Sewing Models

Over the two-year program at NBSS, we make a lot of blank books as models that inform our understanding of the book as a moving system. They give us the opportunity to try out different materials and structures to see how they operate. Most of our models are the same size and use the same textblock paper (which is never true for printed books), and we often vary the sewing supports, thickness of thread, and spine linings for each style of binding. This makes it a bit difficult to make exact comparisons of how those variables affect the book action. One interesting (and quick) project that we did at the end of the first year was designed specifically to compare book action for different sewing structures.

This project is inspired by Tom Conroy’s article titled The Movement of the Book Spine (1987). [Disclaimer: Tom actually taught this as a workshop some years back – both here at the school and other places – so these are based on his design, but may not be exactly as he teaches it.] Conroy’s article attempts to isolate and discuss the variables that affect the action of a codex with a rounded spine, including the supports and sewing, height and shape of the round in the spine, and linings. The introduction suggests that one make a series of binding models “in which only one variable at a time is altered” in order to compare them. Obviously that would be a huge number of models, so our project attempted to do so for just the sewing structures.

Over the course of a class day, we made a series of extremely rough, tight-back leather bindings. Each binding was composed of the same number of sections, each with the same number of folios of Mohawk paper. The same thread and number of supports were also used for each, but the type of support and style of sewing were varied.

Two basic types of linen sewing supports were used: German tapes and 6-ply cord. One book block was sewn with 2-hole lap sewing over the tapes (above, top). The next was sewn with loops over single cords (below, top). A third was sewn with packed sewing over single cords (below, bottom). The fourth textblock was sewn with a herringbone linkstitch over double 6-ply cords (above, bottom).

Conroy (1987) notes that the tension of the supports will affect both the shape of the spine and the book action (p. 15).  Too little tension will leave the book loose and somewhat spongy – causing the boards to skew. Too much tension will put the supports are constant strain when the book is rounded and backed. The effect is more pronounced for thongs (animal) than cords (vegetable), as the cord is less elastic (Conroy, p. 15). All sewing was done on a sewing frame in order to achieve proper tension on the supports. The books were also all sewn in one sitting, compressing each section with the bonefolder as we went, in order to be as consistent as possible.

The spines were all glued-up with hide glue and rounded and backed to get a 90 degree shoulder. Davey board was cut to the size of the textblock.

Conroy classifies the spine linings and covering material of the books as either “tension” or “compression” layers (p. 4). When the book is open, layers adhered directly to the spine will be put under tension, while those further out will be compressed (See Conroy’s paper for diagrams). We often line the spines of our books with layers of textile, paper, and/or leather to achieve the right opening for the size of the book and drape of the paper. Paper (in general) produces a much stiffer spine opening, while leather and cloth are more flexible. In order to simplify the spine lining variable, we finished these bindings as tight-backs with leather as the only spine lining.

Small squares of full-thickness leather were cut for each book, dampened, and pasted out with wheat starch paste. After letting the paste soak in for a couple of minutes, the paste was gently scraped off and a new coat applied. A square of leather was then applied to the head and tail of each book, so that it covered all but the center sewing support. The result is kind of like a cut-away model, allowing you to easily see the sewing structure of each book. Note that there are no headcaps; The leather is cut flush to the boards and textblock, and the boards are not back-cornered. Like I said, these are models are quick!

After the leather was applied, the books were tied up in the usual way. A long piece of cord is wrapped on either side of the bands immediately after covering to keep the leather from pulling away as it dried. For the book sewn on double cords, a third wrap was made across the center of each band (see above).  We made some rudimentary tying up boards out of scraps of binders board in order to keep the cord from marking the leather that goes across the face of the board.

The books were allowed to dry and opened the following day. The pictures that follow attempt to show the different openings for each book. While the results are not so visually dramatic, variations in spine flexibility are easily felt through handling. Note the shape of the spine on each.

The spine of the book sewn on tapes throws up a great deal and opens in a “V” shape.

The book below is sewn on single raised cords (unpacked sewing). It’s spine has more of a “U” shape from the stiffer support, but still exhibits high throw-up.

The book sewn on single packed cords, however, has a very different shape from the one above. The packing of the sewing essentially creates a thicker, stiffer support, reducing the throw-up of the textblock.

Finally, the book sewn on double cords has the least amount of throw-up.

These pictures also illustrate how the movement of the spine affects the leaves. As Mohawk is a pretty stiff paper, less throw up from the spine keeps the pages from lying flat.

