5 °C rapeseed oil triglyceride (vegetable) Ranges between -2 and -10 °C Table 3: Critical substances for spew and deposits on leather (Hollstein 1987)

Leather conservation can benefit from technological expertise of leather manufacture. Recommendations aimed at preventing the formation of free fatty acids and fatty deposits on leather are (Hollstein 1979; Mitton and Pankhurst 1956): - Prevention of harsh climate changes (large and/ or rapid RH fluctuations) and intense radiation - Prevention of crystallisation seeds e.g. dust which increases risk of crystalline deposits of free fatty acids ) - Provide fungistatic conditions and nutrient removal - Monitoring of the pH and the difference number (detection of the presence of free mineral acids) - Avoiding saponification conditions (e.g. presence of alkali or metal salts) - Short storage time for conservation auxiliaries

- Pollution control (SO2, NOx, O3)

134

ICOM-CC Leather and Related Materials WG Offenbach 2012

Conclusion

The relubrication of ancient leather has completely different impacts as compared with the original lubrication manufacturing process. The major difference is the lack of moisture in any relubrication attempt. Water used in the original manufacturing process of lubrication takes on the function of an interfibrous spacer and vehicle of transport which guarantees an outstanding penetration and efficient fixing. The lubrication of dry and aged leather causes a number of follow-up problems. This is one instance where the ‘less is more’ principle is particularly applicable (Figure 8).

Conservation studios specializing in leather restoration should not stockpile fatliquors or other fatty auxiliaries on the long term. Suppliers should be selected by objective criteria like transparency and proof of origin. Producer of lubricants should fulfill commitments with strict seriousness and professionalism which means quality management systems, traceability of resources, product purity and stable processes.

Figure 8: Natural ageing (left) compared to dry lubricated gilt leather panel (right), sample Moritzburg Castle. References Brown, J. B. et. al. 1951. "Synthetic fat replacement materials for leather (III)". JALCA 46, 9, pp. 483-505 Elsinger, F. J. 1982. "Die Tranautoxidation und ihre Bedeutung für die Leder- und Pelzherstellung". Das Leder 33, 8, pp. 125-131 Franke, H. 2012. "Fettausschlag – ein vermeidbarer Lederfehler?" lecture given on 1. Freiberger Ledertage, 25th-26th April 2012 in Freiberg Gnamm H. 1951. Die Fettstoffe des Gerbers. Stuttgart: Wissenschaftliche Verlagsgesellschaft m.b.H. Gutterres Soares, M. 2001. Über die Wechselwirkung von Sulfosuccinaten mit gegerbtem Kollagen – ein Beitrag zu den Grundlagen der Lederfettung. Dissertation, TU Bergakademie Freiberg Hollstein, M. 1979. "Zum Problem der Fettausschläge auf Leder". Das Leder 30, 3, pp. 40-46 Hollstein, M. 1987. Entfetten, Fetten und Hydrophobieren bei der Lederherstellung., Bibliothek des Leders vol. 4, Frankfurt/ M: Umschau Verlag Jäger, S. s. d. "The deterioration after treatment of painted leather wall hangings in three castles in Rhineland".s. l. unveröffentlichte Aufzeichnung des Landesamtes für Denkmalpflege Sachsen John, G. 1996. Fehlermöglichkeiten bei der Lederherstellung. Lampertheim: Druck Partner Rübelmann GmbH Klokkernes, T. 2007. Skin processing technology in Eurasian reindeer cultures. dissertation, University Oslo Larsen, R. 1997. "Zerfall und Konservierung vegetabilisch gegerbter Leder". Restauro 1997, 4, pp. 268 – 271 Leithardt, P. et. al. 1951. "Synthetic fat replacement materials for leather (IV) ". JALCA 46, 9, pp. 505-527 Mariott, R. H. 1933. "Die Absorption von Oel durch pflanzlich gegerbte Leder". JSLTC 17, p. 270 Mitton, R. G. and Pankhurst, K. G. A. 1956. "Fettausschlag auf Chromleder". Das Leder 7, 7, pp. 169-174 Noehles, M. 1984. "Lederpflege, Lederkonservierung". lecture given on international meeting of conservators, 7th July 1984 in Schlitz Raphael, T. and Mc Grady, E. 1984. "Leather dressing: A misguided tradition, de Froment, D. (ed) preprints, ICOM Committee for Conservation, 7th Triennial Meeting, Paris, 84.18.6 Reich, G. 2003. Vom Kollagen zum Leder. Freiberg: Ed. Forschungsgemeinschaft Leder e. V. und Verein für Gerbereichemie und Technik e.V. Reich, G. and Oertel H. 1987. "Beiträge zur Theorie des Fettlickerns". Das Leder 38, 3, pp. 41-47 135

ICOM-CC Leather and Related Materials WG Offenbach 2012

Schulze, A. 2011. "Goldleder zwischen 1500 und 1800. Herstellung und Erhaltung". Arbeitsheft 17 des Landesamtes für Denkmalpflege Sachsen, Dresden: Ed. Sax-Verlag Stather, F. and Laufmann, R. 1933. "Tranausharzungen auf pflanzlich gegerbtem Leder". Collegium 762, pp. 617-630 Tancous, J. J. 1965. "The mechanism of calcium soap formation in heavy hides and its influence on subsequent tannage". JALCA 60, 5, pp. 206-218 Trommer, B. 2008. Archäologisches Leder. Saarbrücken: VDM-Verlag Dr. Müller Trommer, B. 2006. "Spezialleder für den Instrumentenbau". FILK Freiberg: scientific report, BMWA-project n° IW 050201 Trommer, B. et. al. 2010. "Research on materials appropriate for restoration – Impacts of ancient and modern tanning techniques". JALCA 105, 5, pp. 138-172 van Hulten et. al. 2012. "Anti-soiling finishes for automotive leather". lecture given on 1. Freiberger Ledertage, 25th-26th April 2012 in Freiberg

Biographies Bernhard Trommer graduated with a master's degree in Leather Technology from the Technical University Chemnitz in 1986 and obtained his postgraduate degree in Environmental Control Techniques from the University of Applied Sciences Merseburg in 1990. He obtained his PhD in Archaeometry and Material Sciences from the Technical University Bergakademie Freiberg in 2005. He is project manager at FILK Freiberg and a member of VGCT and a member of ICOM-CC. Forschungsinstitut für Leder und Kunststoffbahnen (FILK Freiberg) - P. O. Box 1142 - D-09581 Freiberg, Germany - [email protected]

Andreas Schulze graduated in Restoration and Conservation Techniques from Dresden Academy of Fine Arts in 1989 and with a PhD in 2011. He worked at the Saxon States Authority for the Care of Monuments (Landesamt für Denkmalpflege Sachsen). Andreas Schulze was coordinator of the ICOM-CC working group "Leather and Related Materials" from 2002 to 2008. He received a professorship on art Technology and conservation of works of art at Dresden Academy of Fine Arts in 2012.

Heinrich Francke graduated in chemistry with a PhD at Technical University Bergakademie Freiberg in 1989. He is the head of the R&D department Leather Division of Zschimmer & Schwarz GmbH & Co. and a board member of VGCT.

136

ICOM-CC Leather and Related Materials WG Offenbach 2012

Reception of an 18th century Gilt Leather in the Historicism: Fortunate Exceptions or Characteristics Examples?

Wivine Wailliez, Eloy Koldeweij

Abstract A characteristic 18th century gilt leather pattern has enjoyed a huge reception during the historicism. This paper is focusing on eight embossed wallpapers from the last third and the turn of the 19th century. These are all inspired by a seminal design featuring at least one parrot on a vine stalk bearing two grape bunches, foliage and several flowers, the whole against a ribbed lozenge patterned background. The study of these patterns and their comparison with historic gilt leathers has allowed the authors to identify four original variations of the seminal design, all produced in The Netherlands and dating from circa 1740-1750. A fifth variation turns out to be a revival gilt leather pattern from the third quarter of the 19th century. This pattern has been copied and adapted too, notably by one of the most renown wallpaper manufactures of its time, Desfossé and Karth in Paris.

Keywords Gilt leather, leather paper, wallpaper, embossed wallpaper, parrot, historicism, Paul Balin, Van Herck, UPL, Desfossé.

Introduction

In the context of two different research-projects, one on historicist gilt leather imitations, and those of Paul Balin in particular, and a research on historic gilt leather on the other hand, we have been able to trace a recurrent gilt leather pattern in different wallpaper manufactories, all occurrences dating from the last third and the turn of the 19th century. The comparison with gilt leathers from the 18th century lead us to the identification of four different lineages, variations around a seminal design featuring at least one parrot on a vine stalk bearing two grape bunches, surmounted by a chrysanthemum and flanked by an acanthus leaf and two crossed tulips, a sinuous blooming peony tendril, the whole against a golden ribbed lozenge patterned background. A related fifth pattern lineage turns out to be a historicist creation.

Historical Context

During the second half of the 19th century, a new bourgeois upper middle class emerged brought about by the industrial revolution. Mechanisation made many, previously luxurious, goods, such as wallpapers, cheaper and thus affordable for a wider public. At the same time, world exhibitions and industrial fairs encouraged and promoted the latest discoveries, techniques and inventions.

The period which succeeded the Empire and Biedermeier was one of many ‘revivals’ - Greek, Byzantine, Romanesque, Gothic, Renaissance, Baroque etc. - summed up under the title of Historicism. In the fine arts and decorative arts there was during the second half of the 19th century much interest in revivals of these historic styles. In architecture and interior design, it was not unusual to find one or more reception rooms in wealthy and bourgeois houses decorated in the historicist or exoticist styles. In more common bourgeois households too, it was the tops to have at least the dining room in (Flemish) Renaissance (Thümmler, 2000, p. 123) or Gothic Revival. These rooms, halls, or passageways that led to them were often hung with imitation leather wall coverings. Gilt leather wallpapers are reported on industrial fairs as early as 18411 (Brussels, 1997, p. 67). Such hangings had become very fashionable and were promoted amongst others by the 'Magasin des Demoiselles' (Paris), which stated in 1864 that it was exceeding smart to have hangings of embossed leather or its imitations (Thornton, 1984, p. 223).

Research on Balin's Gilt Leather Wallpapers

In 2010-11 a research project, led by the Workshop for the study on finishes of historic buildings of the Royal Institute for Cultural Heritage (IRPA-KIK) in Brussels (Belgium), was devoted to the famous Parisian wallpaper manufactory of Paul Balin (1863-1898) and especially his imitations of embossed gilt leathers. We 137

ICOM-CC Leather and Related Materials WG Offenbach 2012

tried in particular to unveil the processes of his gilt leather wallpapers, given our interest for this topic (Wailliez et al., 2012). We have, indeed, already studied part of the production of the so-called Japanese leather papers also known as kinkarakawakami, produced for export in the late 19th century (Wailliez et al., 2005; Wailliez, 2007a and 2007b).

Paul Balin has founded his wallpaper manufactory in 1863 and has taken a patent on the embossing with a cold screw press in 1866. His aim is clearly expressed in this funding patent, followed in 1869 by a related patent, and three additional certificates: "par ce procédé nous imitons (...) l'ornementation du carton pierre, doré ou non, (...) les boiseries sculptées, (...) tous les ornements en pâte, (…) les vieux cuirs, (…) les laques de Chine, (…) l'effet de broderie etc."2 Numerous other patents provide evidence of Balin's constant research on the use of metallic effects in order to obtain convincing representations of gilt leathers on one hand and of silk fabrics on the other hand. The latter were his real passion and strong point in the application of metallic effects.

