Léo Marchand's ontlijming treatment

(1)

L

ÉO

M

ARCHAND

’

S

O

NTLIJMING

T

REATMENT

AN INVESTIGATION OF A SIZE REMOVAL METHOD

PERFORMED BY LÉO MARCHAND BETWEEN CA. 1950-1990

Jolijn Schilder

Student nr. 10759654 Supervisor: Dr. Emilie Froment Second reader: Dr. Herman den Otter

C&R UvA professor: prof. dr. Maarten van Bommel C&R UvA professor: prof. dr. Ella Hendriks

University of Amsterdam

(2)

Preface and acknowledgements

This thesis was written for the Conservation and Restoration of Cultural Heritage – paintings specialisation masters programme at the University of Amsterdam. The research would not have been possible without the help of a large number of people, to whom I would like to express my gratitude.

First, I would like to thank Emilie Froment, my thesis supervisor, for her support, discussions, feedback and advise, and It cannot be stressed enough how grateful I am to be given the opportunity to work with her reconstructions. I also thank the Dordrechts Museum for the opportunity to treat Adriaen Beeldemaker’s Dogs

in a Landscape, a painting from their collection. At the Dordrechts Museum, special thanks are due to conservator

Lidwien Speleers for providing the conservation reports of the painting and many other treatment reports by Léo Marchand.

I owe much gratitude to Pauline Marchand for answering endless lists of questions, giving many explanations, descriptions, providing me with photographs and last but not least, her continuous support. I thank Jos Deuss for the invaluable conversation, during which he detailed recalled memories on Marchand’s practice. At the Rijksdienst voor Cultueel Erfgoed (RCE), special thanks are due to Sanne Berbers, with whom I discussed several options for technical analysis, and especially Saskia Smulders-De Jong for performing pyrolysis-Gass Chromatography Mass Spectrometry (py-GC/MS) and interpreting the results. I would further like to thank Bart Ankersmit (RCE) for the use of the thermal camera.

I thank Esther van Duijn (Rijksmuseum) for her thoughts and information on the wax-resin lining practice of Henricus Hubertus Mertens and his contemporaries at the Rijksmuseum, as well as for providing me with helpful articles and documents.

Furthermore, I would like to thank Wil Werkhoven for the conversation on Henk Dik’s practice, and Nico van Bohemen jr. for responding to my questions.

I would like to express my gratitude towards the lecturers at the University of Amsterdam; Maartje Stols-Witlox and Vera Blok, and René Peschar. Furthermore, I thank the professors Ella Hendriks and Maarten van Bommel for their insight and advise.

Last nut not least, many thanks to my fellow students Sofie Dubbeldam, Steinunn Hardardottir, Ivana Jerdonekova and Christa Molenaar for their support and company!

(5)

Summary

This master thesis focusses on the lining practise of Léo Marchand (1913 – 1996), a paintings conservator active in the Netherlands between ca. 1950-1990. Marchand wax-resin lined many paintings throughout his career, amongst which were Adriaen Beeldemaker’s Honden in een Landschap I & II (1697), two paintings from the Dordrechts Museum. Upon establishing their conservation history it became apparent that Marchand routinely cleaned the reverses of paintings before lining. His aim was to dissolve the size layer of the original canvas in order to enhance the penetration of the wax-resin adhesive. To do so, the painting’s reverse was soaked with an aqueous mixture composed of water, ammonia and denatured alcohol, after which it was thoroughly brushed. Léo Marchand considered this treatment as a preparatory step and referred to the specific method of removing the size layer as ‘ontlijming’.

Since Marchand subjected all unlined canvasses and previously glue-paste lined canvasses to the

ontlijming treatment, this research can be related to a large group of objects. A better understanding of the

effects of the treatment may contribute to the diagnostic phases of conservation treatments and leads to the main question; ‘What are the effects of Léo Marchand’s ‘ontlijming’ treatment on reconstructions of 17th-century Netherlandish canvas paintings and how does the treatment influence the distribution of wax-resin?’

After the introductory chapter, a detailed description of Léo Marchand’s lining practise is given. Secondly, the current knowledge on the influence of size layers on the distribution of wax-resin is given, primarily based on literature. Following, ontlijming is performed on a set of samples to establish its effects on preparatory layers. The effects were assessed by means of pyrolysis- Gass Chromatography Mass Spectrometry analysis of extractions in combination with visual observations. Lastly, the influence of ontlijming on the distribution of wax-resin is evaluated by wax-wax-resin impregnation of a set of samples, half of them subjected to ontlijming.

This research revealed a variety of physical alterations on reconstructions of 17th_{century ground layers}

caused by ontlijming, including a (partly) removal of the size layer, an increased stiffness of the canvas support, delamination of ground layers, embrittlement and loss of binding medium and pigment. However, some of these effects were immediately resolved by the wax-resin impregnation afterwards. Lastly, the size removal treatment did not show major differences in the discolouration and spreading of discolouration on oil-bound single and double layers after wax-resin impregnation, although its effects on glue-based (double) ground layers requires further research.

(6)

Samenvatting

Deze masterscriptie focust zich op de bedoekingsmethode van Léo Marchand (1913 – 1996), een schilderijenrestaurator die tussen ca. 1950-1990 werkzaam was in Dordrecht, Nederland. Marchand heeft gedurende zijn carrière vele schilderijen bedoekt, waaronder Adriaen Beeldemakers Honden in een Landschap I

& II (1697), twee schilderijen uit de collectie van het Dordrechts Museum. Tijdens een onderzoek naar hun

restauratiegeschiedenis kwam aan het licht dat Marchand routinematig de achterkanten van schilderijen schoonmaakte voordat deze werden bedoekt. Zijn doel was om de voorlijming van het originele doek op te lossen en hiermee de doordringing van het was-harsmengsel te verbeteren. Hiervoor maakte hij de achterzijde van het originele doek nat met een mengsel van water, ammonia en spiritus, waarna hij stevig borstelde. Léo Marchand beschouwde deze behandeling als een voorbereidende handeling en verwees ernaar als ‘ontlijmen’.

Aangezien Marchand alle onbedoekte en voorheen lijmbedoekte schilderijen ontlijmde kan dit onderzoek worden gerelateerd aan een grote groep objecten. Een inzicht in de gevolgen van ontlijming kan bijdragen aan de diagnostische fase van een restauratiebehandeling of –onderzoek. Dit leidt tot de volgende hoofdvraag: ‘wat zijn de effecten van Léo Marchand’s ontlijming op reconstructies van 17e-eeuwse schilderijen op doek en hoe beïnvloedt deze behandeling de was-hars impregnatie?’

Na het inleidende hoofdstuk wordt een gedetailleerde beschrijving van Marchand’s bedoekingsmethode gegeven. Daaropvolgend wordt de huidige stand van de wetenschap over de invloed van lijmlagen op de doordringing van was-hars uiteengezet. Vervolgens zijn meerdere reconstructies ontlijmd om het effect op grondlagen te beoordelen. Dit is onderzocht door middel van pyrolyse-gaschromatografie massaspectrometrie (py-GC/MS) van monsters, en visuele waarnemingen. Als laatste is de invloed van het ontlijmen op de doordringing van was-hars onderzocht door meerdere reconstructies te impregneren met was-hars; de helft van de reconstructies zijn geimpregneerd in hun oorspronkelijke staat, de andere helft na ontlijming.