Essentially, these bindings show that increasing the diameter (profile) of the sewing support will make it stiffer and reduce the throw-up of the textblock (compare tapes to single raised cord sewing). Also, packing the sewing creates a thicker and stiffer support (compare single cords to single packed cords). Increasing the number of sewing supports makes the opening stiffer as well (compare single to double raised cords). I really enjoyed Conroy’s discussion and many diagrams that illustrate these concepts (p. 10).

But why, you may ask, do we spend all this time thinking about the subtle interactions of the materials that comprise a book spine? A book must function in order to be a book. If it doesn’t open, it is essentially a block of paper; if the sewing and adhesive fail, then it is basically a pile of loose sheets. Books that open well, without creating undue strain on the text or covering materials, are more enjoyable to use and will last longer. Whether creating a new binding or repairing a damaged one, by manipulating the sewing, supports, or linings, one can create a customized book action that is sympathetic to the materials. That is just one of the beauties of a handmade book.


I’ve just shown you some of the roughest looking leather bindings you can make, so next in my next post I’ll go in the complete opposite direction with some pictures of my first French-style fine binding.


ConroyTom. (1987). “The Movement of the Book Spine”. AIC Book and Paper Group Annual, 6, 1-30.

Mention in Felt & Wire

Alyson Kuhn wrote a short feature on NBSS and mentioned a recent project of mine. I’ve been promising a post on springbacks for a while now, after rebinding a few ledgers this year. That post is coming – but, as a teaser, ‘Before’ and ‘After’ shots of one of the books appear in the article.

You can read that story here.

Limp Leather Bindings

In this post I’m going to discuss limp leather case bindings. This style is  perfect for rebinding books of a specific dimension – mainly flexible, pocket-sized volumes (with no shoulder) like language dictionaries and some versions of the New Testament. These case bindings fall at the end of the first year in the NBSS curriculum, since they require a great deal precise leather paring, but no hot tooling.

For my first, I started out with a damaged copy of Kipling’s Kim that I found at the Brattle Book Shop. The front board was missing entirely and the red leather was extensively red rotted.

Very little of the gold decoration on the spine remained. The back board was also completely detached.

This volume came from a set, published by MacMillan and Co. in the early 20th century and probably originally looked like this one:

My time-ravaged copy included a rather interesting addition, however. Tucked inside the pages, I found this very brittle envelope.

The envelope contained a small silver gelatin print of Zamzama (also known as Kim’s Gun).

This edition is not difficult to find or very valuable, so I had no qualms with rebinding it. I discarded the remaining board and mechanically removed the leather from the spine after softening the adhesive with a methyl cellulose poultice. The blank flyleaves from the old endpapers were skinned off and also discarded. The original sewing was still in good shape, so I left it intact. New single-folio endsheets of Hahnemuhle Gutenberg were tipped on at the front and back. The spine was then re-lined with an aero cotton extended lining.

These are case bindings, so the cover is made off the book. The boards are simply 20pt Bristol board, with the hard edges and corners taken down using sandpaper. There are no endbands and the squares are very small. You can basically cut the boards to the width of the textblock – so that when they are pushed out slightly from the shoulder of the textblock to make a joint space, a small square is created. With the boards in place on the book, a scrap of paper is used to measure the spine and joint width. That measurement and the placement of the boards is transferred to the flesh side of the leather with a grease pencil or china marker. I also cut a spine piece out of thicker paper (like Dove Gray) that is the width of the textblock spine and the height of the boards.

We used goat skin to cover these volumes. The majority of the paring was first done in the Scharf-fix, flat paring the skin down to about 0.5 mm. The remainder of the paring was done by hand.

The paring for a limp leather bindings is quite a challenge; any irregularities in the paring can be seen and felt through the thin boards and across the turn-in areas of the spine. With the paring knife, all four sides of the leather are pared so that the turn-ins are a long bevel that goes down to absolutely nothing. The length and angle of the bevel should be gradual enough that a change in thickness is imperceptible when running your fingers across the leather. The headcap areas are cut at a gentle crescent, but also have a smooth bevel that goes down to nothing. While paring, I constantly check the consistency of the paring by laying the leather (grain-side up) on the litho stone and running my fingers across it. If any bumps or ridges are detected, I go back and smooth them out with the knife. Similarly, I will sometimes fold the pared area in half and run it between my thumb and index finger to feel for irregularities. This process can take a long time to get right the first couple of times.

When the leather finally feels right, it is dampened from the grain side, flipped over, and pasted out. While the paste is soaking in, the textblock is wrapped in plastic wrap (“cling film” for anyone in the UK). The first layer of paste is scraped off the leather with scrap board and a new, thin coat of paste applied. The spine piece and boards are then placed onto the leather according to the marks. At this point I’ll put it on the book and check to make sure everything looks right. Adjustments can be easily made at this point.