Balin's Historic Models

Balin demonstrated his excellence in particular in the production of raised designs, copying ancient decorative materials such as rich fabrics and embroideries, embossed leathers, and earthenware, either from public collections, or from his wide-ranging private one (Jacqué, 1991; Hans, 1992).3 Several of his gilt leather imitations are accurate copies of well-known antique leathers such as the vase and valance under a canopy-pattern4 (Koldeweij, 1998c, p. 27,31; Kassel, 2006, cat. 30) derived from Daniel Marot´s Patrons d'Étoffes et de Velours, and the Five senses, and a well-known Dutch gilt leather from ca. 1660 (Koldeweij, 1998a, vol. IV, cat.n° 103, 104; Glass & Koldeweij, 1998, cat.n° 38; Kassel, 2006, cat. 18).5 However, Balin's creations are more often the result of an interbreeding of different historic designs or trends and modern influences. Balin's designer, Arthur Martin, explained how a historic model was adapted to fit its new decorative task (Hans, 1992, p. 101, 104).

A Dutch Mid-18th Century Gilt Leather Pattern Interpreted by the Historicism

The recurrent 18th century gilt leather pattern, known in different variations, has enjoyed a certain critical fortune in the Historicism. Redrafted by Paul Balin, it has also been reproduced in several West-European and Japanese historicist wallpapers from the second half of the 19th and early 20th centuries. All together, five different lineages can be recognised (Table 1). These are being described here, together with their 18th century models.

First Lineage

The Balin pattern N° 4975 - ‘Exécuté d'après un cuir espagnol de l'époque Louis XIII’ (‘Executed after a leather from Spain during the reign of Louis XIII‘ according to the inscription on the usual label on the bottom right hand corner) - displays a parrot on a vine stalk, two tulips and a flexible acanthus leaf against an eye catching light embossed lozenge background motif (Figure 1). Striking is its similarity to a Dutch gilt leather panel dating from 1740-1750, in particular the parrot's head position, the acanthus lip, the grape bunch, and the mosaic-like chased background pattern. Besides a complete wall hanging in Ulriksdal Palace in Sweden several individual panels of this pattern do survive (Koldeweij, 1998a, vol. IV, cat.n° 249). Of these the panel in Musée des Arts Décoratifs in Paris is of particular interest as this has probably been acquired at the auction of Balin’s private collection in 19006 (Clouzot, 1925, pl. XIV) (Figure 2). On the Balin adaptation, however, a great part of the original 18th century pattern is missing such as the second bird, the second bunch, the caterpillar, the carnation etc.

138

ICOM-CC Leather and Related Materials WG Offenbach 2012

Figure 1: Gilt leather wallpaper sample Nr. 4975-N, manufactured by Paul Balin, from a sample catalogue housed at the Musées royaux d’Art et d’Histoire (MRAH-KMKG, Brussels), Inv. n°. BP584/5. © IRPA-KIK (X035501)

The discovery of a contemporaneous wallpaper of Balin with a related design, N° 49987, confirms that Balin has indeed copied the 18th century model, since many missing elements on N° 4975 can be found in pattern N° 4998 (Figure 3).8 Besides, another colourway of N° 4998 kept in the archive of the English wallpaper manufacturer Sanderson is well provided with the same golden light embossed lozenge background pattern as N° 4975. This wallpaper bears a label with the inscription ‘No. 4998 Exécuté d’après un cuir de Cordove (Collection P Balin)’ (= No. 4998 Executed after a leather from Cordoba) (London, 1985, cat.n° 33).9

These two Balin wallpapers differ slightly in the degree of alterations of the original pattern. Although the bird of N° 4998 is a mirror image and the flanking peony has been changed into a modern looking slender rose, the whole is less processed than its parrot-counterpart N° 4975. Both these wallpapers can probably be dated from the late 1860's or the beginning of the 1870's, according to their numbering.10

Figure 3: Gilt leather wallpaper sample N° 4998, manufactured by Paul Balin, from a sample catalogue kept at the Musée des Arts Décoratifs (Paris), Inv. n°. M5051.998.163.2.1. © Les Arts Décoratifs, Paris

139

ICOM-CC Leather and Related Materials WG Offenbach 2012

Lineage 1 Lineage 2 Lineage 3 Lineage 4 Lineage 5

Leathers

Wallpapers

140

ICOM-CC Leather and Related Materials WG Offenbach 2012

The modifications observed in N° 4998 can also be found on a Balin design entitled Grand dessin perse dating from 1869 (Clouzot and Follot, 1935) (Figure 4). This photogravure shows the full pattern of the 18th- century gilt leather – though slightly redrafted – learning that its designer did respect and appreciate the original design. This design with its light background colour could also be a draft for chintz wallpaper.11 The two tulips, chrysanthemum and butterfly copied after the 18th-century model are clearly present, whereas they have disappeared in the pattern N° 4975. A third wallpaper, derived from this Grand dessin perse can be recognized in the quite remote pattern N° 5557 (Figure 5). That pattern, however, has nothing to do with gilt leather, but is a fabric imitation in which floral and vegetal motives from the 18thcentury gilt leather are re-used: tulips, chrysanthemum, peonies, hydrangea, carnation and acanthus lip. This pattern is thereby testifying to Balin's eclecticism.

Figure 4: Wallpaper design entitled Grand dessin perse and dated from 1869, drawn from Clouzot et Follot, 1935, p. 227.

Desfossé

The famous contemporary French wallpaper manufacturer Jules Desfossé, equally renowned as Paul Balin for his gilt leather imitations, has edited a similar wallpaper. On the pages of an order book from the Cowtan Company devoted to Eaton Hall (Cheshire), the Duke of Westminster's country house remodelled by Alfred Waterhouse between 1870 and 1882, a tiny wallpaper sample, of which the bird-on-a-branch-against-a-ribbed-lozenge-background pattern is recognizable, is hand-annotated with the mentioning "Desfossé" (Thornton, 1984, fig. 461). In view of the very small sized and lacunar sample, it is not possible to attribute this specific paper to one of the lineages presented here.

Figure 5: Wallpaper sample N° 5557, manufactured by Paul Balin, from a sample catalogue kept at the Musée des Arts Décoratifs (Paris), Inv. n°. M5051.998.163.2.5. © Les Arts Décoratifs, Paris

Second Lineage

In the Brussels Hap mansion, a small country house expanded around 1900 to bring Hap’s notary practice and his private dwelling under the same roof, wallpaper imitating gilt leather is still existing in one of the rooms of the notary practice, above a high dado, made of a wallpaper imitating carved wood (Figure 6).12 Both the manufacturer and the exact date of this wallpaper are unknown.13 The reversed pattern on one hand and the representation of a single parrot (instead of two birds) turning its head backwards on the other hand, are the main differences with the pattern of the first lineage. Moreover, neither the butterfly nor the caterpillar can be found on this wallpaper. This second pattern is in every respect faithful to another mid- 18thcentury gilt leather pattern. This pattern, which is very similar to the previous one, has been traced in at 141

ICOM-CC Leather and Related Materials WG Offenbach 2012

least three different wall hangings and several individual panels, amongst others in the English country house Levens Hall and the deanery of Zele (Belgium) (Figure 7) (Peckstadt et al., 1992; Koldeweij, 1998a, vol. IV, cat.no. 250).

Figure 6 (on the left): Whole repeat of a gilt leather wallpaper kept in situ in the Hap Mansion (Etterbeek/Brussels), ca. 1900 © IRPA- KIK (X017852)

Figure 7 (on the right): One panel of the gilt leather wall hanging in the deanery of Zele (Belgium) ca. 1740-1750. © A. Peckstadt (Ghent, Belgium)

Third Lineage

A close reproduction – though mirrored – to the second lineage can be recognized on an embossed wallpaper sample from the Antwerp Van Herck manufactory (Glass, 2004, XXII) (Figure 8). Launched in 1873 by Jan-Frans Van Herck (1824-1903) and his son Eugène (1854-1941), the Fabrique de cuirs de Cordoue lasted two additional generations (Koldeweij, 1998b; Brussels, 1997, p.67-68). The private collection of the Van Herck family – a rather well known dynasty of antiques dealers in Antwerp –

comprised a large Figure 8: Whole repeat of a gilt leather wallpaper collection of gilt from the Van Herck manufactory, drawn from leathers aside all Druckmodel für Goldledertapeten, pl. XXII. © sorts of other Kunsthandel Glass, Essen (Germany) objects.14 Unfortunately, in the catalogue of the grandson Charles' (1884-1955) gilt leather collection – in its state before selling – there is not a single gilt leather panel that is related to our topic pattern, or which could have served as inspiration for Van Herck's wallpaper (Glass, 1998). Deviant the previous lineage, a second, very stiff, bird recalls very much the second bird of the first lineage. Furthermore, the painstaking comparison of both patterns reveals that the Van Herck wallpaper comprises the full pattern, but is actually broader and higher than the pattern of the second lineage. It displays an extra strip of peony tendril at one side, and a second range of grape

Figure 9: Gilt leather panel ca. 1740- bunches at the bottom. Therefore it is rather a hybrid between the 1750. © Eloy Koldeweij two models, expanded by side-extensions. Could this to be 142

ICOM-CC Leather and Related Materials WG Offenbach 2012

explained as a designer's trick to let his pattern match the own factory proportions? However, a recently discovered set of gilt leather panels, which showed up on the Dutch art market, shows an almost identical pattern. This set proves that Van Herck's pattern is not a Revival fantasy at all, but a faithful copy of an authentic 18th century gilt leather (Figure 9).

Fourth Lineage

Circa seven gilt leather panels in the collection of the Deutsches Tapetenmuseum Kassel (Germany), acquired in 1922 from the Marburger wallpaper manufacturer J.B. Schaefer, are said to have come from a demolished Amsterdam house (Koldeweij, 1998a, vol. IV, cat.n° 251; Kassel 2006, cat. 38) (Figure 10). Even as the gilt leathers mentioned before, these panels can be dated between 1740-1750 and have been made by an unknown manufacturer in either the Northern or the Southern Netherlands. This pattern displays parrot, vine shoot, grape bunch, crossed tulips, carnation, and butterfly motives referring to the seminal pattern in the focus of our study. It is a another variation on the theme, in which the parrot stands straight, its head turned to the left, and where the vegetal composition diverges from those observed in the other lineages. The background criss-crossing double line punching features a face-centred lozenge trellis recalling the ribbed lozenge background of the other lineages.

A wallpaper produced by the Usines Peters-Lacroix (UPL), established in 1872 in Haren (Belgium), shows a related pattern (Figure 11), even though the composition differs slightly from the 18th century model and its evenly coloured background is flat without any ornamentation. Considering the trellis background pattern, an almost identical embossed wallpaper, i.e. with the same main motive (parrot, tulips etc.) kept in situ in the Paul Tétar van Elven Museum (Delft, The Netherlands)15 shows, on the contrary, a slightly cylinder- embossed lozenge patterned background. Quite likely this paper was hung by the redecoration of that room in the 1920’s (Van Burkom et al., 2001, p. 187).