Dit onderzoek heeft een aantal materiële veranderingen in reconstructies van 17e-eeuwse schilderijen op doek blootgelegd die werden veroorzaakt door het ontlijmen, waaronder een gedeeltelijke verwijdering van de voorlijming, een toegenomen stijfheid van het doek, delaminatie van grondlagen, verbrossing en het verlies van bindmiddel en pigment. Een aantal van deze gevolgen werden door impregnatie met was-hars direct verholpen, anderen hebben mogelijk gevolgen voor toekomstige behandelingen. Tenslotte liet de ontlijming geen grote verschillen zien in de verkleuring en verspreiding van verkleuring op oliegebonden enkele- en dubbele grondlagen na was-hars impregnatie, hoewel de effecten op lijmgebonden grondlagen verder onderzoek behoeft.

(7)

1. Introduction

Many canvas paintings in Dutch collections have been subject to wax-resin relining, a treatment in which an additional canvas is adhered to the original support with a mixture of wax and resin. Wax-resin lining, also known as “The Dutch Method”, was introduced in the second half of the nineteenth century in the Netherlands and has long been, sometimes still is, used. Wax-resin linings served multiple functions; flaking paint was consolidated, the paint layers were flattened and an additional support was adhered.1_{However, there has been a sharp decline}

in the application of wax-resin linings from the 1970’s onwards, due to raised concerns about the method’s drawbacks, the controversial effects and changes in conservation ethics. Wax-resin lining always involves impregnation of the original painting with a wax-resin adhesive.

This thesis describes the research carried out on the wax-resin lining practise of Leopold François Joseph (Léo) Marchand (1913 – 1996), a paintings conservator active in the Netherlands between ca. 1950-1990. Marchand ran a private practise in Dordrecht and from the 1970’s onwards he worked part-time for the Dordrechts Museum. Marchand routinely cleaned the reverses of paintings thoroughly before lining. His aim was to dissolve the size, usually an animal glue layer, of the original canvas in order to enhance the penetration of the wax-resin adhesive. To do so, he added ‘een scheut spititus en een scheut ammonia’ (a dash of denatured alcohol and a dash of ammonia) to a small bucket of lukewarm water. Marchand soaked the reverse of the canvas with the water-ammonia-alcohol mixture and started to brush thoroughly. As a result, foam appeared on the wetted canvas and Marchand believed this to be the animal glue. The brown foam was scooped from the canvas, which was rinsed with plain water afterwards and then let to dry. Léo Marchand considered this treatment as a preparatory step and referred to the specific method of removing the size layer as ‘ontlijming’. Léo Marchand’s lining practise, and especially the ontlijming method that he used, are the topics of research.

1.1. Background

In the context of the Conservation and Restoration of Cultural Heritage master programme – paintings specialisation at the University of Amsterdam the paintings Honden in een landschap I & II (1697) (Dogs in a Landscape I & II) by Adriaen Beeldemaker were treated by Ivana Jerdoneková and the author between 2017 – 2019 [figure 1& 2]. The seventeenth century pendants are owned by the Dordrechts Museum, who acquired them in 1970. In the following year, both paintings received conservation treatment by Léo Marchand. During the current restoration process of both paintings Pauline Marchand, Léo’s daughter, was consulted. She is professionally active as a paintings restorer in Rotterdam and stores her father’s documentation. Pauline

1_{Esther van Duijn, “Vader en zoon Hopman; een tijdsbeeld aan de hand van twee 19}de_{-eeuwse restauratoren”, Cr 3}

(8)

explained more about Léo Marchand’s lining method, which was when it became apparent that he performed the so-called ‘ontlijming’ treatment.2

Fig. 1. Adriaen Beeldemaker, Honden in een landschap, 1697. Oil on canvas. 45 x 49,5 x 2,5 cm (excluding frame). Dordrecht, Dordrechts Museum. DM/970/465. Photograph: Ivana Jerdoneková.

2_{The explanation was recorded, a transcription can be found in Appendix I, 1.1. Private conversation of author with}

(9)

Fig. 2. Adriaen Beeldemaker, Honden in een landschap, 1697. Oil on canvas. 45 x 49 x 2,5 cm (excluding frame). Dordrecht, Dordrechts Museum. DM/970/466.

After conversations with Pauline Marchand and Jos Deuss, a restorer who was trained by and worked with Léo Marchand, it became clear that before wax-resin lining paintings Léo Marchand subjected all unlined canvasses to ontlijming, as well as previously glue-paste lined canvasses which had to be wax-resin relined.3,4

Since the Dogs in a Landscape paintings were unlined before the 1971 treatment by Marchand, it was with great certainty assumed that they received ontlijming. Although a surface phenomenon referred to as blanching is investigated by Sofie Dubbeldam between January and June 2019, the paintings do not show specific phenomena

3_{The glue-paste lining would be removed first, after which residues of the adhesive were removed by the so-called}

‘checkerboard’ method. Marchand swelled the glue with a wet sponge in a checkerboard pattern and removed the residues mechanically with a scalpel. It was only after the removal of the glue-paste residues that the ontlijming treatment with the water-ammonia-denatured alcohol mixture took place.

4_{Private conversation of the author with Jos Deuss, 15/02/2019; and private conversation of the author with Pauline}

(10)

that question the effect of the ontlijming on their present state.5_{However, given the description of the method}

it was hard to believe that it did not affect the paintings at all.

Léo Marchand himself aimed to remove the size layer, which would have resulted in a more even distribution of the lining adhesive. The distribution of wax-resin, assessed by discolouration of the ground layer(s), may not be as apparent on the surfaces of paintings as it is in the structures underneath. Moreover, it must be taken into account that phenomena associated with ontlijming may show over time but are not present yet, or that they are not recognized as such. Since it is known that Marchand subjected all unlined canvasses to the size removal treatment, the subject did not allow a comparison between objects that have and have not been ontlijmd. The only paintings of which Marchand did not try to remove their size layer were the canvasses that he did not wax-resin line at all. The research described in this thesis therefore originates from two objects that have been ontlijmd, Dogs in a Landscape I & II, but mostly relies on experiments performed on reconstructions rather than analyses of the objects themselves.

1.2. Research question

Marchand aimed to improve the distribution of wax-resin by washing out the size layer of the original canvas. Whether this procedure indeed dissolves the size layer and if so, how this influences the distribution of wax-resin are the main points of research. Based on the objects from which the subject arose and previous research into the consequences of wax-resin lining, it was decided to focus on seventeenth century paintings and their build-up. The following research question was formulated:

‘What are the effects of Léo Marchand’s ‘ontlijming’ treatment on reconstructions of 17th_-century

Netherlandish canvas paintings and how does the treatment influence the distribution of wax-resin?’

In literature on the influence of size layers on the penetration of wax-resin, the evenness with which the wax-resin adhesive darkened the ground layers was discussed shortly in ‘Colour Change in Sample

Reconstructions of Vincent van Gogh's Grounds due to Wax-Resin Lining’ by Emily Nieder et al. (2011), although

it focussed on nineteenth century paintings by Van Gogh. The darkening of ground layers due to the impregnation with wax-resin has extensively been studied in Emilie Froment’s PhD thesis ‘The consequences of wax-resin

linings’ (2019). However, the influence of size layers on the penetration of the lining adhesive in general and the

possible discolouration associated with the impregnation are beyond the scope of the research. This thesis

5_{The painting Honden in een Landschap I shows white stripes in the sky area. There is no evidence that these are}

(11)

research primarily focusses on the investigation of materials and techniques used for the ontlijming treatment as well as on the impact of the treatment on the condition of paintings and its consequences for the distribution of wax-resin after lining.