The next step is to do the corners – which we did in the English style. The leather is mitered at a 45 degree angle, cutting down at about 30 degrees, and about a board thickness and a half away from the board corner. The leather is then scooped out a bit right at the corner to make it thin enough to pleat. Now the head and tail turn-ins are done. The very tips of the corners are pleated down, and the fore-edge turn-ins are done – being careful to match up the miters of the corner so that they make a smooth surface on the inside of the boards.  The finished cover is then wrapped around the textblock to dry.

The plastic wrap keeps the moisture from the leather from penetrating the textblock and warping the paper. The damp leather is extremely easy to mark up at this stage, so the whole package is then wrapped in felt and placed under a light weight to dry.

I left mine to sit overnight – but, depending on the environmental conditions, the cover will probably be dry in a couple of hours. When it’s ready, it can be decorated. I did not know how to do any tooling at this point, so I titled it in the Kwik Print.

This “terracotta” goatskin from Harmatan colors very nicely, so I did the stamping in blind rather than with foil. I built a little jig out of binders board to hold the cover in place on the platen of the Kwik print. I then dampened the leather a bit and with the temperature pretty low on the hot stamp, I would make a quick impression and see how much color came out. I would then dampen a bit more and repeat, being careful not to burn or blacken the leather, until an even impression was achieved.

These bindings also typically have a little blind rule that goes around the head, tail and fore-edge of the boards.

These lines were actually done with a bone folder. After gently marking out the lines with dividers and the bone folder on dry leather, the leather is evenly dampened with a cotton ball. The line is then marked again with the tip of the folder. The moisture and pressure are enough to make a permanent, dark line in the skin. This is the result.

The minimal spine lining gives the book a very flexible opening.

I also used this structure to rebind a small, well-used New Testament, that I believe was a gift as part of the client’s confirmation.

The original imitation leather cover and endsheets had become quite stiff and had cracked in several places. Once again, the sewing was intact and only minor page repairs were required.

The greatest thing about this being a case binding, is the ease with which one can title or decorate the new covers. Here I was able to reproduce the titling and name stamping in gold foil, once again, using with the Kwik Print.

This is a project that Jeff just demonstrated one afternoon, and I’m not sure if it can be attributed to anyone in particular. I have not seen any articles or handouts on this particular structure, so I have no other resources to point to for further reading. If you have encountered other instructions on making this structure, please leave a comment.


I’m working through these as quickly as I can.  Basically I have loads of pictures of different finished projects – I just have to find the time to sit down and write about them. Thank you for being patient. Upcoming posts: Sewing models, atlas and album structures, springback account books, parchment over boards, 18th century trade binding, rounded spine leather box, and flag books!


Time Travel

Today I’m going to share a beastly, strange, Frankenstein of a book that I did for the “Time Travelers” show at the Asheville Bookworks. You can see an online gallery of the books in that exhibit here.

The time traveling theme was meant to give artists the opportunity to illustrate and respond to the various ways that book structures and materials have changed over time and across geographic regions. I decided to take the prompt a little more literately and ask what would happen to the book itself  if it traveled back in time.

There is a lot of great literature out there about time travel; The Time Machine by H. G. Wells or A Connecticut Yankee in King Arthur’s Court  by Mark Twain immediately spring to mind when I think of stories that follow this theme. I decided, however, that the high points of the genre have probably received too much attention already and that it would be a lot more fun to focus on one of the not-so-great stories. You know – one that nobody would spend time or materials rebinding. Maybe one of those titles that you often find in library book sales or your local thrift shop. Maybe even one that was adapted into video game and a feature film starring Paul Walker and Gerard Butler.

That’s right – I’m talking about Micheal Crichton’s 1999 book Timeline.

My submission to the show is rebound in the Carolingian style, fully covered in deer skin with a brass fastener (made by Kait Kennedy, a jewelry maker here at NBSS).

In the story, a group of history students travel back to medieval France to rescue their professor, Dr. Edward Johnston. After the their teacher disappears, the students find a  600 year old note from Johnston buried in the ruins of a castle and hijinks ensue. I thought it would be fun to make the book look like it had traveled back in time with the characters, but never made the return trip. I also like the juxtaposition of a machine-made, mass-produced piece of popular fiction in a binding from a time in which books were handmade, incredibly expensive, and unique. Technically, the story takes place in the 14th century, and Carolingian bindings are produced from the late eighth to the end of the twelfth centuries (Szirmai, 2000, p. 100) – but I’ve wanted to try one of these bindings for a while, so let’s just ignore the fact that this should be more of a Romanesque binding.