Figure 10 (on the left): Panel of a gilt leather hanging ca. 1740-1750, from an Amsterdam house. Inv.n°. DTM 19/19. © mhk, Deutsches Tapetenmuseum (photographer: A. Hensmanns).

Figure 11 (on the right): Illustration of the section “Cuirs japonais” in a small commercial brochure from Usines Peters-Lacroix, beginning of the 1960's. © Wivine Wailliez.

The UPL wallpaper is strangely advertised under the section "Cuirs japonais" in a small commercial brochure without date (UPL Histoire de la fabrication du papier peint, n.d.). The advertising message says: "Nos cuirs japonais sont des reproductions de modèles classiques dans les différents styles. Décorés au moyen de 143

ICOM-CC Leather and Related Materials WG Offenbach 2012

planches gravées, et finis à la main, ils offrent grâce à leur fond d'or ou d'argent et à leur fort relief, une très grande richesse décorative." (= Our Japanese leathers are reproduced from classic models in different styles. Decorated by engraved moulds, and finished by hand they offer a highly decorative grandeur, thanks to gilded or silvered background and its strong relief). Though the original gilt leather archetype presents, indeed, a gilded background, the paper reproduced in the UPL advertisement does not. It shows instead a reversed relationship where the gilded embossed motives stand out against a flat coloured background. Was this UPL wallpaper available in various colour schemes? It is however rather odd to use in an advertisement an argument that has visibly not been implemented. Another interpretation of this advertising claim is that the locution "leur fond d'or" is not aimed at the background but at the grounding – that is to say the first finish applied to the substrate. In most cases, the so-called Japanese leather papers actually mimic the gilt leather gilding process, i.e. display a yellow coated tin leaf (Wailliez et al., 2005 ; Wailliez, 2007a).16 According to published pictures by UPL (UPL Usines Peeters Lacroix, n.d. (1972), pp. 74-77) and our own observation on another UPL gilt leather imitation, the gilding process is the first step and consists in fact of gilt lacquered white metal leaf grounding.

Does the title Cuirs japonais indicate that the Belgian firm UPL has been aware of this pattern through kinkarakawakami – produced from 1873 onwards for the Western market –, rather than by historic gilt leathers remaining in Belgium or the Netherlands? In fact, UPL did not only develop their own designs but also adapted designs bought all over the world “principalement en Europe mais aussi dans des pays aussi éloignés que le Japon.” (UPL, (1972), p. 28). This indicates that UPL was aware of the designs of various Japanese wallpapers. But the question of the main source of inspiration for this specific pattern has not yet been resolved.

Fifth Lineage

A fifth lineage displays the main features of the seminal design against a slightly embossed gilded lozenged background. A complete repeat of this gilt leather pattern is in the collection of the Metropolitan Museum of Art in New York (USA) (Figure 12). A number of occurrences of this pattern are known, of which several are still in situ, amongst others in Italy, France, The Netherlands, Poland, and Sweden (Mallé, 1970, pp. 428- 429; Bender, 1992, cat.n° 164, pl. 36; Fournet, 2004, 802, 1436). The foliage with the flowers, amongst which tulips, chrysanthemum and acanthus, and the butterfly is similar to the first lineage. Other elements however do show it is another variation on the theme: the two in-flight birds catching insects, the pear in place of the right grape bunch, and the central full-blown peony converted into a big pomegranate. The inscription ‘droit’ on the backside of the leather in the Swedish castle Skenas, hung ca. 1878, indicates it has been produced either in France or in Belgium. This pattern was already available in 1871, when it was hung in the billiard room of the Dutch country house Kasteel ter Hooge in Middelburg (Koldeweij, 1998a, vol. I, p. 293, ill. 205). Even though this gilt leather has often been labelled as an 18th century leather pattern, there is no doubt at all that it dates from the third quarter of the 19th century and was available at least up to 1910 (Mioland & Lelogeais, 1910, pl. 9).

This widely used pattern has freely been adapted by the Parisian wallpaper manufactory Desfossé et Karth (Figure 13).17 The acanthus leaf has become there a fern tendril and – as for Balin N° 4975 – the foliated scroll has been cut off and converted into a twig. This 19th-century gilt leather pattern has also been faithfully taken over in Japan. In that case, however, the ribbed lozenge background pattern has been replaced in one case (Tokyo, 1984, p. 24) by a creped ground,18 and in another case19 by a specifically Japanese (relief) background pattern, the so-called bishamon kikko i.e. the tortoise-shell armour pattern of the Japanese hero, Bishamon (Figure 14).

144

ICOM-CC Leather and Related Materials WG Offenbach 2012

Figure 12 (on the left): Gilt leather panel in the Metropolitan Museum of Art, New York (Inv.no. 37.17), France or Belgium, ca. 1870. © Metropolitan Museum of Art, New York

Figure 13 (on the right): Gilt leather imitation wallpaper from the Desfossé et Karth manufactory (1875), kept at Musée des Arts Décoratifs, Paris. © Eloy Koldeweij

Figure 14: Left: Kinkarakawakami i.e. Japanese gilt leather wallpaper. Right: Detail. Modern reproduction by Ueda Takashi with the original wooden roller kept at the Paper Museum (Tokyo, Japan), n.d. Illustration drawn from Suga, 2007.

The Context of the 18th Century Prototypes

As mentioned above, these four 18th century gilt leather patterns can all be dated on stylistic grounds circa 1740-1750, and all four they have been produced in The Netherlands. Unfortunately none of the traced surviving examples of any of these four patterns has been documented by archival documentation. As a consequence it is impossible to date these more precisely, or to attribute them to specific workshops in either the Northern or the Southern Netherlands. There is hardly any doubt that these four variations on this specific pattern will have been made by different workshops. The reality of today that successful products are being copied by competitors did also exist during the 17th and 18th centuries. For the gilt leather manufacturers this was a continuous and well known problem they were facing with. Of many 17th and 18th century gilt leathers two or more versions are known and many archival documents are dealing about this issue (Koldeweij, 1998a, vol. 1, pp. 191-193). Of the more than 200 different embossed patterns from the 17th and 18th centuries which have been traced, 44 patterns (= 22 %) with one or more copies are known to exist. As such, ‘our small group’ of four gilt leather patterns which are very alike, is no exception. 145

ICOM-CC Leather and Related Materials WG Offenbach 2012

Another interesting aspect is the date of the 18th century leathers, and the position of these four 18th century gilt leather patterns within the entire group. In The Netherlands embossed gilt leathers were designed and made up to about 1780 (Figure 15). In total 117 different 18th century embossed gilt leather patterns have been identified, of which not less than 89% dates before 1750. ‘Our group’ of four embossed patterns (only 3,4% of the whole) have been made in the last decade of a period of prosperity, and date just before the last embossed patterns from The Netherlands would perform their swan sung. A remarkable aspect of this group of patterns is however, that in contract to most other patterns, the background is always gilded and the pattern painted. Up till this moment no mirror-image of one of these patterns has been traced which have a gilded pattern against a painted background.20

Figure 15: Diagram showing the quantity and distribution in time of the 18th century Dutch embossed gilt leather patterns. © Eloy Koldeweij

The Originals and their Copies: Characteristics and Distinctions

These five lineages of 19th century wall hangings have all been inspired on 18th century gilt leather patterns. Four of these are exact, or rather exact, copies with just some minor differences, only the fifth lineage is a free adaptation, using the historic elements as an inspiring source. Both stylistically and technically there is no doubt at all that the fifth lineage dates from the second half of the 19th century.

Looking to the four 18th century gilt leathers and their 19th century counterparts, several striking differences can be discerned between both these groups. First of all, no embossed papers were ever made during the 18th century, these all date from the second half of the 19th century or later. Secondly, almost all leathers from the second half of the 19th and the early 20th century are of a far less quality than its 18th century counterparts. Due to the industrial tanning methods used, a large percentage of the 19th century leathers (start to) suffer from so-called red-rot, they are colouring red and are slowly breaking up. Another feature which can be found on some of the revival leathers is a wood-pulp filling, which is stuffing the embossed parts from the backside. Such fillings never occur on 18th and 17th century gilt leather panels. Besides, a characteristic encountered in the 19th and early 20th century revival leather hangings is that the original rectangular sizes of the individual embossed leather panels have been ignored: large hangings with relief patterns, so in one piece, were produced without any interruptions or stitchings in whatever way. This was possible by joining and gluing the individual leather panels together before they got their embossed patterns and their finishes. Although many large gilt leather panels from both the 17th and 18th centuries do survive, it was impossible at that time to give these an embossing as was being done post 1850. Large gilt leather panels from the 17th and 18th centuries panels are (partially) stenciled, painted or glued with flock and the gilded parts punched, but never embossed. This was at that time technically impossible. Moreover, the embossed forms of most of the revival leathers differ too by their very sharp and refined, hard details, and they often show stylistic contradictions. 146

ICOM-CC Leather and Related Materials WG Offenbach 2012

While the gilding on the 17th and 18th century gilt leathers is made by silver leaves with a yellowish varnish, resulting in a rich and bright gilding, most of the factories making revival leathers and wallpapers have been using everything but silver. Alternative metals or alloys from tin, zinc, copper, brass or lead were used. Unfortunately, no analytical researches have been done on this, aside from the researches to the Balin wallpapers and some Japanese papers. The gilded parts of the revival leathers have almost always a complete different look and are usually far less shinny. Together with the applied colours this gives quite another appearance to the leather: in most cases the colours of the 19th and 20th century wall hangings imitating the old gilt leathers are dark, tempered and heavy (brown, dark blue, dark red or dark green). Vivid contrasts and violent, bright and pure colours (blue, red, orange, pink, yellow, white, grey, black, and brown), that were so common and characteristic during the centuries before, were shunned. Within the wallpapers that are imitating gilt leather light and bright colours do however occur.

The differences between the 18th and 19th century embossed wall hangings are however not always as clear as may be suggested here. Quite a few exceptions do exist. These generalised characteristics cannot be trusted blindly, sometimes due to the quality of the copies, sometimes due to the aging of the originals. The discoloration of the pigments, the colour of the aged varnishes, the tarnishing of the metal finish, and the attached dust and dirt can make it difficult to make a correct identification. However with the use of modern analyzing techniques like electronic microscopy, and determination techniques as X-ray fluorescence (XRF), Energy-dispersive X-ray spectroscopy (EDS, EDX, or XEDS) or Micro-Fourier Transform Infrared Spectroscopy (FTIR), and High Pressure Liquid Chromatography (HPLC, used for the identification of dyestuffs) it is possible reveal the use of modern (i.e. synthetised) pigments as e.g. Scheele’s Green (post 1775), Chrome pigments (post 1809), or Zinc white (post 1780, most probably begin 19th century), or the use of curious of cheap metal-alloys as spelter (zinc alloy). Artificial dyestuffs used as lake-pigments are also good time-markers. Nevertheless most of these techniques are rather labor-intense or costly. In most cases the experienced eye will be able to catch the stylistic differences or to discern leather from the second half of the 19th or early 20th centuries from leather that is one or two centuries older.

Conclusion

The knowledge about revival imitation, 19th and 20th century gilt leather copied from and inspired by 17th and 18th century gilt leather prototypes, is still rather limited. This article must be seen as a first step of this research process. Up till this moment far four West-European manufactories (Balin, Desfossé & Karth, Van Herck, and Usines Peters-Lacroix) have been identified which have been copying a mid-18th century seminal design. British and German productions have not yet been researched, though these countries are likely to reveal some Revival copies too, for example by Jeffrey and Co, who were very reputed for their gilt leather wallpapers.

