THE BRITISH LIBRARY

Collection Care blog

Behind the scenes with our conservators and scientists

Introduction

Discover how we care for the British Library’s Collections by following our expert team of conservators and scientists. We take you behind the scenes into the Centre for Conservation and the Scientific Research Lab to share some of the projects we are working on. Read more

24 November 2016

Applications of Image Processing Software to Archival Material

Images of archival material are useful to both conservators for monitoring changes, and to researchers for detailed analysis and permanent access to collection items. Image processing allows historical documents and other collection items to be studied without the risk of damage to the primary source. The increase in digitisation projects is generating large volumes of image files that can be processed to enhance the understanding of our collections without physically handling fragile material.

ImageJ is a powerful public domain Java-based image processing package. The nature of open source software allows for the constant update and availability of new plugins and recordable macros designed for specific tasks. ImageJ’s built-in editor and a Java compiler allow for the development of custom acquisition, analysis and processing plugins. In April 2013 I presented a poster at the ICOM Graphics Documents Working Group Interim Meeting in Vienna, outlining the applications of image processing software to archival material . The full poster can be downloaded as a PDF here.

C-DUFFY-poster

While several improvements have been made to the functionality of ImageJ since 2013, I hope this poster provides useful information to those less familiar with image processing techniques.

ImageJ was originally designed for the purpose of medical imaging by the National Institutes for Health by Wayne Rasband, but has since found applications in many fields. It can be run on any computer with a Java 5 or later virtual machine, as an online applet or as a downloadable application (Microsoft Windows, Mac OS, Mac OSX, Linux, Sharp Zaurus PDA). ImageJ offers features similar to commercially available image processing software packages such as brightness/contrast adjustment, frequency domain filtering, binarisation and particle analysis.

Christina Duffy

21 November 2016

The Conservation and Spectroscopic Analysis of a Burmese Concertina Binding

Background

Last year I was asked to conserve a high profile, yet highly degraded late 18th century Burmese Concertina binding (OR.3591). Almost everything that could go wrong with a paper-based manuscript had gone wrong with this. It was mouldy, it had pest damage, the media was coming off the paper, and there were significant losses of the paper substrate which had compromised its handling and legibility. Consequently the item was deemed ‘unfit’ for use by the British Library, and was in desperate need of conservation.

Concertina Bindings

A concertina binding, also known as an accordion binding, is a binding method that sits somewhere between a modern sewn book and an ancient scroll. Concertinas are customarily a long sheet (see Diagram 1), or multiple adhered sheets (see Diagram 2), folded vertically into panels, which can then either be displayed as a continuous sheet, or picked up and read like a book (making it far easier to handle than the often cumbersome scroll).

Diagram 1

Fig. 1 A concertina binding like OR.3591, formed of one continuous sheet of paper.

Diagram 2

Fig.2 A concertina binding formed of multiple sheets adhered together.

Figure 3

Fig.3 OR.3591 The carved relief wooden end board.

Fig.4

Fig.4 OR.3591 Stratified, highly deteriorated concertina folds.

OR.3591

This concertina was formed of one continuous sheet of handmade light brown long-fibred eastern paper (see Diagram 1), folded into 24 panels, between two dark wood boards with carved relief designs. Coloured paint and inks on a white ground adorned both sides of the paper. Side A illustrated Buddhist cosmology and side B illustrated fortune telling. The manuscript was acquired from a collector by the British Museum in 1888. The accompanying letter reads:

This book was found by me in the Laywun hut at Thatwé which is about 20 miles S.E of Yamethin in upper Burmah. Thatwé was occupied on the 10th December 1886 by a column of the 3rd Brigade under the command of the Brig. Gen W.S.A Lockhart.

The Burmese interpreter informed me that the pictorial side represents the progress of a saintly man from the nethermost regions to the acme of all goodness. One man shown upside down is the supposed after reaching an exalted position, to have led an immoral life for which he was sent down.