So a bit about the structure of Carolingian bindings, with the caveat that I have absolutely no experience with historical examples. While I’ve examined a hand-full of manuscripts from this period, none were in their original bindings. All my information comes from Szirmai’s chapter. These bindings are a kind of structural transition between the squared-up, leather-covered Ethiopian structures (unsupported chain sewing that also functions as the primary board attachment) and the rounded spine Romanesque  and Gothic bindings (sewing on supports that are laced into heavy wooden boards).

As I said before, this is a Frankenstein of a book – not only content-wise, but because of the way I had to put it together. As many bookbinders know, it can be difficult to find a printing of contemporary literature that is actually sewn folios. Most are either single, adhesive bound sheets, or were folded sections that have been sawed open at the spine for better adhesive penetration. My copy was glued single sheets, so I had to reform the text into sections. Sections can be made by guarding the leaves together with Japanese tissue, but this method often creates a significant amount of swell at the spine of the book. Swell is good if the book is going to be rounded and backed, but because Carolingian bindings have flat spines, I needed absolutely minimal swell. I decided to use a sewing machine to assemble sections.

This method is a little bit crazy and I wouldn’t recommend it for a book that is meant to last – but it is quick and gets the job done. It might depend upon the model of sewing machine that you use, but the one that I used pierced the sewing holes very close together at the widest setting. Doing so effectively perforates the sheet, and since the adhesive probably isn’t going to be penetrating that far, you are creating a good opportunity for mechanical detachment of single leaves later on. Whip-stitching by hand, a common technique used in atlas bindings, would probably be stronger. But this book is for show, not to be read!

Traditionally a binder would probably prepare the boards as their first step (Szirmai, p. 103), then, after sewing, trim the textblock to the size of the boards with a drawknife (p. 119). I trimmed all the sections down in the board sheer to give a kind of rough-cut edge and then prepared the boards. Szirmai reports that a range of wood is used for the boards of extant Carolingian bindings; most are oak, but beech, fruit tree, and poplar are also found (p. 103). Unfortunately, I couldn’t get quarter-sawn oak boards that were large enough in time for the exhibit deadlines, so I had to fake some up by laminating layers of 1/4″ plywood and millboard.

The result was a super-thick board (around 1/2″) that was heavy and hard like wood, but was extremely easy to drill and shape. Szirmai states that the thichness of Carolingian boards (with leather covering) ranges from 7 to 17mm (p. 103). After marking up the board and drilling angled holes for the lacing, channels are cut between the holes so that the laced cord doesn’t sit proud of the board surface. Szirmai provides a range of diagrams on the variations of lacing patterns through the board (p. 107).
 Szirmai (2000) states that cord was used primarily as sewing supports in the Carolingian bindings surveyed, but that white leather thongs are also found in books from France and Italy (p. 112). An extra long length of cord is laced through the front board, then tied up onto the the sewing frame. The sections are then sewn through the “center” around the double supports in a ‘herringbone’ pattern (Szirmai, p. 101).

After sewing, the second board is laced on. Traditionally, the free end of the cord is secured into the lower board with wooden wedges or dowels (Szirmai, p. 103). I didn’t have much faith in the strength of my boards, so I just frayed out the cord ends, fanned them out with adhesive, and hammered them into the interior surface of the board. The pastedowns are typically put down onto the boards before covering (Szirmai, p. 117).

Headbands are sewn either in the coptic style or around double cords that are laced into the boards (Szirmai, p. 121). A distinguishing feature of these bindings is the leather tabs that extend over the head and tail of the book. Shelving books upright was not a common practice during this time, and books were often stored in chests. The tabs facilitate lifting the volume up and out of it’s enclosure.

When the endband thread is anchored into the secitons, it passes through one layer of the leather tab. I used the same deer skin for the tab that would be used for the covering… this skin was odd to work with, but more on that later. Here is the endband from the back, showing the anchoring. .

It looks very rough, but after covering and doing the parimeter sewing to keep the parts of the tab together, the result is quite nice and super strong. Szirmai has excellent diagrams of several variations of tab construction (p. 125).

After cutting a small channel in the upper board for my fastener, the book was covered. Szirmai states that the majority of extant Carolingian bindings were covered in thick (2mm) ‘chamois’ or buckskin (species of origin unknown, p. 127). The skins of local wild animals were used, as Charlemagne’s edict of AD 774 gave the Abby of St. Denis permission to use deer skin their forest for bindings (Szirmai, p. 127). The deer skin that I used was actually donated to the school (along with a variety of odds and ends) some months ago, so I do not know from whence it came or how it was tanned. It was quite scarred, stained, and difficult to pare, however. Perfect for making a medieval book!