Acknowledgments We are very grateful to Mrs Ann Peckstadt (conservator-restorer), Mrs Véronique de la Hougue (curator, MAD), Mrs Dr. Astrid Arnold-Wegener (curator, MHK), Mr. Horst Glass, and Mr. Jean-Pierre Fournet, for their kindly provided information and pictures.

References Brussels 1997. Le Murmure des murs. Quatre siècles d’histoire du papier peint, exh. cat., Galerie CGER. Kassel 2006. Thümmler S. and Gerner C.E. (ed.), Goldrausch. Die Pracht der Goldledertapeten, exh. cat., mhk, Kassel. London 1985. Ch. Woods (ed.), Sanderson 1860-1985, exh. cat. Tokyo 1984. Kinkarakawa. kawa to kami no Tozai koryu (The connection between leather and paper of the East and West), exh. cat., Ina Booklet, Vol.3 N° 4. Bender A., 1992. Zlocone Kurybany w Polsce, Lublin. Bruignac- La Hougue V. (de), 2000. ‘Une acquisition par dation du Musée des Arts décoratifs, Paris : Une paire de panneaux en papier peint de la manufacture Arthur & Grenard et huit souches d’échantillons de la manufacture Paul Balin’, Revue du Louvre: la revue des Musées de France, vol. 50, N° 3, pp. 23-24. Brunhammer Y. and Nouvel O., 1990. « Les Papiers peints », in Antiquité et Objets d’Art, n° 9 : Tissus, tapis et papiers peints, Paris, Editions Fabbri, pp. 58-74. 147

ICOM-CC Leather and Related Materials WG Offenbach 2012

Burkom van F., Gaillard, K., Koldeweij, E., Schulte, T., and Willink, J. (ed.), 2001. Leven in Toen, Vier Eeuwen Nederlands Interieur in Beeld, Zwolle 2001, pp. 186-187, 285. Clouzot H. 1925. Cuirs décorés, Paris. Clouzot H and Follot Ch., 1935. Histoire du papier peint en France, Paris, Ch. Moreau. Fournet J.-P. 2004. Les cuirs dorés anciens en France, Paris, École du Louvre, mémoire d'étude approfondie (unpublished), 11 vol. Glass H. and Koldeweij E., 1998. Bedeutende Goldledertapeten 1550-1900, Kunsthandel Glass, Essen. Glass H., Koldeweij E. and Tielker E. H., 2004. Druckmodel für Goldledertapeten, Kunsthandel Glass, Essen. Hans P.-X. 1992. "Les tentures de style de Paul Balin (1832-1898)" in J. Cuisenier (ed.), L'œuvre en multiple, Paris, pp. 99-114. Jacqué B., 1991. "La perfection dans l'illusion ? Les techniques de fabrications de la manufacture Balin 1863-1898", Bulletin de la Société Industrielle de Mulhouse, n° 823/4, pp. 107-115. Koldeweij E.F. 1998a. Goudleer in de Republiek der Zeven Verenigde Provinciën. Nationale ontwikkelingen en de Europese context, Amsterdam (unpublished dissertation Leiden University), 5 vol. Koldeweij E., 1998b. “The Van Herck Collection of Gilt Leather“, in Glass H., Koldeweij E., Bedeutende Goldledertapeten 1550-1900, Essen, 1998, pp. 8-9. Koldeweij E., 1998c. “Gilt leather and its patterns: sources of inspiration“, in Glass H., Koldeweij E., Bedeutende Goldledertapeten 1550-1900, Essen, 1998, pp. 24-32. Mallé L., 1970. Palazzo Madame in Torino, Torino. Mioland & Lelogeais, 1910, Sièges artistiques en cuir repoussé: cuirs décorés, ciselés dorés & pyrogravés. Tentures d'art, [Paris] Oman, C.C., and Hamilton, J. 1982. Wallpapers: a history and illustrated catalogue of the collection of the Victoria and Albert Museum, London. Peckstadt A. et al, 1992. “Conservering van het goudlederbehang van de dekenij van Zele”, Monumenten en Landschappen, 11/n° 6, Nov.- Dec. 1992, pp. 55-61. Rein J.J., 1886. Japan nach Reisen und Studien im Auftrage der königlich preussischen Regierung dargestellt, 2 vol., Leipzig, 1881-1886. Suga Y., 2007. Kinkarakami. The Art of Japanese Leather Paper, Tokyo. Thornton P., 1984. Authentic Decor, London. Thümmler S., 2000, Tapetenkunst. Französische Raumgestaltung und Innendekoration von 1730-1960. Sammlung Bernard Poteau, exh. cat., Kassel. UPL Histoire de la fabrication du papier peint, n.d. (before 1964), Brussels. UPL n.d., UPL Usines Peters-Lacroix s.a., n.d. (1972), prefaced by P. van Mollekot. Wailliez W. et al., 2012. "Style and substance: Balin's gilding techniques revealed for the first time", Wallpaper History Review, 2011/12, pp. 5-12. Wailliez W. (2007a). "Kinkarakawakami oder japanische Goldlederpapiere – Teil 1. Historische Grundlagen und Hestellverfahren", Restauro, 3/2007, pp. 168-177. Wailliez W. (2007b). "Kinkarakawakami oder japanische Goldlederpapiere – Teil 2. Fälle und Auswertung", Restauro, 5/2007, pp. 321-327. Wailliez W. et al., 2005. "Les papiers peints japonais en Europe occidentale autour de 1900", Preprints of the 14th Triennal Meeting ICOM-Comittee for Conservation, The Hague, 11-16 Sept. 2005, Vol. II, pp. 869-874.

Endnotes 1 Catalogue des produits de l’industrie belge admis à l’exposition de 1841, envoi d’Alexis Lefebvre, rue de Ransfort à Molenbeek-Saint-Jean-lez-Bruxelles, p. 151. 2 Patent N°71160. Pour l'application du balancier à estamper à froid dans la fabrication des papiers peints. 3 Balin’s private collection included 1406 items (Jacqué 1991, p. 110). It was put up for sale at Drouot after his death. Catalogue des étoffes anciennes composant la collection de feu Paul Balin, vente les 18, 19, 21, 22, 23 mai 1900, Hôtel Drouot, Maître Sarrus Commissaire-priseur: « Soies, damas, lampas, brocatelles, brocarts et velours du XVe au XVIIIe siècles; Tapisseries au point; Broderies et applications sur soie, velours ornements d’église et tissus divers du XVIe au XVIIIIe siècle; Cuirs anciens français, hollandais ou espagnols du XVIe au XVIIIe siècle; Carreaux et faïences anciennes de Perse et de Rhodes. » 4 Balin's production number N°5082. 5 An example of the Balin reproduction is kept in Musée des Arts Décoratifs (MAD), Paris (ref. 9775 23A; inv.n° HH 895), illustration in: Y. Brunhammer and O. Nouvel, 1990, p. 69. 6 According to documents kept at the MAD. Communication from Mr. J.-P. Fournet, November 2012. See also Fournet, 2004, cat. 209, p. 941.

148

ICOM-CC Leather and Related Materials WG Offenbach 2012

7 MAD, sample catalogue 1, inv.n° 998.163.2.1 - N° 4998. In this sample, however, the background is kept flat without any embossment. 8 The close numbering is an indication for a quasi contemporaneous release. 9 Arthur Sanderson was Balin's agent in the United Kingdom. In the Sanderson archive more than 900 Balin wallpapers are being preserved. 10 We have compared their numbering with these of the fifty-seven Balin textile imitation wallpapers exhibited at the Vienna World Exhibition of 1873 and given to the museum by Paul Balin in 1874. Oman and Hamilton, 1982, pp. 278- 281. See also the collection database of the V&A Museum: http://collections.vam.ac.uk/ (consulted in April 2012). 11 We have not been able to trace this specific Balin wallpaper in any of the European wallpaper collections. 12 Address: Chaussée de Wavre, 508, Brussels. See for the dado the photo-collection of the IRPA-KIK: http://www.kikirpa.be/EN/45/63/Photolibrary.htm. Cliché number X017904. 13 The leather-paper imitation might have been produced in Belgium. 14 The ceramics, terracotta and drawings from this collection have been acquired by the non-profit King Baudouin Foundation (see www.collectionvanherck.be); the gilt leathers have been sold to Galerie Glass in Essen (Germany). 15 It was the home of the Academy painter and art collector Paul Tétar van Elven (1823-1869) between 1864 and 1894. 16 Period description in the American journal, The Decorator and Furnisher, June 1886, pp. 84-85, and Wailliez W., 2007a, pp. 174-175. 17 Desfossé et Karth, Sample book 52437 ref. 6829, dating from 1875. Collection MAD, Paris. Communication from Mrs de la Hougue, whom we thank warmly. 18 Technique intended to imitate a leather texturation as usual by the older Japanese leather papers. Rein, 1886, pp. 486-89 and 491. 19 Wallpaper from undetermined date and manufactory but probably by the Imperial Printing Bureau (i.e. after 1880). 20 This is however no prove that such colour-variations have not been made. Of the just over 500 different gilt leather patterns that can be described to The Netherlands from both the 17th and 18th centuries, not less than 222 patterns seem to have survived in just one single surviving example. Koldeweij 1998a, Vol. I, p. 304.

Biographies Wivine Wailliez, a conservator specializing in historic interiors, entered the conservation in 1992 as a laboratory assistant at the Royal Institute for Cultural Heritage (Institut Royal du Patrimoine Artistique – Koninklijk Instituut voor het Kunstpatrimonium aka IRPA-KIK), Brussels (Belgium). Between 1994 and 1998 she studied sculpture conservation at the École Nationale du Patrimoine - Institut de Formation des Restaurateurs d’Œuvres d’Art (ENP-IFROA), Paris (France). After working in the workshop for polychrome wooden sculpture at the IRPA, she founded and headed a new workshop for the research on finishings of historic buildings in 2001, still at the IRPA. She is responsible for an inventory of wallpapers in Brussels. Institut Royal du Patrimoine Artistique - Parc du Cinquantenaire 1 - 1000 Brussels, Belgium - [email protected]

Dr. Eloy Koldeweij (1959) studied History of Art at the Leiden University, The Netherlands, and specialized in the Decorative Arts and historic interiors. In September 1998 he got his doctorate on Dutch gilt leather. Between 1982 and 1997 he had several freelance jobs in several Dutch museums, the Victoria & Albert Museum in London, the Leiden University, an American private collector and a Dutch private trust for the conservation of important architectural monuments. Since September 1997 he is working as a specialist on historic interiors at the Cultural Heritage Agency in The Netherlands. He has written extensively and edited many publications on gilt leather and various elements of the Dutch historic interior. Cultural Heritage Agency - P.O. Box 1600 - 3800 BP Amersfoort, The Netherlands - [email protected]

149

ICOM-CC Leather and Related Materials WG Offenbach 2012

Leather and the Medieval Christian People of the Fourth Cataract Region of the Nile: Two Case Studies Barbara Wills

Abstract Leather or skin garments and objects are being studied as part of a two-year Clothworkers’ Foundation Senior Conservation Fellowship project. The aim is to clean and stabilise a group of fragile, naturally-mummified human remains and provide a suitable mounting and storage system so that they may be fully studied. The group of around 40 medieval bodies came to the British Museum as a result of rescue excavations in the Fourth Cataract region of the Nile. Though still partially obscured by soil, an intriguing range of tissues and associated material including items made from skin products is being revealed. Two of the bodies, the subjects of this paper, are wrapped in animal skins with the hair on. Skeleton 1110’s ‘cloak’ is multi-layered with long, wavy, blue dyed hair. Skeleton 4431 is almost fully clothed in leather/skin garments. Detailed observation and recording, species identification and tannin analysis are underway. As the project is at an early stage, further research is planned which will allow a better understanding and appreciation of this material and culture.

