Determining the place of a single witness within the textual tradition of the Old French prose Tristan is notoriously tricky, in part because of the competing classificatory systems used by modern scholars. This article begins with a reflection on the various «versions» of the Tristan that have hitherto been identified (with an appendix reclassifying extant manuscripts according to Emmanuèle Baumgartner’s schema), before attempting to pinpoint the place of a long-neglected ​​«complete» witness, ms. 164 of the Fondation Martin Bodmer, within the Tristan tradition. This volume’s three constituent parts, bound together in the fifteenth century, are each studied from a codicological and a textual perspective. The lost head of this «acephalous» manuscript will be found, the inclusion of Tristan’s fight with a dragon will shed light on the textual affiliations of one of its parts, and a case will be presented for Burgundian production and assembly of a volume once deemed Italian.