1.3. Current knowledge

Over the past decades, wax-resin lining treatments gained interest within the field of conservation. An important moment was The Conference on Comparative Lining Techniques which was held at the National Maritime Museum in Greenwich, England in 1974. It was the first opportunity for conservators to discuss wax-resin lining and its shortcomings as well as the underlying principles of structural treatment.6_{As described by Mireille te}

Marvelde almost 40 years after the conference, one of the conclusions was that traditional lining methods, ‘especially the wax-resin lining method, in the past had caused dramatic changes to paintings. It was also concluded that on a large scale, and often without real need, relining had been carried out systematically […].’7

Already in the 1970’s Vishwa Mehra argued that lining treatments should be reversible, and reversibility and

minimal intervention became key terms in general.8_{A new attitude towards conservation emerged. The papers}

of the Greenwich conference were published in 2003 and include relevant information for this Master thesis research, especially Georges Messens’s article ‘hand lining with wax-resin using an iron’ (2003) is of great interest because the step-by-step description of Messens’s lining practise comes close to that of Marchand.9_{After the}

Greenwich conference an interest in the history of lining techniques arose, resulting in Westby Percival-Prescott’s paper ‘The Lining Cycle: Causes of Physical Deterioration in Oil Paintings on Canvas: Lining from the 17th century

to the present day’ (2003).

Although general articles on wax-resin lining practises provide basic knowledge and can serve as a starting point, the research into Léo Marchand’s ontlijming treatment requires more specific information on his practice. The wax-resin lining practices of Dutch paintings conservators have been studied to some extent.10

Most recently an overview of the practices of several paintings conservators that worked in the Netherlands,

6_{Stephen Hackney et al, “Lining easel paintings”, in The Conservation of Easel Paintings, eds. Joyce Hill Stoner and}

Rebecca Rushfield (Abingdon-on-Thames: Routledge, 2012), 415.

7_{Mireille te Marvelde, “From Roger Marijnissen’s Historical overview to the present day: some thoughts concerning}

half a century’s development of Conservation History as a discipline”, CeROArt, accessed 18-02-2019, http://journals.openedition.org/ceroart/4623 , section 17.

8_{Hackney et al., “lining easel paintings”, 415.}

9_{Georges Messens, “hand lining with wax-resin using an iron”, in Lining paintings: papers from the Greenwich}

Conference on Comparative Lining Techniques, edited by Caroline Villers (London: Archetype publications, 2003), 70

– 76.

10_{Over the past twenty years, multiple studies of Dutch conservators and their practises have contributed to a broad}

knowledge on the history of conservation practise. Some of these studies include the description of wax-resin lining practises, possible effects and drawbacks. Particularly significant are contributions by Mireille te Marvelde, Esther van Duijn, Ella Hendriks, Emilie Froment and Saskia van Oudheusden.

(12)

including Léo Marchand, was given by Froment in ‘The consequences of wax-resin linings’ (2019).11_{Apart from}

Froment’s PhD thesis, there is a lack of published literature that makes mention of Marchand. However, Jos Deuss wrote down some of Marchand’s treatments, materials and techniques.12_{Another important source of}

information is Pauline Marchand, Léo Marchand’s daughter and paintings conservator herself. The author had several conversations with Pauline, of which the explanation of Léo Marchand’s lining practise was recorded and transcribed.13_{Lastly, many treatment reports of paintings that were restored by Léo Marchand between ca.}

1965-1990 are available. Although Marchand himself did not elaborate on his ontlijming treatment, he did mention the ontlijming in several treatment reports.

With regards to the effect of size layers on the penetration of resin and discolouration due to wax-resin impregnation, there are two main sources in which those topics are described: ‘Colour Change in Sample

Reconstructions of Vincent van Gogh's Grounds due to Wax-Resin Lining’ by Nieder et al. and Froment’s thesis

‘The consequences of wax-resin linings’, which were mentioned before. Nieder et al. found that the darkening of samples with a glue or emulsion bound ground was the greatest in samples without a size layer and least when a gelled size had been applied.14_{For samples with oil bound grounds the opposite was true: samples with a gelled}

size darkened the most whereas the samples without size darkened least. It was for all samples argued that the size layer provided a barrier against the wax-resin, which is why less of the adhesive was absorbed into the ground, also described as a less complete impregnation. Interesting observations were made regarding the topography of the surface; the surfaces of chalk in oil grounds without size darkened evenly, opposite to the chalk in oil ground with size for which only the weave tops of the canvas darkened. The topography indicated that samples without size were more completely impregnated with wax-resin.15_{However, it is worth noting that}

the samples shown in the article measure approx. 3,25 x 3,25 cm, which may be too small to draw conclusions on the distribution of wax-resin. In Froment’s ‘The consequences of wax-resin linings’ (2019) the influence of the size layer on the colour change of linen canvas is described. The application of a size layer itself darkens the linen canvas, but the difference in discolouration between bare canvas and sized canvas after wax-resin impregnation is only small.16_{Also, an interesting observation on the penetration of the wax-resin adhesive is made; it was seen}

that the adhesive penetrated through the glue bound grounds, but not through the oil bound grounds although

11_{Emilie Froment, “The consequences of wax-resin linings for the present appearance and conservation of}

seventeenth century Netherlandish paintings on canvas” (PhD diss., University of Amsterdam, 2019), 81-100.

12_{Jos Deuss, Schriftelijk interview Jos Deuss, 1998-1999. See Appendix I, 1.2.} 13_{For the transcription, see Appendix I, 1.1.}

14_{Emily Nieder et al., “Colour Change in Sample Reconstructions of Vincent van Gogh's Grounds due to Wax-Resin}

Lining,” Studies in Conservation, Vol. 56, No. 2 (2011), 98.

15_{Ibid, 99-100.}

(13)

both types were similarly applied on a sized canvas. This indicated that penetration of the adhesive may depend on the binding media of ground layers.

In sum, a considerable amount of literature on conservation history and articles on the wax-resin lining practises of Dutch paintings restorers are available, but information on Léo Marchand’s practise in specific is lacking. There is one unpublished written document by Jos Deuss in which Marchand’s materials and techniques are described. Moreover, an important source of information is Pauline Marchand, who worked with her father for a couple of years and preserves a box full documentation as well as materials used by Léo Marchand. The thesis by Froment (2019) and the article by Nieder et al. (2011) provide a great starting point for research into the influence of size layers on the penetration of wax-resin in general, but the possible removal of size has not been investigated.

1.4. Sub-questions, structure and methodology

To answer the main question, ‘what are the effects of Léo Marchand’s ‘ontlijming’ treatment on reconstructions

of 17th century Netherlandish canvas paintings and how does the treatment influence the distribution of wax-resin?’, multiple sub-questions are formulated. In order to place ontlijming in its context, Léo Marchand’s lining

practise is addressed. Further, the influence of size layers on the distribution of wax-resin in general is discussed. The last two chapters cover the effects of Marchand’s ontlijming treatment on preparatory layers and the influence of ontlijming on the distribution of wax-resin.