Keywords Leather, skin, Nile valley, Sudan, medieval, human remains.

Introduction

The aim of the two-year project Safeguarding a body of evidence: researching and conserving a group of exceptional naturally-mummified Nilotic human remains is to clean, reveal, stabilise, and begin to study a group of naturally-preserved bodies and their wrappings. The work is being supported with charitable funding from the Clothworkers’ Foundation. The Conservation Fellowship award allows the Senior Fellow to step aside from routine work to focus on the project while a Junior Fellow undertakes the Senior Fellow’s usual duties.

Naturally-mummified bodies are rare and this group survived mainly due to burial in consistently dry conditions. The forty-one medieval bodies were recovered from cemeteries in the area of the Middle Nile Valley, Sudan, and date from the sixth to fifteenth centuries AD, the Christian era. The Fourth Cataract salvage excavations (1999–2007) were undertaken prior to the building of the Merowe Dam and the Sudan Archaeological Research Society (SARS) was licensed by the National Corporation for Antiquities and Museums in Khartoum to excavate sites in the affected region. Study material (of which these bodies form a part) was given to SARS who then donated it to the British Museum (BM).

The bodies show remarkable preservation of human tissues and possess a wide range of original wrappings, mostly garments. Teeth and bones survive well. Soft tissues such as skin, eyes, muscle, tendons, toe/fingernails and hair are present in varying states of preservation from very poor to good. Some bodies have tattoos. Information retrieval possibilities therefore exist beyond that offered by skeletal remains alone. Two bodies, the subject of this presentation, are wrapped with skin or leather ‘cloaks’ which retain the hair. One body has three leather amulets apparently ties to the lower legs. To simplify the definitions, skin or leather is called ‘skin product’, and the skin products described here with the fur or hair still attached are called ‘sheep/goatskin’.

Working with Human Remains The acquisition of human remains requires the holding institution to preserve them with care and respect, and to make them available for study. Their special status carries the obligation to follow international, national and institutional protocols of care. The UK government’s Department of Culture, Media and Sport 2005 Guidance document states that any conservation should be carried out by an accredited conservator trained and experienced in caring for biological materials, and that such work be overseen by an osteologist. In this case, the appropriate guidance is given by Daniel Antoine (BM Assistant Keeper and Institute for Bioarchaeology Curator of Physical Anthropology). The human material is investigated and laid out on his 150

ICOM-CC Leather and Related Materials WG Offenbach 2012

advice, and interventions discussed and agreed before treatment. The approach in this project is to use minimal or no chemicals, adhesives or consolidants because of the risk of compromising present or future analyses. The aim is to mount the bodies sensitively, and implement well-considered handling protocols when the material is studied in future.

The risks inherent in the treatment, storage and study of this group were considered at an early stage. This involved identifying tissues and wrappings, and learning how best to clean and secure these while exploring methods for sensitive mounting. Bodies are inherently complex, diverse in structure and strength, and in this case the surviving tissues are often dislocated, partial and extremely fragile. The wrappings similarly vary. The mounting system is thus designed to accommodate complex contours, areas of inconsistent stability and varying layouts. During cleaning, samples are taken and mapped in relation to the body. These will be stored accessibly on a tray beneath the mount. An adaptable support plan using recently-tested materials is developing which may also be of use to others.

Case Study 1: Skeleton 1110

Object Description The body provisionally known as skeleton 1110 was excavated with others from a cemetery on the island of Mis. The largest burial site on the island, it had over 500 medieval grave monuments varying in size and style, aligned roughly east-west (Ginns 2010). Skeleton 1110 (Sk1110) was found close to the church wall. The body came to the Museum resting on the original excavation mount made from locally-available materials (wood battens, hardboard, foam sponge, polyethylene sheeting, polyethylene Grip Seal bags stuffed with raw cotton, plastic ropes and nylon mesh). On reaching the Museum, the body (with others in the group) went through a freezer treatment to kill any pests. It is stored on shelving in the dedicated Human Remains store on a melamine-laminated, medium-density fibreboard under an acid-free tissue cover.

Sk1110 was at first difficult to understand visually as the surface was covered with stony soil and sand (Figure 1). Examination showed much loss of human skin; what remains is mostly near the upper surface of the hips, shoulders and ribcage. Post-mortem insect activity was one reason for loss (clearly indicated by the characteristic holes and rounded edges of otherwise robust areas of skin). Detached skin and other fragments were found among the loose grave soil. All tissues and textiles had areas with concreted soil attached. The skin is of varying colours (pale buff to grey-brown and deep red brown) and incompletely attached to the bones, often with a void beneath. Tendons or ligaments occasionally survive and some cartilaginous tissue remains, in situ and elsewhere. The head and much of the feet are missing. Though many are disturbed, the bones are mostly intact and (more or less) in place. The soil associated with the gut is of a darker colour than other soil and finer in texture, and is bagged separately for future investigation.

151

ICOM-CC Leather and Related Materials WG Offenbach 2012

Three types of textile were noted (Figure 2). Closest the body lies a tailored, ochre-coloured plain-woven linen garment. Above and around this is a wrapping of fine silk, plain-weave blue, red and white fabric, as well as a colourful and complex woollen textile. The condition varies from good to very poor: all textiles show losses due to natural decay and insect activity.

Around the shoulders and beside the right torso area down to the level of the knees is a multi-layered, folded ‘cloak’ made from skin products. The grain pattern of one layer can be clearly seen at a de-haired edge; it most closely resembles goat. At least six different layers of skin product are visible near the right shoulder; elsewhere three layers are evident (Figure 3).

Alongside the innermost layer, a curly fleece of a deep blue colour is found, generally detached. The staple is c. 50-70 mm in length (Figure 4). The ‘cloak’ shows frequent splits and breaks; it is brittle and cracking, with disconnected pieces elsewhere in the detritus. Although only a limited area can be seen, a profusion of leather/skin thongs hold layers together with running stitches. At least one is z-twisted and circular in section, others are flat strips of rectangular section or with the edges folded over so that they meet in the middle. The purpose of some of the stitching is unclear. Several seams run closely in parallel, and some are perpendicular to each other creating a large grid-like pattern. The average length of visible stitches is between 10-12 mm, width is c. 2-3 mm. Shorter stitch lengths, (average 6 mm) are found, often associated with starting knots. Longer stitches (16-19 mm) run along the edges, particularly the bottom edge. The overall size of the cloak is c. 1160 mm from shoulder to the end below right knee. The width is difficult to estimate because of the folding, but not less than 500 mm across.

The Amulets Three amulets came to light on clearing soil away from the lower limbs of the body. One was detached, lying near the left knee. It is disc-shaped with remnants of a leather tie. One side is more rounded (because of the contents). Running stitches secure the two halves together. A second similar amulet was found attached to the left leg. A third, cylindrical, amulet was discovered beneath the right shin (Figure 5). It has a more complex design with an impressed linear pattern and the thong tie is finely plaited with a decorative knot.

152

ICOM-CC Leather and Related Materials WG Offenbach 2012

A tradition of using amulets (hijabs) for protection is still extant in North African societies. Common in medieval Nubia, they preceded the Christian era and survived the transition to Islam. Hijabs consisted of religious/magical texts on parchment, paper or papyrus which were tightly folded and stitched into ornamental leather covers. Thongs allowed them to be tied to the person or to valued animals or objects (Adams 1993).

Treatment / Results A dedicated tool set is used for cleaning human remains at the Museum. In this instance, a variety of spatulas, from metal to Melinex® was initially used to micro-excavate, carefully checking and sieving soil to find anything of relevance to keep as distinct sample material. The vacuum cleaner (low powered, adjustable with Hepa filter) allowed cleaning of the final dust layer which was lightly brushed towards the nozzle. Alternatively, the nozzle was held over the surface to vacuum through a protective screen. Where it was potentially damaging, concreted soil attached to body tissues or textile wrappings was removed by crushing between blunt-ended tweezers. Samples were put in labelled polythene zip-lock bags and the positions noted, numbered and mapped on a grid diagram of the body. Samples of materials used on excavation to lift, wrap and pad were also kept for future reference.

Where the various layers of human tissue or textile were delaminating and required re-securing in position, this was done by applying Plastazote ‘clips’ which were made to clamp the layers gently. Tyvek® ties were also useful in holding areas together. Additional internal and external support and padding was placed where needed during the process of cleaning, such as within the body cavity when the ribs were temporarily removed.

The blue woollen fibres from the colourful textile and deep blue woolly fleece fibres were analysed by Fourier transform infrared spectroscopy (FTIR) and confirmed to contain indigotin dye (Figure 6).

Several layers of the leather cloak were tested for the presence of vegetable tannins using the iron tannate test (Van-Driel Murray 2002) and proved negative. Where the leather was very dark brown no testing was done as the diagnostic blackening would not be discernible.

Mounting A new mount board was made of Cellite® 220 and covered with a sheet of Plastazote LD33 (expanded polyethylene) covered with Tyvec®. The body was transferred, sometimes piecemeal as necessary, keeping as large an area intact as possible and using polythene sheets as support, and several pairs of hands to help. Supporting blocks and wedges were cut from Plastazote, modified and shaped to achieve the right contour and smoothness. Methods included the use of a thermocutter and sharp blades. Edges in contact

153

ICOM-CC Leather and Related Materials WG Offenbach 2012

with the body were smoothed using one of several techniques: covering with Tyvec®, with polytetrafluoroethylene (PTFE) sheet (Relic Wrap™), smoothing with a finger while warm from the thermocutter, or rubbing with rasps and/or abrasive papers. Where a padded surface was needed on the Plastazote, a layer of polyester wadding was inserted beneath the Tyvec®, alternatively, to make it softer still, Relic Wrap™. The shaped wedges and supports were secured with stable steel pins.

Figure 6: FTIR spectra of undyed and blue fibres with indigo reference.

Case Study 2: Skeleton 4431

Description and Condition The body also came from the church cemetery site at Mis but was less obscured by the soil (Figure 7). It appears very complete and relatively robust. Enclosed from head to knee in a sheep/goatskin ‘cloak’, the interior areas of the body remain hidden. The size of the outer wrapping garment is difficult to measure accurately because few edges are accessible. It is made from three broad rectangular bands of sheep/goatskin. The colour of the hair varies, alternating dark brown and yellowish. Strips of precisely-cut fine dark skin product (35-40 mm width) back the joins, and a similar material is used as edging (10 mm on flesh side, less on the hair-side), secured with a thong running stitch (as found with other stitched joins). The sheep/goatskin shows signs of wear, especially over the head where the hair side can be seen through parallel breaks in flesh side of the skin. A number of later (or original processing) repairs are evident where skin product patches are stitched over a hole. Many areas are cracked, detached, split and broken, as fragile and brittle as Sk1110. A whitish plain-woven textile partially wraps the upper back. Figure 7: Sk4431 in box from excavation. © British Museum

154

ICOM-CC Leather and Related Materials WG Offenbach 2012

Examination of the grain surface of the samples to identify the species with Yvette Fletcher (Director, Leather Conservation Centre) did not lead to any positive identification because the surfaces were too smooth and featureless.

