With this post, we’d like to take the opportunity to welcome our new intern, Cécilia Duminuco. As we mentioned back in May, our internship programme is celebrating its 10th year, and we’re very happy that Cécilia has joined our team in this anniversary year. We hope you enjoy hearing about Cécilia’s work prior to joining the CBL Conservation team.
Before starting my internship at the Chester Beatty Library this November, I completed an MA research project at West Dean College, University of Sussex, in the U.K.
My research developed in collaboration with the Heritage section of the Maurits Sabbe Bibliotheek at the Catholic University of Leuven (Theology Faculty) in Belgium. I studied the sewing threads and sewing structures of twenty-four manuscripts and printed books from Western Europe, ranging from the 12th to the early 19th century, by means of two methods: visual assessments and Reflectance Transformation Imaging (RTI) with a Microdome, a digitisation tool created by the Catholic University of Leuven.
Figure 1: the Microdome. Credit: Marc, P., Vandermeulen, B. & Watteeuw, L. (8/09/2014) “See the Surface. Imaging and measuring surface characteristics of library materials by photometric stereo (RICH Project)”, Digital Humanities@Arts Summer School, Leuven, p.8
In this post, I will briefly present the Microdome, the imaging technique attached to it and the results of the sewing thread analysis with this apparatus.
Reflectance Transformation Imaging (RTI) is defined by the Cultural Heritage Imaging organisation as “a computational photographic method that captures a subject’s surface and colour and enables the interactive re-lighting of the subject from any direction. RTI also permits the mathematical enhancement of the subject’s surface shape and colour attributes. The enhancement functions of RTI reveal surface information that is not disclosed under direct empirical examination of the physical object”. The technique was invented by Tom Malzbender in the Hewlett-Packard Laboratories in 2000-2001. In 2005, following Tom Malzbender and the Cultural Heritage Imaging organisation’s previous research, a new imaging device -a Portable Light Dome (PLD) – was developed in the Catholic University of Leuven.
The Portable Light Dome (PLD), also called Microdome, is a dome-shaped tool including two hundred and twenty eight LED white lights of four thousand Kelvin. A high resolution digital camera is mounted on top of the device. The object to be studied is placed in the centre of the dome, and several pictures with different lighting angles are automatically recorded. The images are then processed and viewed with specific software. The viewing software, PLDViewer, allows the researcher to analyse the created image through the use of several tools, such as a zooming and rotating functions, a measurement tool, and a 3D model exportation tool. This technique allows the surface characteristics of an object to be studied precisely, non-invasively. Furthermore, several filters can be applied to the image, allowing the creation of enhanced views of the surface (such as the Sharpen or the Shaded filters).
Figure 2: Ambient light source and Colour filter, with light positions. Cod.4, pp.4-5.
Using a survey, I carried out visual assessment of each book’s sewing structure, before applying Microdome imaging on two pages of each book where the sewing threads were visible.
Figure 3: Albedo light source and Shaded exaggerated filter, with light positions. Cod.4, pp.4-5.
In bookbinding, a sewing thread is defined as “a filament or group of filaments used for securing the leaves or sections of a book.” These filaments or fibres are most commonly of natural origin; sometimes from animal sources, such as silk and wool, or plants such as flax, hemp and cotton. Fibres were traditionally harvested by hand, and according to their nature they were subjected to different processes such as cleaning, combing and twisting before spinning. This is what results in each thread’s unique characteristics.
Thread twist count
The twist of a thread is defined as the “turns per metre of yarn, used to hold filaments together”. The number of thread twists per centimetre was analysed for forty-six thread samples using the measuring tool -a one-millimetre wide vertical or horizontal grid applied to the images.
Figure 4: Twist count, Ambient light source and Sharpen filter. INC 805 EF_Etym, gathering beta.
Regardless of their date, the majority of the threads have approximately five to eight twists per centimetre (t/cm). In nine of the books, thread samples examined showed a consistent twist number range per book. However, in eleven books, the sewing threads had different twist numbers in each of the various samples analysed.
Figure 5: Twist count, approximately 6t/cm. PBM 248.158_Hens Viri, pp152-153.
The twist angle is defined as “the amount of twist in a yarn measured in degrees”. The images of forty-eight sewing threads were studied after enhancement. In order to calculate the angles trigonometry formulas were applied by measuring the thread thicknesses (the opposite side of the right triangle) in combination with the twist measurements (the hypotenuse).
Figure 7: A right-angled triangle ABC.
Figure 6: Twist angle (A) measurements. Cod.4, p4-5.
All of the sewing thread twist angles measured ranged from 15° to 36°, with fourteen threads presenting a twist angle ranging from 20° to 25°. The smaller twist angles did not appear to be characteristic of a certain time period. However, the higher twist angles (more than 30°) were more commonly found in books from the 16th to the 19th century. It should also be noted that when considering threads in one volume, twist angles may vary widely. However, the majority of the threads studied presented relatively similar twist angles, within a narrow range.
Figure 8: Twist angle measurements. PBM 248.158_Hens Viri, average ~23°, pp.152-153.
Using the Microdome software measuring tool, thread thicknesses were also calculated in the samples. The majority of the sewing threads (forty-one samples) have an average thickness ranging from 0.40 to 0.80 millimetres. Five threads, from the 17th to the 19th century, are thinner than 0.40 millimetres; and twenty others, with no specific time range, have a thickness ranging from 0.50 to 0.65 millimetres. Twelve threads, nine of them from the 16th century or earlier, are thicker than 0.65 millimetres. Within a single book, sewing thread thicknesses vary. Similarly, within a single thread sample, variations in thicknesses can be observed
Figure 9: Thread thickness measurements. INC 805 EF_Etym, average ~0.60mm, centre of gathering alpha.
In conclusion, all of the features studied revealed variations in the sewing thread structures. It seems likely that these variations could be explained by the hand-processing of the threads that were used.
A woman spinning fibres on a Picardy wheel. Engraving from Arts et Metiers, l’art du fabricant d’étoffes by J.M. Roland de la Platière, ~1780.
Analyses of these results has allowed me to define the advantages and limitations of both visual and analytical techniques for assessing books. In particular, the complimentary use of these two methods served to highlight the possibilities offered by the Microdome for use in the conservation field.
More details about the Microdome and my research will be presented at the Care and conservation of manuscripts conference in April 2016 in Copenhagen.
I am very grateful to Professor Lieve Watteeuw and to the Maurits Sabbe Bibliotheek of the Catholic University of Leuven who gave me such a wonderful opportunity to work with one of the most promising digitisation tools in the conservation field.
Cécilia Duminuco, Heritage Council Intern in Conservation