The tables on the reverse are those from which the horoscopes are worked out- every Burman always carries a horoscope."

P.D Jeffreys Lt. Col.
The Connaught Rangers
Late Brigade Major
3rd Brigade Burmah Field Force

Figure 5  Figure 5 and 6
Figs. 5 & 6 OR.3591 Accompanying letter from L t. Col Jeffreys.

Condition: the paper

Fig.7

Fig.7 OR.3591 Manuscript torn into two parts.

The paper was soft, weak and fragile to handle, distorted, holey and creased. The manuscript was in two parts, having torn down one of the fold-lines. There were also extensive losses on each panel, due in part to rodent damage.

The manuscript had also obviously been exposed to water at some point in its lifetime as there were clusters of black spots, which were identified as inactive mould. The water had also stuck the folded panels together, which had later been forced apart, skinning (or tearing off) the top layers of the paper, leaving them stuck to the opposing panels.

Figure 8  Figure 8 and 9
Left: Fig.8  OR.3591 Skinned, creased and torn panel edge with off-set accretions and media. Right: Fig.9  OR.3591 Water damaged media and inactive vegetative mould.

Condition: the media

Water damage had also solubilised the media, causing it to smudge and offset again onto apposing panels. When examined under microscopy the paint was powdery or ‘friable’ and was coming off the paper, suggesting the paint’s binder was exhibiting stronger cohesive rather than adhesive properties, and had ultimately failed in its utility.

Fig.10 and 11

Left: Fig.10  OR.3591 Friable powdery pigment media. Right: Fig.11  OR.3591 Example of visually obstructive off-set media on opposing panels.

Pigment Analysis via p-XRF Spectroscopy

Portable X-ray Fluorescence (p-XRF) Spectroscopy is a non-destructive elemental analysis technique for quantification of nearly any element from Magnesium to Uranium, which enables conservation professionals to identify pigments. Precisely identifying pigments governs the succeeding conservation treatment as some pigments can be adversely affected if treated incorrectly - for example some discolouration if exposed to some solvents. Identifying the pigments also gives invaluable insight into the history, materiality and manufacture of the manuscript.

Fig.12

Fig.12  OR.3591 Analysing pigments via use of p-XRF spectroscopy.

P-XRF analysis was carried out using a Bruker ‘Tracer-III SD’ portable device, and the resulting data was then compared against suitable reference data to determine the identities of the pigments. Four pigments from the manuscript were examined: red, white, yellow and green. In addition, the composition of the underlying paper was also assessed, and this datum was used as a ‘background’ so its influence could be removed from the results.

The data suggested that the four pigments are: red - vermilion; white - lead white; yellow - orpiment;
green - orpiment plus an organic blue (possibly indigo).

* Elements given in parentheses are present in minor quantities.

Table

Fig.13  OR.3591 p-XRF spectra.

Treatment

The manuscript was first photographed, and then the media cross-sectionally solubility tested with deionized (D/I) water, ethanol and a 50:50 mix. All media was soluble in water and 50:50.

Mould reduction

The dry surface mould was brushed gently with a sable brush into a HEPA filtered vacuum according to COSHH standards. Extreme care was taken during this process to limit loss of original media. To ensure absolute mould deactivation, the recto and verso of mould damaged areas were treated via local application of ethanol using a fine brush.

Humidification

The manuscript was humidified in a controlled environment using a blotter, Bondina and Gore-Tex stack covered with a polythene sheet. This process relaxed the paper and reduced its distortions, enabling me to accurately realign the fibres, creases and tears using tweezers.

Fig.14

Fig.14  OR.3591 Humidifying the manuscript.

Media consolidation

Straight after humidification the media was consolidated. Jun-Funori, a pure Japanese algae-based consolidant was made up to a 1.4% and inserted into a nebulizer. The nebulizer, which dispersed the consolidant in a fine mist, was then passed cross-sectionally over each panel on each side. The process was repeated twice until friable pigment was sufficiently consolidated, and the manuscript was left to dry. Jun-Funori was selected for the consolidant because it has documented success at consolidating friable media and it has a low refractive index so it is suitable for matte pigments. When emitted via nebulizer the particles penetrated between the friable pigments, adhering them to the paper.