The covering leather is adhered directly to the spine, which actually produces a nice opening for the book.

The thickness of the boards meant that the turn-ins have to be left extra long and made doing the corners a bit of a challenge. Ordinarily, a leather corner can be pared quite thin and pleated along the inside of the board to fully cover the corner and deal with excess material. The toughness of the deerskin made this difficult, so I cut the leather as one would do a cloth corner. On the inside of the boards, I mitered the leather and stitched it – as diagrammed in Szirmai, Fig. 7.29 (p. 129).

Szirmai reports that the majority of surveyed bindings from this period are left untrimmed at the turn-ins (p. 129).

As for fasteners, Szirmai says that they were a regular feature on carolingian bindings, but that few of them have survived; all that remain are stubs of leather straps and holes from pins (p. 131). The author includes drawings of a few examples  from the Abby of St. Gall with extant rings and pins that I modeled mine after.

The pin is just hammered directly into the lower board. The ring is attached by a looped piece of calfskin (stitched with a strip of parchment) and secured to the upper board with brass escutcheon pins.

As I said before, these books were not stored in the way that we are accustomed to shelving books, so I inked the title onto the fore-edge. I have not read all that much on the titling of books, so I do not know how titling or distinctive markings were made during this period. If you have suggestions for reading on the topic, please post them in the comments.


I think I’m going to try to put up a post on sewing models next. Look for it soon!


Szirmai, J. A. (2000). The archaeology of medieval bookbinding. Aldershot : Ashgate.

NBSS on Wikipedia

One of the projects that has occupied my writing time as of late has been the creation of a Wikipedia article for the North Bennet Street School. Surprisingly, one had not been attempted before.

I wrote this article using Sarah Henry and Mary A. Williams’ book, titled North Bennet Street School: A short history 1885-1985, as a source for the history of the building and organization. It’s actually a very interesting read and includes photos of some of the earliest students and instructors.

This is my first ever contribution to Wikipedia, so I am rather pleased to have it up. I am also interested to see how the article develops as other users make changes and add information.

Millimeter and Rubow Bindings

Caution: I have had one of the craziest weeks of my life – so I apologize if this post is disjointed or rambling.

The next binding that I am going to share is actually the first binding project that was assigned in the second year. Students at NBSS are not given a binding to do for the first couple of months of their second year, so that there is no distraction from practicing gold tooling. It’s probably a wise choice – as I was looking for anything else to work on after the first couple of days of tooling.

The millimeter binding is a very modern and elegant adaptation of the paper case binding, well suited to thin volumes. It is also a logical next step in the curriculum because it is so similar to the leather bindings done in the first year (here). I’m going to discuss two variations: the millimeter binding and the Rubow binding.

According to John Hyltoft, former Head of Conservation for the Smithsonian Institute Libraries, the millimeter binding style was developed by the Danish binder Henrik Park during WWII. At a time when luxury items like leather were at a premium, there was a place for finely bound volumes that used leather as sparingly as possible. These bindings are characterized by leather (or parchment or even cloth) that covers the spine and extends about a millimeter onto the boards. Small “hidden” or “semi-hidden” corners are also quite popular.

Park’s book from 1949, titled Moderne Danske Bindtyper, is a short volume with nice photographs of several millimeter binding styles, including detailed photos of a variety of corner and headcap styles.

Other variations on the millimeter include one by August Sandgren, in which small strips of leather appear at only the head and tail of the spine and extend a millimeter onto the boards (Park, Bindtype I). Two variations that are not pictured in Park, that are probably more modern adaptations include a kind of T shape in leather at the head of the spine, with small arms going onto the boards, and doing a full strip of leather along the fore-edges of the boards instead of corners. The “Bind-O-Rama” from  2005 (found here) has images of a range of  bindings that encompass the variations on the style. Don Rash’s book is definitely my favorite of them.

The “Rubow” binding, supposedly suggested by Jorn Rubow (maybe this Jorn Rubow?) features leather all along the head and tail edges of the boards. Park’s description of this binding, “papirbind med slidskinner”, fittingly translates to paper binding with wear rails.

Some common features of the books include a single-folio endsheet reinforced with a Japanese paper or textile hinge, sewing on flattened cords, 90 degree shoulders with open joints, rolled leather or cloth endbands, and edge decoration (Riley, 2006; Verheyen, 2005). In reality, these are an economical form of fine binding.