Chapter 2 provides an answer to the question what was Léo Marchand’s common lining practise? A step-by-step description of his wax-resin lining treatment, including the preparatory steps, gives context to the

ontlijming treatment. Ontlijming itself is further described. For this section, which mostly includes conservation

history, the techniques to gather relevant information include interviews, conversations and the analysis of documents such as treatment reports and photographs.

Following, the third chapter discusses the current knowledge on the influence of size layers on the penetration of resin. It answers the question what is the influence of size layers on the distribution of

wax-resin in general? This chapter relies mostly on literature.

The fourth chapter deals with the effects of Marchand’s ontlijming on preparatory layers. To find out whether only the size is affected, or whether the treatment has consequences for the whole preparatory system, the following question was asked; how does Marchand’s ontlijming treatment affect preparatory layers? Answering this question is done by means of experiments and instrumental analyses.

Lastly, in chapter 5 it is questioned whether differences in the distribution of wax-resin can be established between samples which are and are not ontlijmd, with other words, what is the influence of

(14)

ontlijming on the distribution of wax-resin? Again, this chapter will mostly rely on reconstructions. A set of

samples, half of them ontlijmd and half of them in their present state, will be impregnated with wax-resin. The distribution of the adhesive will be assessed by the discolouration and spreading of discolouration of the ground layers.

1.5. Relevance

As of now, no research on the wax-resin lining practice of Léo Marchand as outlined above has been carried out. Since he subjected all unlined canvasses and previously glue-paste lined canvasses to the ontlijming treatment, this research can be related to a large group of objects. As mentioned before, the effects of the treatment may not or not yet be visible on the surfaces of paintings, but they may occur in deeper parts of paint structures. Moreover, it has to be taken into account that possible phenomena are not yet recognised as such. This thesis research aims to investigate these effects of Marchand’s ontlijming treatment.

In a broader context, wax-resin lining treatments are currently of great interest within the field of conservation and during the past decennia a growing number of articles that describe wax-resin lining treatments and their consequences have been published. Wax-resin lining practises in the Netherlands have been studied to some extent too.18_{However, there are considerably less publications on private practice restorers that were}

active in the Netherlands during the second half of the twentieth century. It is now urgent to gain insight into the conservation treatments of these restorers. Literature describing the practise of Leo Marchand is very limited, although the paintings that he treated are widespread through Dutch collections.19_{Moreover, in past}

publications little attention is devoted to preparatory steps before lining and their possible consequences. By describing Marchand’s lining practise and the ontlijming in specific, this thesis will contribute to the growing field of conservation history. Despite the research on wax-resin lining carried out over the past decennia, there are still major gaps of knowledge and this thesis is only a start in a broader field of research into wax-resin lining practises.

18_{For instance:}

Esther van Duijn, “The studie of Goltzius’ Jupiter and Antilope – a study into its conservation history”, Cr 3 (2005). _____., “Vader en zoon Hopman; een tijdsbeeld aan de hand van twee 19de_{-eeuwse restauratoren”, Cr 3 (2006).}

_____., “Restauratoren uit het verleden; Dr. Ir. Angenitus Martinus de Wild; schilderijenrestaurator, chemicus, onderzoeker, kunstkenner en docent,” Cr 1 (2007).

Mireille te Marvelde, “How Dutch is the Dutch Method? A History of Wax-Resin Lining in its International Context,”

Past Practise – Future Propspects, The British Museum Occasional Paper, nr. 145 (2001): 143-149.

19_{Leo Marchand worked part-time for the Dordrechts Museum from the 1970’s onwards, but ran a private practise}

(15)

2. Wax-resin lining method

Léo Marchand was trained as an artist and specialised in paintings conservation after the Second World War. His main education within the field of conservation took place in the Rijksmuseum in 1953 – 1954, during which Henricus Hubertus Mertens (1905 – 1981) was the main paintings conservator. It was assumed that Marchand’s lining practise resembled the method used in the Rijksmuseum at that time. The description of Marchand’s wax-resin lining practise is mainly based on conversations with Pauline Marchand, Jos Deuss and the written interview by Deuss in 1999. It must be acknowledged that the procedure could slightly differ between individual paintings, dependent on their condition, materials, techniques and sizes. It is therefore decided to focus mainly on those steps of the treatment that affect paintings as a whole, rather than local work such as tear mending or the treatment of holes and distortions.

2.1. Wax-resin lining as performed by Marchand

2.1.1. Preparation and composition of the wax-resin adhesive

Léo Marchand’s wax-resin adhesive typically consisted of approximately 5 parts of wax and 1 or 2 part(s) of resin. The wax and resin were melted in a casserole pan on an electric chafing. Due to the relatively high melting point of resin, the resin was melted first and the beeswax was added only after the resin had softened. The amount of wax-resin that was prepared did not correspond to the amount necessary for one painting, the mixture was cooled down and heated up again when needed.

A study of treatment reports by Marchand present in the Dordrechts Museum dating from ca. 1965 – 1990 revealed that the wax was either referred to as ‘was’ (wax) or ‘bijenwas’ (beeswax), except for one occasion in which a modern painting was lined.20_{Of the resin it is known that Marchand commonly used colophony in the}

early years of his conservation career, although he made mention of ‘Dammar – Resina’ as early as 1969.21,22 _It

was probably in the second half of the 1970’s that Marchand substituted colophony with dammar as the main resin in his lining adhesive.23_{Apart from the wax and resin Jos Deuss made mention of Venetian turpentine, a}

natural balsam that acts as a plasticizer, as an additive.24

20_{The wax-resin adhesive for modern paintings is discussed later in this section.}

21_{Jos Deuss, Schriftelijk interview Jos Deuss, 1998-1999: bedoekingsspecie. See appendix I 1.2.}

22_{The use of dammar is mentioned in the 1969 treatment report of Jan Veth, Portret van Prof. Dr. P.J. Veth, 1886. Oil}

on canvas, 99,5 x 79,5 cm. DM/969/371. Dordrechts Museum. (treatment report by Léo Marchand, 1969, Dordrechts Museum).

23_{This indication is based on Deuss, Schriftelijk interview Jos Deuss, a private conversation of the author with Jos}

Deuss and a private conversation of the author with Pauline Marchand. Jos Deuss worked with Léo Marchand in the early 1970’s and mentioned colophony as the main resin used, whereas Pauline worked with her father in the early 1980’s, by then dammar had become the main resin for wax-resin mixtures.

(16)

For modern paintings the unbleached beeswax was replaced with ‘witte bijenwas’ (bleached beeswax) and elemi was added to prevent the paint surface from changing colour.25_{According to Deuss, the same mixture}

was used for older paintings in which parts of the ground layer were left exposed. Marchand was aware of the possible colour change caused by the impregnation and thus tried to minimize the impact by adapting the adhesive.

2.1.2. Preparatory steps

Before the actual lining of a painting several preparatory steps had to be taken, which included the removal of the painting from its strainer, the application of a facing, the stretching of the painting onto a framework, cleaning the verso of the painting and local treatments on tears, holes and distortions. As mentioned before, only the steps that affect the painting as a whole are discussed and local treatments are not dealt with within this thesis research.

