Reuse as packaging is a filter, and it selects against precisely the books a library catalogue would have listed. Most of the seventeen are popular texts that survive in several copies across England and France, among them the Anglo-Norman *Lumere as lais* [6]. That is the expected result: the codex you cut up for a cover is the one nobody in the sixteenth century thought scarce. So the sample tells you what circulated widely, not what the island owned that was unusual.
The language split is the more durable finding. Six of the seventeen are in French, with many of the rest in Latin [6], which is about 35 percent of the collection in the vernacular [1]. On that basis the team reads late medieval Guernsey as a place whose readers, when they left Latin, read French and not English [7].
The outlier matters for anyone who edits texts. *L'Estoire de Griseldis*, a secular play attributed to Philippe de Mezieres around 1395 [8], survives in this version in one other manuscript and two printed editions from the sixteenth century [10]. Counting Guernsey, the manuscript witnesses go from one to two [2]. One witness cannot be collated against anything; two can, and the places they differ are evidence about transmission rather than guesswork. Rundle's group also argues the fragment is the first sign of the play being read outside the French royal court circles where it probably started [11]. De Mezieres was working roughly forty years after Boccaccio put the Griselda story in the *Decameron* in 1353 [9][3].
How it was found is the part other archives should read closely. Callum Tostevin-Hall of the Island Archives went looking for a sixteenth-century copy of Edward III's 1331 Extente of Guernsey and noticed that the shelfmark itself mentioned a vellum cover with Latin text, taken off the notebook during an earlier round of conservation; the cover proved to be a reused bifolium from a medieval manuscript [13]. No crate had to be opened. The lead sat in a descriptive field about the object's wrapping, written by someone who had no reason to care what the wrapping said.
Rundle expects more of these to surface, on Guernsey and elsewhere [12]. The load-bearing condition is that somebody, at some point, wrote down that a volume had skin on it. And several of the seventeen were already known to Guernsey archivists; what was absent was anyone able to say what the texts were [3]. That is the real bottleneck in this method. A detached cover is not evidence until a reader can name it, which means the supply of finds is limited less by archives than by the number of people who can identify a half-page of fourteenth-century French on sight.