Areas of the leather cloak were tested for the presence of vegetable tannins using the iron tannate test, which proved negative.

Treatment / Results The body and wrappings remained in the original wooden crate during cleaning and examination (Figure 8) as removal requires colleagues with expertise and the appropriate facilities which will be available later in the project. All accessible areas have been cleaned, recorded and sampled as with sk1110. When carefully removed from the crate the body will be mounted similarly.

Discussion

What factors accelerate or inhibit the decay of desiccated archaeological leather or skin in this context? Within this group, burial patterns vary but a common feature is the laying of the body directly in the grave cut without a coffin. Often part or all the body was protected by stone lintels placed across the upper part of the grave, retaining the soil above. Air could circulate. This practice and the high salinity of the background soil would have helped in preservation.

Although apparently intact, leather and skin over time becomes increasingly inflexible and brittle as characteristically resilient collagen chains are severed in a process of progressive depolymerisation. Studies of the degree of denaturation are being undertaken by Differential Scanning Calorimetry (DSC) in collaboration with the University of Northampton, facilitated by Dr. Paula Antunes, PhD Senior Lecturer and Field and MSc Course Leader, Leather division. Results will be reported in due course.

Alongside (and often preceding) the increasing fragility of the skin products, deformation from the original shape occurs in situ due to local pressures. Changes in shape may happen swiftly if, for example, the grave superstructure collapses on to the body, or slowly in response to the action of gravity over time. In this case the cloaks resemble their original conformation, being supported internally by a wrapped body, and appear little damaged by soil fallen from above. However, decay usually happens very fast if in association with a putrefying body. Autolysis (where enzymes that colonise the gut break down the body post mortem) starts close to the gut and initially affects adjacent wrappings. Such decay is not always visible from above; indeed the cloaks may be in very poor condition where hidden beneath the body. Intriguingly, the human skin has in some areas greater strength and flexibility than the processed skin product.

155

ICOM-CC Leather and Related Materials WG Offenbach 2012

Conclusion

Although still at an early stage in the project, much information is emerging as the bodies are cleaned and mounted (Figure 9). More data will emerge and a number of potential studies are being put together which will lead to future projects. This initial presentation is offered to introduce others to some of the intriguing aspects of the two bodies and their wrappings; comment and further information are welcome.

Acknowledgments Thanks are due first to the Clothworkers Foundation for funding the project. Daniel Antoine has given crucial guidance throughout, and Derek Welsby, Julie Anderson and Elisabeth O'Connell of the Department of Ancient Egypt and Sudan have advised regarding archaeological and cultural aspects. Catherine Higgitt, Marei Hacke and Caroline Cartwright of the Department of Scientific Research and Conservation have provided valuable scientific support. Paula Antunes, Yvette Fletcher and Roy Thomson have been generous with time, facilities and leather-related advice. Sherry Doyal and David Saunders helped edit the paper.

Materials Cellite® 220 (Aluminium skin board with a lightweight honeycomb core): TRB Lightweight Structures, www.technicalresinbonders.co.uk, tel +44 (0)1480 447400. Tyvec® (nonwoven spunbond olefin fibre): Preservation Equipment Ltd, Vinces Road, Diss, Norfolk IP22 4HQ, UK, Tel: +44 (0)1379 647400, Email: [email protected] http://www.preservationequipment.com. Relic Wrap™ Packaging sheet (polytetrafluoroethylene): Gaylord Bros. International,PO Box 4901 Syracuse, NY 13221-4901, USA, [email protected], tel (001) 315-457-5070 ext. 8243 Plastazote (polyethylene foam): Ramplas Ltd, 84 Birmingham Road, Dudley, West Midlands DY1 4RJ, UK, tel +44 (0) 1384 453160, Email: [email protected]. http://www.ramplas.com/pages/ramsheet.asp. Thermocutter: Preservation Equipment Ltd, http://www.preservationequipment.com/Store/Products/Equipment-$4-Tools/Hand-Tools/Thermocutter

References Adams, W. 1993. « Medieval Nubia, Another Golden Age ». Expedition: Bulletin of the University Museum of the University of Pennsylvania, vol. 35, pp 29-39. Driel-Murray, C. van. 2002 « Practical Evaluation of a Field Test for the Identification of Ancient Vegetable Tanned Leathers ». Journal of Archaeological Science, vol. 29, Number 1, Academic Press pp. 17-21. Ginns, A. 2007. « Preliminary report on the second season of excavations conducted on Mis Island (AKSC) ». Sudan & Nubia, vol. 11, pp. 20-25.

Biography Barbara Wills works on the conservation of a wide range of organic materials and specialises in leather, human remains, basketry materials and ancient Egyptian objects. She is presently undertaking a two-year Senior Clothworkers Fellowship, cleaning and investigating a group of naturally-mummified human remains from Sudan. She lectures and publishes widely and enjoys passing on skills to others, contributing to workshops and University course units. Her interest in leather is reflected in long-term Archaeological Leather Group committee membership and as a Trustee of the Leather Conservation Centre. Recently she organised the BM conference ‘Going Green; towards sustainability in Conservation’ reflecting her concern for sustainable practice. Department of Conservation and Scientific Research - The British Museum - Great Russell Street - London WC1B 3DG, UK - [email protected]

156

ICOM-CC Leather and Related Materials WG Offenbach 2012

The Conservation of Archaeological Wet Leather at Schloss Gottorf During The Last 65 Years Gabriele Maria Zink

Abstract The Stiftung Schleswig-Holsteinische Landesmuseen Schloss Gottorf possesses a huge number of archaeological wet leather artefacts. As in many cases, there were very few reports in former times but since the late 1950s until today the conservators have documented their data. In the laboratories different techniques, which were modern in their times, were practiced: “Dégras”, PEG, cedar wood oil, “BLM”, “Neutralfett SSS” and “DLM 4060” (a mixture of fatliquors) were used to conserve archaeological wet leather finds. These substances were applied in baths, brushed on or even applied warm. The appearance of these leather artefacts in 2012 is mostly dark and fatty. The text seeks to give an overview of the different treatments. Using the example of the excavations Elisenhof, Haithabu/Settlement and Schleswig/Schild, the problems (dark, fatty and tacky surfaces, hydrophobicity, stiffness, fatty spue) caused by using fats and oils for conservation will be shown and discussed.

Keywords Archaeological wet leather, degreasing, fat, hydrophobicity, oil, retreatment, fatty spue (mineral soaps), stiffness.

Introduction

Due to the slightly acidic soil of Schleswig-Holstein, proteinacious materials like wool and leather are especially well preserved so the conservation of those materials has a long tradition in the archaeological museum of Schloss Gottorf. The first were excavated in the 19th century and, at times, the notes of archaeologists like Johanna Mestorf provide the only information about conservation treatments. In the first half of the 20th century, it seems that nobody noted anything – neither the archaeologists nor the conservators. Some conservators even kept their techniques secret. The documentation of conservation started in the 1950s when fundamental information was noted in tables. Over the decades the documentation did improve and today we work with a data-base which is used to organize information from the initial excavation until the conserved object is published. In this way, fundamental information does not have to be noted several times and everyone in the Institute can follow the object digitally.

The substantial lack of documentation, indecipherable handwriting and, due to a lack of education of conservators at different times, the use of substances and techniques borrowed from handcraft and industry are a big problem for conservators today. While retreating objects we often don‘t know exactly what substance was used for conservation, how this reacted with the original and consequently if and how this chemical/physical reaction can be resolved again. We have studied all sorts of scientific analyses and techniques, but in daily work there often is no money for analyses and the retreatment is just an approach to improve the visual appearance and the touch of an object, step by step. Both are very important senses for conservators but are actually subjective impressions and thus the opposite of “material-scientific” conservation.

Manual abilities and the knowledge of material science are the foundation of conservation. Learning a handcraft or a comparable long-time manual education, combined with studies in chemistry, physics and material science and the exercising of restoration techniques help us to avoid the mistakes of former times: using the techniques from handcraft and industry for conservation! Fats and oils may be suitable for manufacturing leather but they are not for the conservation of archaeological and historical leather objects. The interactions between fats, oils and archaeological leather have produced an urgent need for conservation research today as we do have to retreat those objects to conserve them in a material-scientific correct way.

157

ICOM-CC Leather and Related Materials WG Offenbach 2012

Description of the Problem

Systematic documentation of conservation treatments started in the 1950s when the main data were noted in tables. According to those short notes, the finds of “Elisenhof” were treated in the 1960s with Dégras (oxidized fish oil), Poly 200 and later on with Poly 600 (polyethyleneglycol or PEG) . Over fifty years later, in 2012, they seem to be less problematic as they are not so overtreated: stiff and brittle but light in colour and not hydrophobic. The reasons for the migration of the conservation substance into the wrapping material can´t be answered yet (Figure 1). Supposedly because of their stiffness and brittleness many of the objects were retreated in the 1980s with fats and oils and now they are dark, fatty, tacky, stiff, hydrophobic and partially with fatty spue (mineral soaps) on the surface (Figure 2).

Figure 1 (on the left): Loop from “Elisenhof”; according to the documentation it was treated with Dégras and Poly 200/600; one of the substances migrated over the decades and stained the paper wrapping material.

Figure 2 (on the right): One-piece shoe from “Elisenhof”; treated a second time in the 1980s with fats and oils. The leather is now dark, tacky, stiff and shows white fatty spue

In the 1960s, Prof. Dr. Schietzel started the first of his two large Viking age excavations in “Haithabu”. Upon the recommendation of an archaeologist, who specialized in leather, the first wet leather finds were conserved in the 1970s by pressing them between heavy glass panels and adding a fatty liquid which was used originally in the fur industry. These “first steps” in conservation show very well how urgently in those former times handcraft-based but also material-scientific educated conservators were needed alongside the archaeologists in the museums. They also show how - due to the lack of this kind of modern conservator - techniques for new materials were tragically copied for archaeological material. As Prof. Dr. Schietzel wasn´t satisfied with the results of dark, fatty and nonflexible leathers and being in charge of the conservation department at the same time, he wanted to enhance the techniques for conserving wet organic archaeological materials. Based on intensive contact, especially with North-European museums which were already more experienced in conserving those materials, he started with what was, at the time, a new type of treatment: cooling, freezing and using again polyethyleneglycols (PEG) - one of the first such treatment in Germany. Due to his experience with “Haithabu/Settlement” and its large number of finds, he started the excavation “Haithabu/Harbour” but not till rebuilding the conservation department which was finished in 1979: large water-basins, cooling room (0°C), freezing room (-20°C) and later in the 1980s the freeze-drying equipment. Leather finds from “Haithabu”, “Schleswig/Schild” and “Großer Schlichtenberg” were cooled and frozen – a technique which was called “Naßgefrieren” - and they look fine (Figure 3): light in colour, soft in touch and according to the notes of the archaeologists there was hardly any shrinkage. (Schnack C. 1992)

In the 1970s two more methods were practised. From the conservation point of view, due to the always problematic necessity in archaeology to show or to publish interesting finds promptly, a method faster than cooling and freezing had to be invented. Conservation with fats and oils, combined with drying by solvent is a common technique as it seems to be easy and fast. The author’s personal experiences with these methods 158

ICOM-CC Leather and Related Materials WG Offenbach 2012

are mostly negative and once more they were confirmed by some single finds from “Schleswig/Schild” which were done in cedar wood oil and carbon tetrachloride in order to conserve them more quickly. The idea of this technique is probably based on the British Museum leather dressing. (Waterer 1986) In 2012 these artefacts are dark, stiff and they have tacky residues on the surface (Figure 4).