Assembling the frames and stretching the lining canvas

Marchand used two working frames; one to stretch the lining canvas and a second one to stretch the painting. Thus, in this thesis the frame to stretch the lining canvas is referred to as stretching frame and the second frame, on which the painting is stretched by means of paper strips and glue, is called the pasting frame. The two strainers fit in each other, the one for the lining canvas being the smallest although the inner measurements of the stretching frame still had to exceed the outer measurements of the painting by at least 10-15 centimetres.

The frames were made out of wood, Marchand stored loose bars in multiple sizes that were often reused. After assembling the bars into a stretching frame that was hold together by cleats on the joints, the lining canvas was stretched from the middle of the length and breadth towards the corners by the use of a canvas plier and fixed with carpenter nails. Marchand never moistened the lining canvas to enable a second stretching, it was only stretched once. The side of the lining canvas that would be adhered to the painting was sanded to remove possible knots and irregularities.26

When it comes to the lining canvas, a coarse, basket woven linen was the standard [figure 3]. The Belgian linen came in rolls and Léo Marchand ordered those together with paintings restorers Henk Dik (? - ?) and Pieter

25_{In the 1978 treatment report of J. Jordens, Portret van Coosje, 1924. Oil on canvas, ?. DM/978/529, Dordrechts}

Museum, Marchand mentions that expressionists paintings, but generally almost all modern paintings, require a different treatment method. To prevent the paint layers from discolouration, he used a bleached beeswax, elemi balsam and dammar mixture.

26_{Jos Deuss, Schriftelijk interview Jos Deuss, 1998-1999: Voorbereiden bedoeking bedoekingsdoek. See appendix I,}

(17)

de Dood (1881 – 1959). Although the basket woven linen was almost always used, an extremely fine painting would occasionally be lined with a less coarse, plain woven linen canvas.28

Fig. 3. Detail of lining canvas, taken from Adriaen Beeldemaker, Honden in een landschap II, 1697. Wax-resin lined by Léo Marchand in 1971.

Stretching the painting

To stretch the painting on its temporary frame, it was placed on a table face up. After the back of the painting was filled up with wooden planks to avoid sagging, the painting was loosened from its strainer. Then, the stretching frame and pasting frame were put around the painting and the evenness of their heights was checked by means of a thin slat that was laid on top. If the painting was not on the same plane as the frames, wooden blocks could be inserted underneath the painting to raise it to a proper height.29

Once all frames were on a continuous level, the stretching frame was removed from its middle position and the painting was loomed (stretched on the larger frame). To attach the painting to the pasting frame, Marchand used strips of whte paper and a rice starch glue. First, the strips of paper were cut into the desired measurements; they had to cover approximately 1-2 centimetres of the painting and at least 3-4 centimetres of the frame. The rice starch glue was then prepared in a rubber pot. Following, the adhesive was applied to the paper and the painting was stretched [figure 4]. The paper strips were first applied to one of the long sides of the painting, after which the opposite side was done. Similarly, the strips were applied to the short sides of the

27_{Private conversation of the author with Wil Werkhoven, 22/03/2019.} 28_{Private conversation of the author with Jos Deuss, 15/02/2019.}

(18)

painting. The length of a large painting was never covered with one strip of paper, but always with multiple short strips that partly overlapped.

Once the paper and adhesive had dried thoroughly and the painting was stretched drum-tight, it could safely be taken from the table. The painting’s strainer as well as the wooden support could be removed and the facing could now be applied. Large paintings were placed in an upright position to do so.

Fig. 4. Léo Marchand applies paper strips to stretch a painting. Photograph: Regionaal Archief Dordrecht, < https://beeldbank.regionaalarchiefdordrecht.nl/index.cfm?action=search.viewimage &img=serveimage%2Ecfm%3Ffile%3Dlarge%2F2%2F3%2F947_2309%2Ejpg >

Application of a facing

To protect the paint surface it was covered entirely with one or more sheets of paper attached with an adhesive, also known as facing. The paint surface was not cleaned, nor subjected to varnish removal before the protection

(19)

was applied. Just as for the stretching of the painting, the white paper and rice starch glue were used for the facing.

A piece of paper large enough to touch the top and bottom of the wooden frame was cut off. To protect the paint surface from an imprint, the paper strips were applied without overlap [figure 5] and their edges were sometimes torn off to avoid sharp lines. The adhesive was brushed onto the paper and after several minutes it was ready to be placed on the painting’s surface. Any air bubbles were brushed out with a soft brush. The facing was allowed to dry before continuing with further steps of the treatment.

Fig. 5. Facing of unknown painting after lining. Photograph: courtesy of Pauline Marchand.

Cleaning the verso of the painting

Once a painting was stretched on the pasting frame and protected with a facing, Marchand would continue with flattening the tacking margins and cleaning the verso of the painting. To facilitate these steps the painting was put face down and secured to the table with clamps [figure 6].

Some water was dripped in the folding edges of the tacking margins before they were flattened. Besides moisture, Marchand also used heath; the margins were flattened with small hot irons. Following, some of the wax-resin mixture was applied in the folding edges, they were saturated with the adhesive and stuck to the paper. In order to avoid the aqueous mixture that would be used for cleaning later on from running to the front of the

(20)

painting, the wax-resin mixture was broadly spread around the tacking margins too. This can be seen on [figure 6], which shows the darkish discolouration of the paper around the painting due to saturation with the wax-resin adhesive.

Fig. 6. Léo Marchand removes an old lining. Photograph: Regionaal Archief Dordrecht, < https://beeldbank.regionaalarchiefdordrecht.nl/search/detail/id/BEEB5C66AC8711E489FB00163E535DC5/showbrowse >

Once the tacking margins were flattened, a previous lining or previous linings could be removed if applicable [figure 6]. Lining canvasses were removed by pulling the canvas off, sometimes in smaller strips [figure 7]. Residues of old adhesives were mechanically scraped off with a scalpel, spatula or abrasives [figure 8]. In the case of glue-paste linings, the glue adhesive was moistened and swollen with compresses in alternate rectangular areas (checkerboard pattern) and then scraped. Old inserts, patches or other tear mending materials as well as fillings were removed too and locally treated.

(21)

Fig. 7. Removal of old linings from unknown artist, Gezicht op het klooster Port-Royal-des-Champs, ca. 1700-1749. Oil on canvas, 94 x 116 cm. Museum Catharijneconvent, Utrecht. Treatment; 1972.

Photograph: courtesy of Pauline Marchand.

Fig. 8. Léo Marchand scrapes off (wax-resin???) adhesive residues. Photograph: courtesy of Pauline Marchand.

(22)

Before the actual lining took place, the verso of the painting was cleaned with an aqueous mixture. The intention of the cleaning was to remove dust and dirt that would, if left in the canvas, be pushed into the paint structure by the wax-resin adhesive. Moreover, the aim was to remove the size layer of the canvas in order to enhance the penetration of the adhesive. To do so, Marchand filled a small bucket with lukewarm water and added a dash of ammonia and a dash of denatured alcohol.30_{The reverse of the canvas was soaked with the}

water-ammonia-alcohol mixture and Marchand would start to brush thoroughly. As a result, foam appeared on the wetted canvas and Marchand believed this to be the animal glue. The brown foam was scooped from the canvas, which was rinsed with plain water afterwards and then let to dry.