The literature presents two methods for doing this binding: either as an “on-set boards” binding in which the boards are attached to a waste sheet built into the endpapers before covering (Riley’s method), or as a simple case structure that is covered before attaching it to the bookblock (Verheyen’s method).

The endsheet construction will depend upon the method of assembly. For the on-set boards version, the endsheet is composed of a single folio of decorative paper with a 2cm strip of mull tipped to the outer edge. A folio of  plain paper is then wrapped around, forming a plain flyleaf and wastesheet. This whole assembly is then folded 5mm away from the edge to make a hook, which is wrapped around the first or last section of the text (Riley, 2006, p. 10).

The endsheet for the case version is either a single folio of decorative paper tipped to the first or last section of the text or a double folio that is sewn on (Verheyen, 2005, p. 25). Regardless of the number of folios, Verheyen advocates adding a guard of Japanese paper or thin cloth around the first section and endsheet.


The forwarding of the book is basically a more refined version of what I have described in the past concerning case bindings or on-set boards bindings. I will note, however that according to Riley, it is only the head edges of these books that are typically trimmed and decorated (p. 12). Solid color, sprinkling, graphite or gilding are options for edge treatment. I did a solid graphite edge for one and a graphite and gauffered edge for another.

If endbands are present in millimeter bindings, they are almost always rolled leather endbands. The spine lining is done much the same way that I have described before – patch lining and sanding to get a very smooth, but rigid, spine.

The leather covering is done first, and, for this style of binding to look right, the edges of the leather cannot be visible under the paper covering. Once the leather is covered over, any sudden changes in transition from leather to board surface will be very apparent and look sloppy.

There are two methods for dealing with the thickness of the leather. The first is to flat pare the leather, cover, then in-fill the remaining space on the boards before covering with decorative paper (Verheyen, p. 28). The second method is to edge pare the leather, cover, then further pare and sand the leather to achieve a long, gradual bevel onto the board surface (Riley, p. 15).

Regardless of how you deal with the leather, it is pared quite thin (0.5mm). Covering is done in the usual way with paste. I found the Rubow covering to be quite challenging, however, because so little leather is holding it onto the boards while you are doing the turn-ins and headcaps.

If following Riley’s method of beveling the leather, the  millimeter of skin that will be exposed is marked out with dividers and the edges further pared at a shallow angle with the knife. The edge is also sanded to get a very smooth surface. The leather patches at the corners are done at this point and sanded smooth in the same way.

To finish, the boards are covered in decorative paper and the endsheets pasted down. Paste papers are quite common in the examples that I’ve seen.

This binding style is actually quite fun to do and the results look very refined. I highly recommend Riley and Verheyen’s articles for step by step instructions. John Hyltoft also demonstrated this binding at the GBW Standards conference in 1995. You can purchase a video of that demonstration here.


Application deadlines are starting to loom and I have to sort out what I’m doing after I graduate in June – but I will try to keep posting the vast array of projects that I have to share on a regular basis. Here is a short list of what is in the queue: Carolingian binding, atlas and album structures, book action/sewing support models, limp leather, and springbacks.


Park, H. (1950). Modern Danish Bookbindings. Copenhagen, Anker Kysters.

Riley, D. (2006). The Millimetre Binding: Danish economic structure. Bookbinder, 20, 9-17.

Verheyen, P. D. (2005). “Millimeter Binding / Edelpappband” The Bonefolder: an e-journal for the bookbinder and book artist, 1(2), 25-29.

Limp Paper and Vellum

In the last post, we saw a paper binding that was basically a glorified wrapper, attached to the textblock by adhesive. This time I am going to look at a structure with earlier roots, that relies mostly on mechanical attachment of the cover to the book block. This binding structure is also interesting in that the textblock and cover are prepared separately, establishing it as a precursor to case binding.

Right after our class covered lapped component bindings and stiffened paper cases, we made some laced paper case models. I feel like I should start this one off with a few disclaimers. I have very little experience with historical examples of laced paper bindings – either limp or in-boards. I have not yet read all that much about their historical context, evolution, or typical features. Therefore, this discussion will focus on only one particular structure that, while based on the binding techniques of antiquity, is heavily influenced by modern conservation practice. While I was making these paper-covered models, I went ahead and made a limp vellum binding using the same structure. There is significantly more literature available on limp parchment structures, so most of this post is actually derived from those sources. Therefore, I ask you (the gentle reader) to forgive the limitations of this discussion and its deviation from my paper-centric series.

Now on to the topic at hand!