Figure 3 (on the left): Upper from “Schleswig/Schild” (Schnack, Nr. 3353), cooled and frozen: light in colour, soft in touch and almost no shrinkage Figure 4 (on the right): Ornamented upper from “Schleswig/Schild” treated with cedar wood oil: changes in colour, tacky residues on the grain, stiff

The same principle is applied in the conservation method using Neutralfett SSS dissolved in toluene, published by Deutsches Ledermuseum Offenbach. Due to health concerns, the solvent was later modified to ethanol. (Schmitzer 1991.) The method gave good results even if the leathers still feel a little bit fatty. But Neutralfett SSS and its replacement Neutralfett QL are actually mixtures of different soaps with ammonium oleate as the main component. As fatty acid anions like to react with cations, the small proportion of saturated fatty acids like stearic acid is enough to form almost insoluble fatty spue (mineral soaps) which do harm leather so these two substances shouldn´t be used for conservation anymore. (Zink 2004.)

According to one of the few records available, at least one leather find from “Schleswig/Schild” was treated experimentally as described by Dr. Bruno Mühlethaler in “Kleines Handbuch der Konservierungstechnik” (Mühlethaler 1979.). Due to the minimal notes it cannot be determined which of Mühlethaler’s techniques was practised. Assuming that the recommended technique for archaeological leather was used, this find from “Schleswig/Schild” was heated in Polydiol (polyethyleneglycol) 400 and 1000 up to 40-50°C. The dark black colour and the total stiffness of the leather raises the fear that perhaps the proteinacious structure has been irreversibly destroyed. It is very fortunate that one part of the boot’s upper wasn´t treated like that. It shows how good the leather could look and hopefully analysis will help to explain the mechanisms leading to the two different appearances so that correct retreatment can give a better match between the two pieces of leather (Figure 5).

Figure 5: The two parts of one upper of a boot – obviously treated completely differently 159

ICOM-CC Leather and Related Materials WG Offenbach 2012

Retreatments

The problem with cooling and freezing – the so-called “Naßgefrieren” - is the resulting dryness and the brittleness of the leathers, as there is no bulking agent between the fibrils and consequently between the bundles of protein fibres allowing them to slide against each other. Supposedly due to this result, the majority of leather finds were retreated in the 1980s with fats and oils and the results are as bad as with the finds of “Elisenhof” (see Figure 2).

Those that had not been retreated in the 1980s are still too dry and too brittle like the finds of “Großer Schlichtenberg” which were published by Groenmann-van Waateringe, who sent them back to the author in 2007. To increase the sliding of the protein fibres and thus the flexibility and the humidity, the leathers were gently sprayed with PEG 600 (30% in deionized water) several times over the course of several days while being wrapped in a plastic bag. By this method, the aqueous PEG 600 had the time to penetrate the leather structure slowly and thoroughly and made the leather objects more flexible and even bigger. (Figure 6) Theoretically, this result can be explained by the filling of the interfaces between the bundles of protein fibres. Whether the PEG penetrated the leather objects really thoroughly and how far the PEG got in between the protein fibres, or even between the fibrils, can´t be assessed yet.

Even if the conservation treatment is just a “simple spraying”, we have to consider the material´s structure and nature of the material: the flesh side has an open fibrous structure while the grain side is dense due to its natural protective function for the creature. So – comparable to the so-called “Tafelschmiere” in tannery (stuffing the tanned leather with dubbin in the currying process) - the object should be sprayed from the flesh side to get a better migration into the leather. The migration of one substance into another is always affected by capillarity and this is dependent on moisture content. So as a pretreatment, a piece of leather should be sprayed with water or humidified in a climate chamber before applying the conservation substance on the flesh side.

The big disadvantage of these early procedures is the pressing with glass panels during freezing for better documentation; this result couldn´t be reversed. In particular, heel stiffeners were damaged irreversibly by this practice as the leathers were broken.

Figure 6: Sole from “Großer Schlichtenberg”; the inner line shows the size after cooling and freezing in the 1970s; the outer line shows the size after spraying with PEG 600 in 2007

160

ICOM-CC Leather and Related Materials WG Offenbach 2012

The retreatment of leathers which were conserved with fats and oils is based on degreasing by solvents. After spot-tests with different solvents (water, alcohols, ketones, ethers, aliphatic hydrocarbons, spirits) on hidden parts of the object, a bath in the appropriate solvent followed. Until now, 50% mixtures of deionized water and denatured ethanol were always adequate (Figure 7). Corresponding to the thickness and size of the object and to the amount of fats and oils, the bath can last one, two or three minutes. Afterwards the leather object is tightly wrapped in tissue and together they are then covered in a box filled with sand. (Figure 8) The purpose of the sand is to provide a 100% adaptation of leather and tissue surfaces to achieve an equal migration of the fats and oils into the tissue all over the object surface. Due to the author’s philosophy: “No more than necessary” this migration is limited to one day. After this 24 hours period in the sand, the fat content should have reached equilibrium and the use of a second or third bath can be calculated. Afterwards the leather objects are placed in boxes with saturated salt solutions to check their humidity absorption (flexibility, weight) as indicator for the degreasing. The basic objective is the isolation of all bundles of protein fibres or even of the elementary fibres with the minimum of bulking agent – here: fats and oils – but surely this is a quite theoretical aim as this can´t be investigated in the daily work. So – as written in the introduction – the end of the degreasing is only determined by look and feel and humidity absorption of the leather.

To avoid inserting more substances, normally no more is done – apart from reshaping and assembling if necessary. Just one shoe had to be sprayed after degreasing with PEG 600 to enable the reshaping for an exhibition as it was still totally stiff.

Figure 7 (left): Shoe from “Haithabu” placed in a red box filled up with a bath of 50% deionized water and denatured ethanol; the fat and oil content is leaching as white and yellow substances.

Figure 8 (right): The same shoe from “Haithabu” after the solvent bath and the 24 hours in sand: the fat and oil content migrated in the tissue as a brown imprint (layer sequence in the red box: sand-tissue-leather-tissue-sand)

Substances like cedar wood oil may present a problem as they are not soluble in water or ethanol. Ethyl acetate (acetic ether) gives very good results but it has to be said that this solvent is quite powerful! Used in a bath, it previously caused irreversible shrinkage so it is just used by swabbing the surface with a tissue moisturized with ethyl acetate (Figure 9).

Figure 9: The upper of a shoe from “Schleswig/Schild”; the leather´s grain is still tacky after the bath (adhering residues of tissue) so the surface is swabbed by ethyl acetate on a tissue (light and dim area in the middle of the leather). The small picture shows the difference between ethyl acetate (left) and denatured ethanol (right). 161

ICOM-CC Leather and Related Materials WG Offenbach 2012

Discussion and Current Approach

Newly excavated wet archaeological leather finds are also treated using the principle of “No more than necessary” but this means the following questions have to be answered first: To which degree should we clean leathers from metals, acids and salts ? How important is shrinkage to us ? Do we want our leathers to be light in colour ? Do leathers have to be supple ?

Leather is skin conserved by a tanning agent. Both are natural products and so in former times it was not 100% possible to standardize them. Two different tanning agents can give different colours and different rigidities for two pieces of the same type of leather. And consider the different taphonomic reactions in the different soils like sand, gravel, vamp or humid soil ! How can one think that one treatment will conserve all leathers the same way ? How can one think that one treatment will produce always the same light, supple and non-shrunk piece of leather ?

In former times the author’s approach was to be concerned with chemistry, but more and more now the author favours the approach of also considering physics. The question can´t only be “What do we put in ?” but also “How do we put this in” ? In conservation, we often use a liquid which we apply in a bath and after dehydration (air-drying, solvent-drying, freeze-drying) we measure the shrinkage to judge the conservation treatment.

If you do tanning you have several steps called “handling, suspending and laying away” (“Farbengang, Versenk, Gerbung”). You start with low concentrations of tanning liquor and gradually increase the concentration of tannins to achieve a thorough tanning everywhere in the corium. Shouldn´t we think much more about fibrils, fibres and fibre bundles, the spaces in between and how to reach them instead of measuring the shrinkage ? Surely, it is a cheap and easy method but even if you use 3D-scannings, which is an exact method, it can´t tell us more than tendencies as shrinkage is the sum of a lot of different aspects.

And we should stop ignoring the principles of natural sciences. As soon as objects are being excavated, oxygen and light activate the degradation of the materials. Leathers of a ship-wreck, excavated in 2011, arrived in the laboratory in 2012. In the meantime, the iron oxidized on the uncovered surface as it was not possible to wrap them excluding oxygen (Figure 10).

The covered bottom is still black as shown on the turned right corner. Actually iron stains could be reduced by chelating agents but the use of these is not ”non-toxic”: in addition to all the metals arising from the taphonomic situation (iron, calcium, etc.), these chelating agents also complex tanning agents! So if leathers have to be treated for a long time to complex iron stains they also get “de- tanned”. Thus the leathers are harmed twice: first the oxidation takes place and then the de-tanning follows. In situations like these, conservation is once again an approach which tragically could have been preventable by correct “first aid”. Figure 10: Piece of leather from a ship-wreck, excavated in 2011; the iron-content oxidized at the surface and harmed the leather structure

Göpfrich´s warning about treatment with Komplexon III / Titriplex III was published already in 1986: “Ein längeres Einlegen erwies sich als nicht ratsam, da die Lederfaser zu stark quillt.” (Longer exposure times proved to be not advisable, since the leather fibre swells too much (Göpfrich 1986.). It simply describes this complicated balance between conserving the necessary tanning agent and destroying iron-contamination.

162

ICOM-CC Leather and Related Materials WG Offenbach 2012

One of the latest works about this problem was carried out by Löwe who tried to test the efficiency of EDTA, citric acid and oxalic acid and the use of different temperatures, pH values and ascorbic acid as reducing agent which is also used in textile conservation. (Löwe 2005.) She also couldn´t give clear advices as this is a fundamental question and a problem. In situations like these, conservators still have to find the happy medium.

Thus, one recommendation for the first important conservation treatment for freshly excavated leathers has to be the immediate wrapping in oxygen-free packages as far as possible and transfer to the laboratories as fast as possible! A cheap method is packing the leathers in zip-lock bags made of PE/PP with a small amount of water in it and closing this bag laying slightly sloped below water surface in a water-basin while pressing most of the air out of the bag! This way, most of the oxygen is removed and the leathers are still water-saturated.

Materials Acetone, Ethyl acetate (acetic ether), Carbon Tetrachloride : solvent “BLM” : supposedly a mistake of handwriting, it should be written “BML”: “British Museum Leather Dressing” Lanolin (anhydrous) 200g, Beeswax 15 g, Cedarwood Oil 30 ml, Hexane 350 ml. Cedar wood oil : Juniperus virginiana, 80% Cedren, 12% Cedrol. Dégras : oxidized fish oil “DLM“ : “Deutsches Ledermuseum“ “DLM-Lösung Nr. 1101“ : 500 ml ethanol, denatured (instead of toluene), 250 g Neutralfett SSS. “DLM-Lösung 4060“ : 20% Lipoderm-Licker SA, 10% Lipoderm-Licker N, 10% Karion F, 60% water.