As can be read in some of the conservation reports in the Dordrechts Museum, Léo Marchand considered the previously described treatment as a preparatory step and referred to the specific method of removing the size as ‘ontlijming’.31_{In the 1973 treatment report of a painting by Abraham Teerlink (1776 – 1857),} De Grot van Neptunus te Tivoli, it is moreover mentioned that the original canvas was sanded down. As for the

rationale behind those steps, Marchand wrote that both the ontlijming and the sanding of the canvas were carried out ‘om een goede impregnatie te verkrijgen’ (to obtain a good impregnation).32

2.1.3. Lining the painting

With regards to the lining practise, there is some discrepancy between (short) notes in Marchand’s treatment reports, Jos Deuss’s written interview and Pauline Marchand’s descriptions. The differences mainly concern the number of applications of wax-resin and of the ironing. In the treatment reports, Léo Marchand sometimes specified an ‘impregnation’ of the painting before lining. Deuss described the lining as a two-step process; first, wax-resin adhesive was brushed onto the reverse of the painting and ironed, subsequently, the lining canvas was put on the original and again wax-resin was added, after which the canvasses were joined by a second round of ironing. However, Pauline Marchand mentioned the application of wax-resin on the reverse of the original, after which the lining canvas was introduced without additional adhesive and the canvasses would be ironed only once. Given Marchand’s own mentions of an ‘impregnation’ before lining it is assumed that the two-step procedure as described by Deuss includes the impregnation; the impregnation being the first ironing of the wax-resin adhesive into the painting. Since the ‘impregnation’ is specifically mentioned in a few of the reports, always when the condition of the paint layer was described as poor and/or when problems with adhesion of the paint film arose, it is furthermore assumed that not all paintings were impregnated. Therefore, the main lining practise

30_{Private conversation of the author with Pauline Marchand, 15/11/2017.}

31_{1973 treatment report of Abraham Teerlink, De grot van Neptunus te Tivoli, ?. Oil on canvas, 96 x 130,5 cm.}

DM/878/175. Previously in the collection of the Dordrechts Museum. (treatment report by Léo Marchand, 1973, Dordrechts Museum).

(23)

of Marchand seems to have been a one-step procedure, with the possibility of an impregnation (two-step procedure) if the condition of painting required this. Both steps will be discussed below.

Before any action within the lining process took place, the frame on which the painting was loomed was secured to the table with clamps as can be seen on [figures 6 & 8]. Deuss noted that (a) blanket(s) were slipped in between the table and recto of the painting.33_{Then, a sufficient quantity of the melted wax-resin mixture was}

brush applied on the reverse of the original canvas. If the painting suffered from flaking paint or the paint layer was otherwise in a very poor condition, the painting would be impregnated first. Otherwise, the lining continued with the joining of the two canvasses.

Impregnation

In some treatment reports present in the Dordrechts Museum, Marchand wrote about the impregnation of the painting before wax-resin lining. In the case of a painting by Anton Mauve (1838 – 1888), for instance, it is described that the paint film delaminated from the ground layer, and that the painting was thus ontlijmd from the back and then impregnated to adhere flaking paint to the ground.34_{For two 17}th_{century paintings by Samuel}

van Hoogstraten (1627 – 1678), that were both in a very poor condition and suffered from flaking paint- and ground layers, the impregnation is mentioned too.35_{Therefore the rationale behind the impregnation seems to}

be the adhesion of delaminated ground- and/or paint layers by the wax-resin mixture.

To impregnate the painting, a hand-held iron was heated. Since the heat could not exactly be controlled, the temperature was checked by spitting on the iron and once the saliva would bubble, the iron was considered to be hot enough to work with. The ironing would always start in the middle of the painting, the hot iron was passed over it in circular movements, working towards the outer edges of the painting. Assessment of the impregnation was done by lifting the painting from the table and observing its facing, which would in the case of proper impregnation saturate with wax-resin and darken.36

Joining the canvasses

For paintings that were impregnated, the stretched lining canvas was placed inside the loom of the painting and more adhesive was brush applied on the verso of the lining canvas. In most cases, however, the painting had not been impregnated and the previously applied adhesive on the reverse of the painting was sufficient to join the

33_{Deuss, “Schriftelijk interview”, bedoekingsspecie.}

34_{Treatment report of Anton Mauve, Duinlandschap met Schapen, ?. Oil on canvas, 38,5 x 64,5 cm. DM/889/31,}

Dordrechts Museum. (treatment report by Léo Marchand, 1972, Dordrechts Museum).

35_{Treatment report of two paintings by Samuel van Hoogstraten, Portret van Maria Stoop & Portret van Jacob van}

Ouzeel, ?. Oil on canvas, 81 x 65,5 cm & 81,3 x 65,5 cm. DM/937/524 & DM/937/513. Dordrechts Museum.

(Treatment report by Léo Marchand, ?, Dordrechts Museum).

(24)

canvasses. Since the wax-resin adhesive is a thermos-reversible material, it requires heating and cooling down to facilitate the adhesion between lining canvas and original. Once the stretched lining canvas was brought in touch with the original painting, it would be secured by stones or weights to keep it in place. Again, the ironing was performed with the hand-held iron, heated as described above, moving in circular movements from the middle of the painting outwards. During the process the iron would lose its heat and the circular movements were repeated until it had cooled down completely. Any surplus of the wax-resin adhesive that had been pushed beyond the tacking margins of the painting, was collected and put back into the pan. Subsequently, the canvasses were ironed with a cold, heavy iron until the adhesive on the lining canvas became glossy.37_{Lastly, the painting}

itself was cooled down completely by lifting it from the table, after which the paper strips were cut.38_{The painting}

had now been lined.

Jos Deuss mentioned in his schriftelijk interview that an extra blanket was sometimes put underneath pastose paintings to avoid flattening of the impasto paint, those paintings would also be subjected to less pressure during the ironing.39

2.1.4. Cleaning the recto of the painting

After the lining canvas had been adhered to the original and the paper strips were cut, the cleaning of the front of the painting remained. To enable any work on the front, the painting was turned around and put face up on the table. As described by Pauline Marchand, a protective facing that had not been fully saturated with the wax-resin adhesive could be moistened. In addition, Jos Deuss described that the surface would be wetted all over with a sponge to swell the starch glue and remove the paper [figure 9].40_{However, the facing was almost always}

impregnated with the adhesive, which allowed for an easy mechanical removal by just pulling the strips of paper from the surface [figure 5].41

By pulling the paper facing off, the varnish film was (partly) pealed from the painting too. In general, the varnish film had been strongly affected by the wax-resin lining; Deuss mentioned that it usually powdered due to the heat, moisture and pressure applied during the lining treatment.42

Lastly, residues of the wax-resin adhesive on the verso of the painting were removed with “wasbenzine” (alkane hydrocarbons). Although not cleaning, Marchand typically still applied some more wax-resin over the sides (tacking margins) of the painting after stretching it on its old or new stretcher; the adhesive can thus be found to cover the nails, which might be indicative for a Marchand-lining.