Fig.15

Fig.15  OR.3591 Consolidating the friable media with Jun-Funori emitted in a fine mist via nebulizer.

Structural consolidation

A Japanese kozo paper ‘Nao 2.2’ was selected for infills and repairs due to colour, weight and availability. The paper was not toned as this would have taken up a large amount of time, and its likely inconsistency would have been more visually distracting than the selected paper.

The tears and holes were repaired with the kozo paper, torn and trimmed so the edges were neat but fibrous, and adhered with wheat starch paste (WSP). The losses were infilled with three layers of the paper, needled out along the loss edge, and adhered onto the manuscript again with WSP. This tri- laminate infill was of equal-weight to the manuscripts paper as desired.

Fig.16

Fig.16 OR.3591 The manuscript consolidated with Nao 2.2 kozo tri-laminate infills adhered with wheat starch paste.

Fig. 17

Fig. 17 OR.3591 The 2 parts of the manuscript, part 1 post conservation treatment, part 2 pre-conservation treatment. 

The infills were trimmed with a straight-edge and scalpel and the manuscript refolded using a bone folder. The manuscript was finally re-housed in its original (cleaned) box.

Figs.18 and 19

Figs.18 & 19  OR.3591 Folding the conserved manuscript back into its original concertina format.

The overall treatment took 113 hours, due to the sheer size of the object and the number of infills that needed to be carried out. The manuscript is now back in use and can be requested by readers - job done!

Fig.20

Fig.20 OR.3591 The conserved item having been returned to its original box.

Fig.21

Fig.21 OR.3591 Foredge showing consolidated concertina folds.  

Fig.22

Fig.22 OR.3591 A section of the conserved manuscript.

Fig.23

Fig.23 OR.3591 A section of Side B showing Burmese horoscopes.

Fig.24

Fig.24 OR.3591 The completed manuscript.

Fig.25

Fig.25 OR.3591 A section of Side A showing Buddhist cosmology.

Fig.26

Fig.26 OR.3591 A section of Side B showing the extent of infills that were required.

Daisy Todd

11 November 2016

The British Museum Bindery heroes

In the British Library Conservation Centre there is a small plaque commemorating four members of the former British Museum Bindery staff who were killed in the First World War.

1

Until 1927 the work of binding and restoring the library's collections was contracted out to private concerns. The earliest contractor was appointed in 1760 and in 1881 the firm of Eyre and Spottiswoode was awarded the contract to supply binding services to the Museum Library. In 1927, His Majesty's Stationery Office (HMSO) took over the Bindery and continued until April 1982 when the staff of the binderies became employees of the British Library.

The four men commemorated are Pte Horace Crawley, L/Corp Horace Davis, Rifleman Thomas Wickham and Pte Alfred Williams. I have been trying to discover a little about who they were and when and where they died. The main source of information is the Commonwealth War Graves Commission (CWGC)’s website, which gives details of all casualties and where they are buried or commemorated. Further information can be found at The National Archives in the War Service records and the battalion War Diaries. The War Service records show when and where each person joined up – unfortunately they are incomplete as many were damaged or destroyed by enemy action during the Second World War. The Commanding Officer of each battalion wrote a brief summary of each day’s events in a War Diary: these usually state where each unit was and may give details of the actions they were involved in.

Three of the men could be identified without difficulty from the CWGC website.

Horace Charles Davis

Lance-Corporal, 21st Battalion, The London Regiment (1st Surrey Rifles). Service no. 1786. According to his War Service record, Horace Davis joined up on 7 August 1914, one of many who volunteered in the wave of patriotic enthusiasm that swept the country immediately after the declaration of war on 4 August. He gave his age as 17 years and 4 months, and his occupation as Catalogue Assistant at the British Museum. He was the son of Edwin and Maria Davis of Chryssell Road, Brixton. He died on 15 September 1916, aged 18, at the battle of Flers-Courcelette and is buried in Warlencourt Military Cemetery, on the road between Bapaume and Albert, Pas-de-Calais. (If the age on his gravestone is correct, he would only have been 16 when he joined up.)