Komplexon III /Titriplex III : chelating agent, Na2EDTA Lanolin : “Adeps lanae“ wool wax respectively “Lanolinum“ (65% wool wax, 15% paraffin oil, 20% water). Neutralfett SSS or QL : Ammonium oleate by Schill & Seilacher Poly, Polydiol : Polyethylenglycol of different molecular weight (200; 400; 600; 1000)

References Göpfrich J. 1986. «Römische Schuhfunde aus Mainz», Saalburg-Jahrbuch 42, Mainz 1986, p. 8 Löwe J. 2005. «Archäologisches Feuchtleder. Komplexbildner zur Behandlung von Ab- und Einlagerung aus Eisenverbindungen. Archaeological wet leather. Complexing agents for treatment of deposits and encrustations of iron compounds», unpublished Diploma-thesis, University of Applied Science, Berlin, 2005 Mühlethaler B. 1979. «Kleines Handbuch der Konservierungstechnik», Bern/Stuttgart, pp. 111-113 Schmitzer W. 1991. «Lederrestaurierung - Tips für Sammler», Offenbach, pp. 9 Schnack C. 1992. «Die mittelalterlichen Schuhe aus Schleswig, Ausgrabung Schild 1971-1975», Ausgrabungen in Schleswig, Berichte und Studien 10, Neumünster, pp. S. 21-22 Waterer J. W. 1986. «Guide to leather conservation and restoration», Northampton, pp. 12-13, 43 Zink G. 2004. «Conservation of archaeological wet leather - the problem of calcium-soap formation and its removal», Proceedings of the 9th ICOM Group on Wet organic Archaeological Materials Conference, Copenhagen 7-11 June 2004, Bremerhaven, pp. 591-611

Biography Gabriele Zink is a conservator for archaeological objects specialized on wet organic archaeological materials. She works since 2005 for the Stiftung Schleswig-Holsteinische Landesmuseen Schloss Gottorf in Schleswig (North-Germany). Her education started with the school for textiles, later she became a goldsmith and after almost three years of working for museums and excavations in Germany and Europe she started her studies in Berlin where she finished with a diploma about archaeological leather conservation. Stiftung Schleswig-Holsteiische Landesmuseen - Schloss Gottorf - 24837 Schleswig, Germany - [email protected] - [email protected] - 04621-813 504

ICOM-CC Leather and Related Materials WG Offenbach 2012

POSTER - Charisma Projects on Gilt Leather

Céline Bonnot-Diconne, Marcella Ioele, Claire Pacheco, Mariabianca Paris, Laurianne Robinet

164

ICOM-CC Leather and Related Materials WG Offenbach 2012

POSTER - Les Collections Ethnographiques, Spécificités dans la Conservation

Marina Regni

165

ICOM-CC Leather and Related Materials WG Offenbach 2012

Participants List

Name Professional Affiliation City Country Email Address

Arens Thomas Staatliche Museen / Kunstgewerbemuseum Berlin Germany [email protected]

Beaumont Ian Ian Beaumont Leather Conservation Northampton Great Britain [email protected]

Blaschke Kristina University Library Bern Switzerland [email protected]

Blasum Gertrude Museum für Völkerkunde Hamburg Germany [email protected]

Blöcher Heidi Kunstgewerbemuseum Berlin Germany [email protected]

Böhm Ingrid Naturhistorisches Museum Nürnberg Germany [email protected]

Bonnot-Diconne Académie de France à Rome – Villa Médicis / Rome / Italy / France [email protected] Céline 2CRC Moirans

Bruns Fabian Norddeutsches Zentrum für Materialkunde von Hannover Germany [email protected] Kulturgut e.V.

Calnan National Trust Suffolk Great Britain [email protected] Christopher

Drover Louise L. Drover Conservation Ltd. Northampton Great Britain [email protected]

Eggert Gerhard State Academy of Art and Design Stuttgart Germany [email protected] Prof. Dr.

Einsiedl Renate Universalmuseum Joanneum, Museum im Palais Graz Austria [email protected]

Eska Margarete State Academy of Art and Design Stuttgart Germany [email protected]

Fahrner Diana Archäologie und Museum BL Basel Switzerland [email protected]

Fischer Antje Staatliche Kunstsammlungen, Rüstkammer Dresden Germany [email protected]

Fischer Rainer Musée national d'histoire et d'art Bertrange Luxembourg [email protected]

Fournet Jean- Art Historian Paris France [email protected] Pierre

Frankenhauser Deutsches Ledermuseum/Schuhmuseum Offenbach Germany [email protected] Nina

Gervais Angelika Norddeutsches Zentrum für Materialkunde von Hannover Germany [email protected] Kulturgut e.V.

Göckeritz Freelance Textile Conservator Bern Switzerland [email protected] Stefanie

Göpfrich Jutta Deutsches Ledermuseum/Schuhmuseum Offenbach Germany [email protected]

Gottsmann Reiss - Engelhorn Museen/Museum Weltkulturen Mannheim Germany [email protected] Sandra

Harand Birgit Historisches Museum Frankfurt Germany [email protected]

Harwart Vivian Paramentenwerkstatt der von Veltheim-Stiftung Helmstedt Germany [email protected]

Hassel Barbara Buch- und Graphikrestaurierung Frankfurt Germany [email protected]

Iafrate Sara Freelance Conservator Roma Italy [email protected]

166

ICOM-CC Leather and Related Materials WG Offenbach 2012

Jensen Eva Lilja National Museum of Denmark Kongens Denmark [email protected] Lyngby

Jervis Anna Istituto Superiore per la Conservazione e il Roma Italy [email protected] Valeria Restauro

Jung Michael Museo Nazionale d’arte orientale Roma Italy

Karlsdotter Göteborgs Universitet Olofstorp Sweden [email protected] Rebecka

Keller Isabel State Academy of Art and Design Stuttgart Germany [email protected]

Kilchhofer Upholstery Conservator Bern Switzerland [email protected] Nadine

Kingham Emilia The British Museum London Great Britain [email protected]

Kobbe Anke Arkeologisk Museum, Universitetet i Stavanger Stavanger Norway [email protected]

Koldeweij Eloy Cultural Heritage Agency Amersfoort Netherlands [email protected]

Kraemer Herzogin Anna Amalia Bibliothek Weimar Germany [email protected] Johanna

Kress Petra Germanisches Nationalmuseum Nürnberg Germany [email protected]

Kunze Friederike Museumssenteret I Hordaland – Salhus Norway [email protected] Bevaringstenestene

Larsen René Dr. The Royal Danish Academy, School of Kopenhagen Denmark [email protected] Conservation

Lehmann Monika Niedersächsisches Landesamt für Denkmalpflege Hannover Germany [email protected]

Ljubic Tobisch Technisches Museum Wien Austria [email protected] Valentina

Mackert Freelance Conservator, Textile and Leather Bonn Germany [email protected] Katharina

Mannina Restorer of Cultural Heritage Palermo Italy [email protected] Loredana

Mehlis Mareike State Academy of Art and Design Stuttgart Germany [email protected]

Moog E.Gerhard Reutlingen Germany [email protected]

Middleton Angela English Heritage Portsmouth Great Britain angela.middleton@english- heritage.org.uk

Moretti Federica Freelance Conservator Roma Italy [email protected]

Moroz Ryszard LWL- Museumsamt für Westfalen Münster Germany [email protected] Dr.

Mühlen Axelsson School of Conservation Kopenhagen Denmark [email protected] Kathleen

Müller Sonja Müller & Schweizer GbR Waiblingen Germany [email protected]

Müller-Radloff Staatliche Kunstsammlungen, Museum für Dresden Germany [email protected] Christine Völkerkunde

Pino Neubohn ARA-Italia Venezia Italy [email protected] 167

ICOM-CC Leather and Related Materials WG Offenbach 2012

Sibylle

Paris Istituto Superiore per la Conservazione e il Roma Italy [email protected] Mariabianca Restauro

Petzold Laura Klassik Stiftung Weimar Germany [email protected]

Posthuma de University of Amsterdam Amsterdam Netherlands [email protected] Boer Martine

Rabin Ira Dr. Bundesanstalt für Materialprüfung (BAM) Berlin Germany [email protected]

Regni Marina Freelance Conservator - - [email protected] Lucia

Reichel Brigitte Kulturhistorisches Museum Rostock Germany [email protected]

Riester Melanie State Academy of Art and Design Stuttgart Germany [email protected]

Schmidt Anne National Museum of Denmark Kongens Denmark [email protected] Lisbeth Lyngby

Scholz Ines Historisches Museum Frankfurt Germany [email protected]

Schubert München Germany Susanne Freelance Conservator [email protected]

Schulze Andreas Landesamt für Denkmalpflege Sachsen Dresden Germany [email protected] Dr.

Schweizer Robert Müller & Schweizer GbR Waiblingen Germany [email protected]

Selm Rosemarie Konservierungsatelier Preetz Germany [email protected]

Siennicki Martin Akademie der Bildenden Künste Wien Austria [email protected]

Stoltz Marc Hermès Paris France [email protected]

Sturge Theo Sturge Conservation Studio Northampton Great Britain [email protected]

Swann June Formerly at Northampton Museum, now Northampton Great Britain [email protected] independent consultant

Thomson Roy Consultant Peterborough Great Britain [email protected]

Trommer Research Institute of Leather and Plastic Sheeting Freiberg / Germany [email protected] Bernhard Dr. (FILK) Sachsen

Vest Marie The Royal Library Kopenhagen Denmark [email protected]

Vogel Regine LVR-Landesmuseum Bonn Germany [email protected]

Wailliez Wivine Royal Institute for Cultural Heritage Bruxelles Belgium [email protected]

Wankova University of Pardubice, Faculty of Restoration in Koprivnice Czech [email protected] Veronika Litomysl Republic

Weisser Ilka Naturhistorische Gesellschaft Nürnberg Germany

Wills Barbara The British Museum London Great Britain [email protected]

Ziegler Julia State Academy of Art and Design Stuttgart Germany [email protected]

Zimmer Judith Deutsches Historisches Museum Berlin Germany [email protected]

Zink Gabriele Stiftung Schleswig Holsteinische Landesmuseen Schleswig Germany [email protected] 168

ICOM-CC Leather and Related Materials WG Offenbach 2012

169

ICOM-CC Leather and Related Materials WG Offenbach 2012

This book contains the papers presented at the 10th Interim Meeting of the ICOM-CC Leather & Related Materials Working Group which was held in Offenbach (Germany) from the 29th to the 31st of August 2012. The conference was organised at the Deutsches Ledermuseum Shoemuseum and brought together more than 85 participants representing 14 countries.

Ces actes regroupent les articles présentés à la 10ème Réunion Intermédiaire du Groupe de Travail Cuir et Matériaux Associés de l’ICOM-CC qui s’est tenue à Offenbach (Allemagne) du 29 au 31 Août 2012. La conférence était organisée au Musée Allemand du Cuir et de la Chaussure. Elle a réuni plus de 85 participants de 14 pays différents.

170