This monograph, based on the author's PhD thesis, focuses on the animals depicted in mosaics from Roman Britain and is intended as a reference work for specialists. The volume is part of a larger project and will be integrated into a forthcoming work on deities, heroes, humans and personifications. More than 700 creatures are recorded from the 140 animal mosaics known to the author up to June 2015. As stated in the Preface (xix), Witts concentrates on questions of iconography (‘what’ and ‘why’) over context (‘when’, ‘how’ and ‘by whom’): she aims to investigate the meaning and significance of these images and to assess their relationship with mosaics from other provinces of the Roman Empire.

The text is effectively structured. Ch. 1 offers a synthetic overview of the subject, followed by an up-to-date literature review. This is generally very thorough, although no mention is made of Gullini's work on the animals in the Palestrina Mosaic (I Mosaici di Palestrina (1956), 20–32) and some major studies on how nature was regarded in the ancient world are similarly omitted (e.g. G. Siebert, Nature et paysage dans la pensée et l'environnement des civilisations antiques (1996)). In the second part of the chapter, W. explains the issues in identifying different species of animals, the terminology she has adopted and her methods of selection and classification. Chs 2, 4 and 6 are dedicated to animals associated with land (quadrupeds and snakes), sea (sea creatures) and sky (birds) respectively. Chs 3, 5 and 7 provide further details about each group, listing the percentage of different species, their gender, setting etc. Ch. 8 summarises the information collected in the previous sections and compares the characteristics of the three categories of creatures. Throughout these chapters, the author reviews vexed questions such as the inscriptions in the Rudston Venus mosaic (32) and the nature of the human figure with bird's head from Brading Villa (101), her judgements being always carefully considered.

Ch. 9 assesses the role and meaning of animals in mosaics and discusses the context of these images in relation to the rooms where they were found and their geographical distribution. Despite a certain degree of repetition, W. makes some intriguing points here. For instance, she proposes that the depiction of birds valued for the table in Orpheus mosaics from the Cotswolds might suggest bird-rearing activities in the area (142–3). She also convincingly argues that the room decorated by the Gloucester House mosaic at Cirencester was octagonal in shape and probably part of a set of baths (144).

The volume is supplemented by a catalogue of 140 animal mosaics, each briefly described and accompanied by comprehensive bibliographical references. In case of doubtful entries, W. gives sound justifications for inclusion. The catalogue and extensive bibliography are followed by five appendices: 1 and 2 offer alternative classifications of the mosaics under study; 3 groups together animal mosaics with wave-pattern borders; 4 lists the Orpheus mosaics from Britain; and 5 deals with the dating.

The book is enriched by a substantial number of illustrations. A colour image of almost every animal mentioned in the text is given (a total of 334 figures mostly of good quality). Where available, a small b/w picture of each mosaic accompanies the catalogue entries, with a further four b/w figures and six colour plates depicting particularly significant floors. These include items not previously published such as a beautiful watercolour by John Lickman of a mosaic discovered in the eighteenth century at Wellow, Somerset (210, pl. F). Mosaics from other provinces are not illustrated; while the abundance of British material justifies this decision, at least a selection of the most obscure comparanda would have simplified the life of the reader. The text is well-finished, with very few slips and typos (e.g. Thermae Romano-Britannicae, not Romanae-Britannicae (235)).

Overall, this is a thorough study and a very useful resource, strengthened by W.’s impressive knowledge of antiquarian material. Despite its primary function as a reference work, W.’s plain style makes it an enjoyable read.