The 1st Surrey Rifles moved to the Somme front in July 1916, and on 10 September joined the line at High Wood. As part of the battle of Flers-Courcelette they moved into Mametz Wood on 15 September and then attacked a German trench, the Starfish Line. According to the War Diary:
4.45pm. Battn advanced in artillery formation to the attack with a fighting strength of 19 Officers and 550 ORs [Other Ranks]. Arrangements could not be made for artillery support or adequate covering fire, and as the leading platoons came under observation they were subjected to an intense artillery bombardment and later to heavy rifle and machine gun fire. The casualties in this advance amounted to 17 Officers and 490 other ranks, of whom a large percentage must have been killed by heavy shells. The remaining officers and a few NCOs and men dug themselves in and held on to what ground they could occupy until the Battn was ordered to withdraw at 7.30am on the 16th.

This was the first battle in which tanks were used, though in a different part of the field.

2

Horace Davis’s gravestone in Warlencourt Military Cemetery.

Thomas Frank Wickham

Rifleman, 1st Battalion, The Rifle Brigade. Service no. B/203615. Son of Thomas and Elizabeth Wickham of Twisden Road, Kentish Town. Died 23 June 1917, aged 41. No known grave, but commemorated on the Arras Memorial, which honours those who died in the Arras sector between Spring 1916 and August 1918.

In June 1917 the 1st Battalion was operating east of Arras, south of the River Scarpe. The War Diary states:
June 22/23. During the night of 22/23 Battalion HQ moved to [map reference] H30 d2.5 in bank near Lone Copse [south of Fampoux and west of Pelves]. I Coy moved up to Welford Trench.
June 23. At 1030pm B Coy made a raid on the enemy’s trenches in I25 b, resulting in the capture of 1 officer and 6 ORs, and accounting for about 35 others.

The account is incomplete but it is possible that Wickham was killed in hand-to-hand fighting during this raid on the German trenches, which would explain why his body could not be recovered.

3

Thomas Wickham’s name appears on one of the huge walls of the Arras Memorial.

4

The Arras Memorial lists the names of 34,785 men who have no known grave.

Alfred George Williams

Private, 18th Battalion, The Middlesex Regiment (1st Public Works Pioneers). Service no. 204113. Son of Alfred and Charlotte Williams of Eresby Road, Kilburn. Born in Willesden on 20 December 1897 and worked in the Book Department at the British Museum. Joined up on 28 June 1915, giving his age as 19 though he was only 17 years and 9 months. Died of wounds 11 October 1918, aged 20. Buried in Etaples Military Cemetery, Pas-de-Calais.

The Pioneers provided support services to the front line, repairing trenches, tunneling, laying rail tracks, revetting canals etc. Many of the men were miners or had experience in heavy construction. In September 1918 the 18th Battalion was working west of the Canal de St Quentin, between Cambrai and St Quentin, making roads passable and demolishing damaged bridges. The War Diary gives little information, but records that on 26 September the battalion was working on the 33rd Divisional HQ and approaches, and on the preparation of the Etricourt – Heudicourt horse transport tracks. 2nd Lieut. Howell and four men were wounded and eight killed. Williams was seriously wounded in the left arm, left leg and chest and was evacuated to the 3rd Canadian General Hospital in Dannes-Camiers (between Etaples and Boulogne) where he died on 11 October.

DSCF2782

Alfred Williams’ gravestone in Etaples Military Cemetery.

6

General view of Etaples Military Cemetery.

