This volume is the result of the collective work of the members of the ‘Groupe de Recherches sur l'Afrique Antique' (GRAA) based at the University Paul-Valéry (Montpellier iii, France). The GRAA is collectively responsible for the edition, translation and annotation of the De miraculis Sancti Stephani libri duo. A number of individual contributions precede, written by members of the GRAA together with a few other scholars. The collection of the miracles Stephen performed in the north African city of Uzalis was compiled on Augustine's recommendation to his friend Evodius, bishop of the city, around 424–5, once the miracles of Stephen began to be well known in the area (pp. 24–5; see also Y. Duval, ‘Sur la Genèse des libelli miraculorum’, Revue des études augustiniennes xxv/1 [2006], 97–112). The GRAA offers here the first critical edition of the text, available otherwise in the Patrologia Latina as an appendix to the works of Augustine (PL xli. 833–54). It is based on the collation of twenty-one manuscripts (presented by J. Meyers, pp. 207–51). The Latin text is faced by a French translation and followed by ninety annotations which are mainly elucidations of difficult points in the text. These notes are crucial to an understanding of a text that is otherwise not always straightforward. One could regret that the content of the contributions preceding the edition and translation of the text was not incorporated into a continuous commentary. As the annotations contain only general references to these contributions (under the form ‘Chap. III, 1’ to be understood as pt iii, ch. i), it is sometimes difficult to find the relevant information quickly. The contributions are distributed in three sections: Stephen's relics and their cult, Uzalis, and the De miraculis. Particularly worth mentioning among these are the two detailed, though slightly repetitive, articles by the late Y. Duval that provide an analysis of all the information about the cult and the monuments revealed in the text, and the study by M. Griffe of the language of the De miraculis which establishes that the author, though interested in quoting verbatim the sermo quotidianus of the people (examined by M. Griffe and J. Meyers in their contributions to the colloquium Les Structures de l'oralité en latin, Paris 1996), was well educated both in classical and Christian authors and well trained in rhetoric.