Defining the interior structure and chronology of a legionary fortress is a daunting task and gets more complicated the denser the successor settlements on top of the Roman layers. It has taken in the case of Caerleon, Inchtuthil and Nijmegen nearly 80 years to get as far as we have, but the large fortresses underlying the modern towns of Mainz, Strasbourg, Vienna and Brigetio have seen for the most part only small-scale interventions, disclosing disjointed walls and floor-levels that are hard to interpret and difficult to use as a basis of a larger chronological system.

Within this context the 1,544-page publication on the excavations of the Judenplatz in Vienna represents a milestone. Located within the fortress close to the via principalis sinistra, it provides insights into three barrack blocks of a normal cohort, as well as a tiny section of the praetorium. For an inner city excavation the area is substantial and, possibly due to its location under a square, remarkably well preserved.

The excavation is meticulously presented, with every feature documented as far as possible in plan and section and often colour photograph, while the finds reports are accompanied by lengthy discussions of all aspects of the material and with full catalogues and numerous illustrations. Special features, such as hearths and ovens, are separately listed, as are the construction details of all walls. The context descriptions are extremely detailed. In sum, every care has been taken not just to present as much information as possible, but also to allow future excavators the chance of cross-referencing any possible detail or paralleling their material with the relevant period here, or even in future years of revising the results in light of further research. In a time of dwindling publication budgets and an increased tendency to relegate more and more information onto disks, supporting websites or even just the site archive, this approach is to be warmly welcomed, and despite its hefty weight in excess of 5 kg and the intimidating price tag of €134, it will be a very welcome addition for anybody interested in legionary fortresses and Late Antiquity, particularly on the middle Danube.

The results of all this labour of love are considerable: the Judenplatz excavations offer a continuous sequence of features from the first timber buildings to the early medieval/sub-Roman period of Vienna, associated with numerous finds in closed contexts providing a good chronological framework. The area is large enough to prove that the patterns encountered within one building are repeated in others, thus allowing for the first time the creation of a reliable overall periodisation of the fortress from the late first century to the fifth/sixth century.

Of particular interest is the successful identification of one, probably two timber periods in Vindobona fortress during the second century. The construction of the barracks in stone or at least on stone dwarf-walls at least 0.7 m high appears to start at the very end of the second century and dates probably to Septimius Severus, with a further stone period datable to c. a.d. 280/300–350/360, during which time considerable wealth was spent on channel hypocausts and wall-painting.

The most impressive evidence comes, however, from the last stone phase (Period 5: a.d. 360/375–390/410). The structures, which still follow the outlines of the earlier barracks are no longer used as formal military barracks but appear to have been converted into a series of workshops among which a series of glassworking ovens with all stages of the production process stands out. Mosser and his colleagues link this development with the contemporary abandonment of the canabae and the gradual conversion of the legionary fortress into a fortified garrisoned town from the reign of Valentinian onwards.

The end of this period, at the beginning of the fifth century, is associated with large amounts of rubble and abandonment layers over much of the site and in Period 6 only a few rooms remain in use, with internal walls being removed and a new floor being inserted on two separate occasions. In the third phase of this period the area is levelled with clay and broken tiles and there is a horizon of post-holes and pits which cut the levelling. This latest phase of occupation, which dates probably to the mid-fifth century, is eventually sealed by black earth deposits and the absence of any later pottery finds from the site.

This new sequence of dates, together with the recent publication from the civilian settlement and the cemeteries of Vienna, is likely to define our view of the fortress for the future and M.'s gazetteer and cautious interpretation of the other smaller excavations within the fortress and their possible relationship to the Judenplatz already point the direction for promising future research.