Most bookbinders and conservators will naturally recall the structure pictured above when the term ‘limp vellum’ arises in conversation; however, Lindsay (1991) reminds us that it is a very general term that covers a wide variety of structures (p. 4). Szirmai (2000) mentions limp parchment bindings from as early as the Carolingian era (p. 286), but both he and Clarkson (2005) conclude that their popularity in the thirteenth and fourteenth centuries grew as a result of growing demand for book supply. These sturdy bindings played an important role as an economical and utilitarian alternative to stiff-board bindings in early universities, as well as library and archive collections.
Szirmai (2000) describes a variety of structures that rely on different book block to cover attachment methods. In previous blog entries, I’ve covered a few limp structures in which the cover attachment is through primary sewing (here). There are also a number of styles that make use of primary and secondary tackets for this attachment. Many of these early limp structures feature a  flat spine that, as we saw in the last post, has a tendency to develop a distorted concave shape.  Binders in antiquity resorted to a number of methods to try to get the book to stay square.  In some cases, the binder would sew through a rigid spine plate (discussed here). Binders were also known, on occasion, to lash wooden sticks or metal rods onto hidden interior sewing supports and then attach the covers with indirect tacketing (see Szirmai, 2000, p. 305 for diagrams). Because the opening of the volume – especially if large – puts a significant strain on the spine, these efforts were largely unsuccessful. Wooden supports would often snap and metal support rods would actually bend to further reinforce the concave round in the spine (Szirmai, p. 314).
These flat-back limp structures gradually give way to bindings with “moderately rounded spines, laced-in endband supports and parchment covers with neat turn-ins and pastedowns (Szirmai, 2000, p. 317). The exact origin of limp cover attachment to the book block with laced-in sewing supports is unknown, but the better known examples date from late fifteenth and sixteenth centuries (Clarkson, 2005; Szirmai, 2000).  Whether the rounding of the spine developed intentionally or as a natural result of the lacing process, is unclear. Regardless, a rounded spine does tend to keep its shape over the life of the book and is a far better solution to textblock bulge than stiff plates or rods.
Several features of the laced models of limp bindings seem common from the literature; most notably hooked endsheet construction, sewing on cords or thongs, and thick parchment covers with two or three slits at the joint for lacing. Rhein (1942) describes incunabula from Amplonian Collection at Erfurt as having single, reused parchment leaves hooked around the outer sections. Clarkson (2005) includes descriptions of several interesting endsheet structures for conservation rebinding; including of a hooked endsheet with “expanding gusset” to relieve unwanted tension in pasted down endsheets, as well as a parchment leaf with alum-tawed hinge for non-adhesive case attachment (p. 12-14). My models all used non-adhesive endsheet attachment (discussed below). I used either single leaves of heavy handmade paper from the University of Iowa or a single folio section (like in the previous posts) for my paper case models.
Szirmai (2000) reports that sewing  was done all along, and occasionally packed around the supports (p. 312). The literature describes two methods of cover attachment: either single supports, laced through in a straight line across the cover or double (or split) supports that are each laced through at an angle, forming a ‘V’ pattern (see Clarkson, 2005). All of my models are sewn on single alum-tawed goatskin thongs as you would tapes.

While modern sewn endbands are merely a decorative feature, late medieval and Renaissance endbands had a structural component. Primary sewing firmly attached the core to the book block through tie-downs in the center of each section. The endband sewing was often done through the spine lining (as above). In the Italian volumes that were commonly used in libraries and archives, endbands are plain, with a back bead. The endband cores were laced into the covers, while the remaining sewing supports were “stubbed to the width of the textblock” (Clarkson, 2005, p. 16). The German variety from the same period went in the opposite direction, however, often have no endbands at all, but laced sewing supports (Szirmai, p. 315). Both Lindsay (1991) and Greenfield & Hill (1990) have excellent instructions for making this endband.

The next step in making this structure is to prepare the cover. Lindsay’s (1991) article does an excellent job of describing the formation of a modern cover that uses cuts and tabs for the mechanical attachment of the turn-ins at the corners. Clarkson (2005) includes diagrams of two other versions with  interlocking tabs. Historically, I think the turn-ins were sometimes just pasted down. As the cuts are somewhat complex and vellum is an expensive material to waste, Lindsay recommends making a paper template first (above). I used 10 pt folder stock for mine. All the measurements are taken directly from the textblock, resulting in a cover that is an exact fit. The vellum is cut out according to the paper template; Lindsay recommends cutting windows out of the paper template, so that the best part of the skin may be chosen.

The sewing supports are first laced out through the holes at the point of the shoulder. These holes are much smaller than the width of the supports, so that the thongs are held very securely after lacing.

When the sewing supports are laced back through to the inside of the cover, they also pass through the extended spine lining and endsheet material.