37_{Private conversation of the author with Pauline Marchand, 09/05/2019.} 38_{Ibid, bedoekingsspecie.}

39_{Deuss, “Schriftelijk interview”, bedoekingsspecie.} 40_{Ibid, schoonmaken in de regel na bedoeking.}

41_{Private conversation of the author with Pauline Marchand, 01/03/2019.} 42_{Deuss, “Schriftelijk interview”, schoonmaken in de regel na bedoeking.}

(25)

Fig. 9. Léo Marchand removes the facing of an unknown painting. Photograph: Regionaal Archief Dordrecht. < https://beeldbank.regionaalarchiefdordrecht.nl/search/detail/id/BED47BF6AC8711E489FB00163E535DC5/showbrowse >

2.2. Ontlijming; a closer look

As mentioned in the description of Marchand’s lining practise, an aqueous solution of (mainly) water, a dash of ammonia and a dash of denatured alcohol was used to moisten the reverse of the original canvas. No measures were taken, but given the fact that both Jos Deuss and Pauline Marchand described the amount of ammonia and alcohol to be ‘een scheutje’ (a dash), it could not have been much. However, the effect of this mixture on the canvas and paint structure is not yet known. According to Pauline Marchand and Jos Deuss, who both performed linings and thus the ontlijming treatment with Léo Marchand, a brownish/greyish foam would appear on the reverse of the canvas whilst brushing. Marchand believed this to be the animal glue size and regularly scooped it from the canvas until the surface ceased to foam. Especially Deuss stressed the importance of rinsing the

(26)

canvas with plain water afterwards to wash out the remainders of dust, dirt and glue. In sum, lots of water, both in combination with ammonia and alcohol as well as plain water afterwards, was introduced in the original painting. It has furthermore been mentioned that the painting would be dried close to the heater after the

ontlijming treatment.43

As described in the above section, Marchand would impregnate the tacking margins and a broad margin of the facing paper with wax-resin adhesive before subjecting the painting to the size removal. The main aim was to avoid any of the aqueous mixture from running to the front of the painting and thus affecting the facing. However, it sometimes did happen that the aqueous solution penetrated the paint system all the way to the protective paper, especially when paintings suffered (large) cracks and losses. The facing was affected by the moisture, but not re-applied or otherwise fixed. Because of the drying afterwards, the paper would re-adhere to the paint surface.44

The study of treatment reports by Marchand in the Dordrechts Museum showed that ontlijming as part of the lining process is only sometimes mentioned, almost exclusively in early reports (ca. 1965 – 1975). In general, the early treatment reports are somewhat more extensive in describing the condition and treatment of the painting and as the years pass by, the documentation becomes more concise. Furthermore, in later reports Marchand often mentioned that a painting was lined by his ‘common lining practice’ or the ‘already known method’. Since he worked on paintings for the Dordrechts Museum on a regular basis, the earlier treatment reports included most of the information and it was not necessary to repeat this for each individual painting. The fact that ontlijming is not mentioned in a treatment report does therefore not mean that it was not performed; according to Deuss, all paintings were subjected to cleaning with the aqueous mixture. Furthermore, no distinction was made between canvasses that had previously been lined or that were in an unlined state; both would be ontlijmd, previously lined paintings only after the removal of the old lining(s).45_{The removal of glue}

residues from a glue-paste lining and the removal of an original animal glue layer were clearly separated within the procedure and served different purposes.

2.3. Ontlijming in the practice of Marchand’s contemporaries

The main education of Léo Marchand within the field of conservation took place in the Rijksmuseum. However, Pauline Marchand mentioned that Léo came into contact with paintings conservation in France shortly after the Second World War, almost a decade before his Rijksmuseum internship.46_{According to a letter of Léo Marchand}

to Arthur François Emile van Schendel (1910 – 1979), conservator of the Rijksmuseum, he had already performed

43_{Private conversation of the author with Jos Deuss, 15/02/2019} 44_Ibid.

45_{Private conversation of the author with Jos Deuss, 15/02/2019.}

(27)

some treatments on paintings from 1948 onwards. Nonetheless he saw the need of further education and hoped to acquire the necessary knowledge and expertise at the Rijksmuseum. Marchand followed an internship in the conservation studio of the Rijksmuseum between mid-1954 and the 17th_{of November 1955, which was}

paid for by the Sociale Bijstandsregeling voor Beeldende Kunstenaars (SBBK).48,49_{During that time he must have}

learned the wax-resin lining practise of Henricus Mertens.

Unfortunately, little is known about the conservation practise of Mertens. According to Esther van Duijn, paintings conservator and conservation history specialist at the Rijksmuseum, there are almost no records in which he described his lining practise.50_{A few years before Marchand’s internship, however, the American}

paintings conservator Louis Pomerantz (1920? – 1988) also trained at the Rijksmuseum.51_{Today, his student}

notebook is owned by the Smithsonian Archives of American Artists, Washington, as part of the Louis Pomerantz papers. Pomerantz gave a detailed account of the wax-resin lining procedure as learnt from Mertens. His step description of the procedure, including coloured sketches and a series of photographs covering all steps of the treatment, is strikingly consistent with Marchand’s practise. However, there is no mention of a size removal or other aqueous cleaning that resembles the ontlijming at all. Since there are no conservation reports, nor other documentary sources or stories that make mention of the practise of a size removal treatment in the Rijksmuseum, it was suggested by Van Duijn to look into the procedures of private practice restorers.

In the search for the origin of the size removal treatment Wil Werkhoven, a paintings conservator who runs a private practise in Makkum, was contacted. Werkhoven worked several years with Henk Dik, one of the conservators that Marchand shared his lining canvas and paper with. Although Werkhoven acknowledged that Marchand visited the studio of Dik every now and then and that they occasionally discussed conservation treatments, she did not recognize the description of ontlijmen. It was definitely not used in the studio of Dik.52

Another contemporary of Marchand was Nico van Bohemen sr., both Marchand and Van Bohemen wax-resin lined paintings for the Royal Palace in Amsterdam in 1963. Nico van Bohemen’s son, Van Bohemen jr. is still active as a paintings conservator in The Hague. He too ensured that no treatment similar to ontlijming was practised in his father’s studio.53_{Both Werkhoven en Van Bohemen jr. expressed that treatments containing such quantities}

of water were by all means avoided.

47_{Correspondence between Marchand and A.F.E. van Schendel, 16/12/1953, present in the staff administration of}

the Rijksmuseum. Courtesy of Esther van Duijn. The type of treatment is not explained.

48_{Correspondence between Gemeente Dordrecht, Rijksmuseum and Léo Marchand, present in the staff}

administration of the Rijksmuseum. Courtesy of Esther van Duijn.

49_{Ministerie van Onderwijs, Kunsten en Wetenschappen, “Verslagen der Rijksverzamelingen van Geschiedenis en}

Kunst.” Part 27, (The Hague: Staatsdrukkerij, 1956), 14.

50_{Private conversation of the author with Esther van Duijn, 23/04/2019.}

51_{Louis Pomerantz’ notebooks cover the period between ca. October 1950 – february 1951.} 52_{Private conversation of the author with Wil Werkhoven, 22/03/2019.}

(28)

The notion that introduction of water into the original canvas was feared seems to be the general conception of Marchand’s contemporaries, and even of earlier generations of paintings conservators. Stanley Cursiter and Martin de Wild described in their 1937 article ‘Picture relining’ already that the amount of water used ‘should be restricted to the minimum’ and that the ‘whole canvas should not be dampened at one time […]’.54_{It must be acknowledged that reverses of original canvasses were often cleaned with moisture, this was}

not restricted to Marchand. However, for the cleaning of canvasses with aqueous solutions it is almost without exception recommended to work in alternate square patterns to avoid dampening of the whole canvas at once and/ or to thicken the solution with a starch.55_{With the present-state knowledge, Marchand’s ontlijming that}

included lots of water and a full staining of the original canvas seems to stand on its own and there are no indications that it was commonly employed by paintings conservators at that time.

