A Greek and Arabic lexicon: materials for a dictionary of the mediaeval translations from Greek into Arabic

impression of unfinished business persists. A proper description of the Middle English manuscripts and a much more detailed exposition of their relationship remain essential desiderata. One can accept the decision to print a single witness, but preference for the Wellcome MS over all the others would have been more comprehensible if a fuller picture of their character had been allowed to emerge. The description of the MS itself unfortunately displays the same kind of inaccuracy as bedevils the edited text. The question of common sources, a vital problem in any consideration of the vernacular English writer's procedure, is obscured by the fact that the editor divulges nothing concerning Gilbertus's sources, yet his incorporation of significant portions of Salernitan texts such as Roger Frugard's Chirurgia or Johannes Platearius's Practica brevis is a salient feature of the Compendium. Although the editor has compared the 1510 print with London, B.L. MS Sloane 272, the textual tradition is not discussed and my own impression is that the text of the print is susceptible of much improvement. Another omission which calls for future remedial work is the complete lack of any reference to Anglo-Norman (the single reference to "Norman French" on p. xvii is antiquated and inaccurate)-no mention of Rothwell's extensive work on the linguistic concurrence of Anglo-Norman and English, no awareness that major medical treatises in Anglo-Norman existed almost two centuries before the work studied here, no acknowledgement that over a third of the lexical items asterisked in the glossary as being first usages not in the MED, or unique usages not in MED or OED, already appeared in Anglo-Norman in the thirteenth century. We must be grateful for Faye Getz's labours, whilst regretting that in the ten-year interval following its acceptance as a PhD thesis her study has not been more carefully revised.