After trimming them back a bit, they are hidden underneath the outermost leaf, which is tucked under the turn-ins. The lacing of the endband cores and the fasteners keeps the whole assembly firmly in place.

The yapp edge and fore-edge fasteners are a distinctive and charming feature of this binding style. They serve to keep the book in a nice, self-contained package and protect the textblock from drops or bumps. Szirmai (2000) indicates that most fastenings were historically chamois leather, but that metal fasteners are sometimes encountered. Clarkson (2005) states that many types of fore-edge tie anchorage designs were in use by the sixteenth century and double as tackets for non-adhesive endsheet structures (p. 10). Szirmai includes several helpful diagrams of these systems and their attachment to the covers (p. 315).

I made a different types of fastener for each of my models to see how they worked. The simplest design is the bow (above), which is merely alum-tawed strips anchored to each cover and tied. While simple to construct, the main disadvantage of this system is that “many people do not have the patience to tie the bow” (Clarkson, 2005, p. 10).

A second fastener consists of a toggle attached to the lower board that hooks around the fore-edge and catches on loops anchored into the upper board. The forming of the toggle is depicted in Szirmai (2000). It is relatively simple to make – two holes are punched a few inches from the end of a strip of tawed skin and the end is rolled up to the first hole. The remainder is laced through the two holes and pulled until the roll is tight. Overall, I am not satisfied with this fastening system. I’ll admit that my execution could be better, but it seems that the toggles can be quite difficult to get in and out of the loops. I suppose they might function better when made out of different skin; it seems the tawed skin is too puffy and the toggle sticks inside the loop when attempting to pull it through. I will need to find and examine a set that function well to really know, however.

Finally, there is the ‘peg and frog’ fastening system. While this one is, by far, the easiest to loosen and secure, Clarkson (2005) reminds us that it originates in Asia and may clash with the cultural or historical consistency of the binding. Regardless, the peg is made of bone and the loop of the strap is secured with small laces of alum-tawed skin. I used a lok-eye needle to get the tiny laces into the strap (below). The image below also illustrates an alternative form of mechanical endsheet attachment – this time, outside of the turn-ins.

Limp parchment bindings with laced-on covers are incredibly durable, as witnessed by the number of bindings that survived the 1966 flood of the Florence Biblioteca Nazionale (Clarkson, 2005). This has led to the widespread adoption of this structure for conservation rebinding (Clarkson 1975; 1982). Further efforts to adapt the structure for conservation work have also been made, the most popular of which being the design by Robert Espinoza (1993). This binding operates very differently than the lapped component, in that it does not really lie flat . Both Clarkson (2005, p. 11) and Barrios (2006, p. 24) note that this structures functions best on thinner books, since the binding functions best in the hands rather than on a lectern.
While this structure is relatively easy to complete with paper, it is deceptively hard to do well in vellum. I am not yet able to really describe why it works out well sometimes and not others – but I think that variation can be attributed to the selection of the skin for the book. The thickness, flexibility, and overall combination of forces within the skin itself have to be examined and chosen specifically for the book block. If the selection is haphazard, the binding will feel off. Clarkson (2005) concludes that this a master binding, not a student one.
The 16th century form of this binding is considered by Clarkson to be the high point of the structure; as decoration takes precedence over structure in the following years, the quality of these bindings suffers a gradual decline. For some interesting historical examples of limp vellum bindings from the thirteenth century to the present, I would suggest browsing through the online exhibit created by Doug Rice.


Barrios, P. (2006). Notes on the limp vellum bindingThe Bonefolder 2(2), 24-27.

Clarkson, C. (2005). Limp Vellum Binding. Oxford: Christopher Clarkson.

Espinosa, R. (1993). The limp vellum binding: A modification. The New Bookbinder, 13, 27-38.

Greenfield, J. & Hille, J. (1990). Headbands: How to work them. New Castle, Delaware: Oak Knoll Books,

Lindsay, J. (1991). A limp vellum binding sewn on alumn-tawed thongs. The New Bookbinder, 11, 3-19.

Pickwoad, N. (1995). The interpretation of bookbinding structure: An examination of sixteenth-century bindings in the Ramney collection in the Pierpont Morgan Library. The Library, Sixth Series, 17, 209-249.

Rhein, A. (1942). Alte heftungen in pergamentumschlange. Zu einem katalog “Die Wiegendruckeinbande der Stadtbucherie Erfurt”. Archiv fur Buchbinderei, 42, 11-13.

Szirmai, J. A. (2000). The archaeology of medieval bookbinding. Aldershot : Ashgate.