54_{Stanley Cursiter, and A. Martin de Wild, “Picture relining,” Technical Studies in the Field of the Fine arts, Vol. V part}

3 (January 1937), 164.

(29)

3. Influence of size layers on the distribution of

wax-resin

Marchand’s main aim of the ontlijming treatment, namely to remove the size layer in order to enhance the penetration of wax-resin, implies the assumption that size layers on paintings influence the impregnation with wax-resin. Previous research has shown this to be true. However, the impact of size layers on the distribution of wax-resin is much dependent on the ground composition, binder and application of the size. This chapter discusses consequences of the presence and application of size layers on the distribution of wax-resin.

In literature, the influence of size layers on the penetration of wax-resin adhesives for lining has not been dealt with to a great extent. It is mainly described by Nieder et al. in the article ‘Colour Change in Sample

Reconstructions of Vincent van Gogh's Grounds due to Wax-Resin Lining’, which was part of the Historically

Accurate Oil Painting Reconstruction Techniques (HART) Project. Apart from the focus on 19th_{century ground}

layers, the lining adhesive and method used for the experiments were the ones used by conservator J.C. Traas (1898 – 1984).56_{It is therefore important to note that the effects described in the article by Nieder et al. may be}

accurate for Van Gogh-like grounds combined with the lining method of Traas, but that the effects may slightly differ for other ground compositions and wax-resin lining adhesives as well as techniques.

Since the goal of the research by Nieder et al. was to investigate the visual impact of wax-resin lining on ground layers of paintings by Vincent van Gogh and his contemporaries, the effects were assessed by means of the darkening and discolouration of a set of samples after lining. Different ground layers have been used, composed of chalk, lead white, barium sulphate or bone black as well as combinations of those, bound in either animal glue, oil or an emulsion.57_{For the size layers an animal glue was chosen and distinctions were made}

between no size, gelled applications and fluid applications of the size. Half of each sample was lined with an adhesive consisting of seven parts unbleached beeswax, four parts colophony and one part larch turpentine. The wax-resin was applied to the reverse of the sample and impregnated with a hot spatula set at 90o_C.

3.1. Effects of size on the darkening and discolouration of

ground reconstructions

In general, wax-resin lining of the ground reconstructions within the research of Nieder et al. caused darkening and discolouration of the samples. The binding medium was found to be the most influential, with the greatest changes observed in glue-bound samples, the least changes in oil-bound samples and the emulsion-bound

56_{Nieder et al., “Colour change”, 93.} 57_{Ibid, 96.}

(30)

samples were in between [figure 10].58_{Size layers influenced the visual impact of lining on the ground layers as}

well; according to Nieder et al., the presence and application of size affected both the change of colour and the evenness (spreading) of the discolouration. For glue-bound samples, grounds without size darkened evenly and changed most in colour, unlike grounds with gel-applied size which darkened only little. Samples with a fluid size were again in between, showing a strong change in colour, but an uneven spreading of the discolouration [figure 11].59_{Surprisingly, the article lacks information on the procedure of colour measuring. Nieder et al. suggested}

that the discolouration in samples with a gelled size was only very little, but since the procedure and spots on which the colour was measured are not described, it is not known whether the measured colour difference is representative for the discolouration. It may be that the colour measurements before and after lining were taken on the exact same spot, which can in the case of uneven spreading be problematic due to the risk of measuring an unchanged part of the ground layer. As a consequence, the measured colour is an average colour of the surface after lining, combing both the discolouration and spreading, instead of a measurement of the discolouration on its own. Since this thesis addresses both the discolouration and spreading, there will be a sharp distinction between discolouration (the measured change of colour) and spreading (the area on which the colour change occurred).

Unlike the assumption by Nieder et al. that the size influences the colour change, Froment found that the presence of a size layer did not impact the discolouration of ground reconstructions used in her thesis research.60_{The procedure of colour measuring by Froment, however, may have differed from the one by Nieder}

et al. Froment described that colour measurements were taken on the same spot before and after lining if possible, but in the case of an uneven spreading of the colour change, the spot was moved inside a discoloured area.61_{Thus, only the discolouration itself was measured, excluding the unchanged areas. Taken together, the}

presence of size might influence the discolouration, but may be interpreted differently depending on the colour measuring procedure.

58_{Ibid, 97.} 59_{Ibid, 96-97.}

60_{Froment, “The consequences of wax-resin linings”, 102-103.} 61_{Ibid, 234.}

(31)

3.2. Effects of size on the impregnation with wax-resin

The presence of an animal glue size was by both Froment and Nieder et al. found to affect the spreading of the discolouration.62_{Especially for glue-bound ground reconstructions with size the impregnation and thus the}

discolouration was uneven, whereas samples without size darkened evenly.63,64_{Moreover, Froment noted in her}

PhD thesis ‘The consequences of wax-resin linings’ that glue-bound grounds on unsized canvas ‘underwent a more rapid and more even colour change than the same ground applied on the sized canvas. The latter showed local unchanged areas even after three applications of adhesive and ca. 7 minutes of ironing.’65_{The way in which}

samples discolour to a varying degree seems indicative for their impregnation. The local unchanged areas in

62_{Ibid, 102-103.} 63_{Ibid, 103.}

64_{Nieder et al., “Colour change”, 96-97.}

65_{Froment, “The consequences of wax-resin linings”, 103.} Fig. 10. The influence of binding medium on the effects of

lining, illustrated by samples with chalk grounds bound in glue, emulsion and oil, applied on fine linen with fluid size. Sample dimensions approximately 3,25 x 3,25 cm. Photograph: Nieder et al., “Colour change”, 99.

Fig. 11. The influence of the size layer on the effects of lining, illustrated by samples with chalk grounds bound in glue medium on fine linen prepared with gelled size, fluid size and no size. Sample dimensions approximately 3,25 x 3,25 cm. Photograph: Nieder et al., “Colour change”, 99.

(32)

bound samples on sized canvas, even after multiple applications of wax-resin adhesive and longer exposure to hot ironing, shows that the impregnations is incomplete.

In contrast to glue-bound samples, the oil-bound ground reconstructions without size darkened the least and the samples with a gelled size darkened the most.66_{Nonetheless it was observed that the size layer acted as}

a barrier against the wax-resin mixture. For grounds composed of chalk in oil with gelled size layer, only the weave tops of the canvas darkened, ‘suggesting that the gelled size prevented saturation of the ground layer, whereas the surface of the sample of chalk in oil over no size darkened evenly.’67

As for the barrier function of the size, Nieder et al. suggested that it may partly be due to the hygroscopic nature of the size layer, in contrast to the non-polar wax-resin.68_{In the article, the observed spreading of}

discolouration of the ground reconstructions was used to draw conclusions on the process of impregnation. It was described that the wax-resin ‘is first absorbed into the fibres of the support and then into the ground layer.’69

66_{Nieder et al., “Colour change”, 99.} 67_{Nieder et al. “Colour change”, 99.} 68_{Ibid, 96.}