This leaves us with Horace Crawley. Nobody of this name appears in the CWGC database; the only casualties named Crawley from the 6th Battalion, London Regiment are T H Crawley (no first names or age given) and Ernest Victor Crawley. Further searching revealed that T H Crawley is commemorated on the Mill Hill War Memorial, and the related website suggests that this was probably Thomas Horace Crawley, who is also commemorated on the Crawley family gravestone in the churchyard of St Mary’s Catholic Church, Kensal Green. His date of death matches that in the CWGC database, so we can be reasonably certain that this is the right person. However, his parents lived in Barnsbury, so it is not clear what links Kensal Green, Mill Hill and Barnsbury.

T(homas) H(orace) Crawley

Rifleman, 6th Battalion, The London Regiment (City of London Rifles). Service no. 3396. Son of Horace and Bridget Crawley of Barnsbury. Died 7 April 1916, aged 19. Buried in Hersin Communal Cemetery, 5km south of Bethune, Pas-de-Calais.

The 6th Battalion, London Regiment was known as the Printers’ Battalion because many of its members were recruited from Eyre & Spottiswoode’s printing works. Early officers included George and William Spottiswoode. The battalion was stationed near Vimy Ridge from March 1916, and according to the War Diary:
6th April. Battn in billets at Villers au Bois [about 5km NW of Arras]. Baths at the disposal of the Battn from 8am to 12 noon and from 2pm to 6pm. All ranks bathed according to orders issued. All ranks washed and greased their feet before going into the trenches. The major, adjutant and Company Commanders went to reconnoitre the left sub-section, Carency section prior to taking over. 5 officers and 250 ORs detailed to work under REs [Royal Engineers] on Cabaret Rouge. Lieut Col M A Mitchell CMG returned from England.
April 7th. Battn in billets at Villers au Bois. Battn paraded at 9.45pm for the purpose of proceeding to the trenches, left sub-section, to relieve the 17th Battn London Regt. Owing to heavy shelling by the enemy of our communication trenches relief was delayed for an hour. Order of relief, Front Line C Coy, Quarries B Coy, A Coy in cellars at Ablain St Nazaire, D Coy in dugouts, Sunken Road, Ablain St Nazaire. Relief complete at 1.25am. 1 officer and 60 ORs remained at Villers au Bois as a permanent working party (under RE).
April 8th. Battn in the trenches. There was good deal of sniping activity immediately after dawn. The enemy shelled Cabaret Rouge with shrapnel. 1 officer and 60 ORs wounded.

It is possible that Crawley was killed by the shelling of the communication trenches or by a sniper. He was presumably known to his colleagues in the Bindery as Horace rather than Thomas, which is why that name appears on the memorial.

7
8

Horace Crawley is buried in the British section of Hersin Communal Cemetery.

9

The Crawley family grave in Kensal Green (from the Gravestone Photographic Resource).

From this gravestone we can see the sadness of the Crawley family: Horace’s younger brother Joseph died when he was only 3 and Horace was killed when he was 19. Mr Crawley died in 1929 and Mrs Crawley lived on until 1943.

So there we have it: three young men who joined up, presumably thinking that fighting would be more exciting than book-binding, and an older man who certainly did not have to fight (conscription for men up to the age of 41 was not introduced until January 1916), but who maybe volunteered out of a sense of duty. We salute them.


Sources

Commonwealth War Graves Commission: http://www.cwgc.org/find-war-dead.aspx

Mill Hill War Memorial: http://www.roll-of-honour.com/Middlesex/MillHill.html

Gravestone Photographic Resource: http://www.gravestonephotos.com/

Alfred Williams: see Mike Hall, A Miners’ Pals Battalion at War, vol. 2, pp 273-274. Kibworth Beauchamp: Troubadour Publishing 2015. ISBN 978 1784620349

The National Archives
War Service records: file series WO 363 and WO 364
War Diaries:
6th Battn, London Regt. WO 95/2729/2
21st Battn, London Regt. WO 95/2732/5
18th Battn, Middlesex Regt. WO 95/2417/5
1st Battn, Rifle Brigade. WO 95/1496/4


Dr Barry Knight