These two volumes present descriptions, transcriptions, notes and images of sixty-two manuscripts in the papyrus collection of the Austrian National Library in Vienna. There is also an introduction explaining the concept behind the edition, and at the end are a list of Greek words read by these editors but not by earlier editors, a list of corrections to previous editions, a two-way concordance between the numeration in this edition and those earlier Vienna corpora, and one between the Gregory-Aland numbering of New Testament Greek manuscripts and this edition.

The sixty-two manuscripts are divided into four categories: those on papyrus, those on parchment, lectionaries, and a single fragment of an apocryphal Gospel. In writing material, categories 1 and 4 are papyrus, and 2 and 3 all parchment. In each section they are arranged by order of New Testament text, so that in the papyri P96 (Matthew) comes first and P116 (Hebrews) last. As the introduction points out, these are not all the Greek manuscripts of the New Testament in the National Library. According to the Kurzgefasste Liste der griechischen Handschriften des Neuen Testaments, there are a total of eighty-seven, but it should be noted that there is not a one-to-one correlation between the two sets of items. Manuscripts with different class marks in the collection are brought together in the Liste, but here treated as individual items. Thus P41 is found as eight separate entries (for the explanation see pp. 64–5), 070 as five, and P56, 059 (=0215), 0219 and L1575 as two each. The total of Gregory-Aland manuscripts included here is therefore fifty-three. All the manuscripts given a papyrus or a majuscule numbering are included.

One manuscript has here its editio princeps. It is an eighth-century parchment leaf containing verses from Matthew xxviii (Gregory-Aland 0148). The introduction notes that ‘there are a number of biblical textual novelties, some of them caused by phonetically based spellings’ (p. 95).

The methodology for each entry is to provide an introduction, a diplomatic transcription containing what is visible, notes on this, and a reading text with missing words supplied, nomina sacra and abbreviations such as superline for final nu spelled out in full and accentuation, some readings which the editors consider to be errors regularised (e.g. 0148 in its recto, column 2, line 18), and further notes. The diplomatic transcription presents page lay-out, including ekthesis, and in the case of several later manuscripts is at pains to present the punctuation, such as no. 40 (Gregory-Aland 0105). The editors state that they are more cautious than most editors in confirming readings, and that they use the underdot more extensively. Where the diplomatic transcription has an underdot without a letter to indicate the presence of ink which cannot be verified as a particular letter, the letter required by the context is supplied (with an underdot) in the reading text. There is detailed discussion of reconstructions which differ from previous editions.

In terms of convenience of use, it would be useful to have running titles which told one which manuscript was being treated on each page. As it is, one has to go to and fro from the introduction and concordances to find one's way around. There is also no cross-referencing between the text and the plates, which are given the edition number and the inventory number. This is especially confusing in the case of plates x–xvii, all of which refer to P41 and give us no guide to find the precise piece of text we are wanting to match up to the transcription. The volume is also unhelpful in never providing chapter and verse numbers against the transcription, so that locating the biblical text against it is troublesome.

One wants to check for accuracy above all. My own experience of making editions like this suggests where the weak points may be. The first hazard is linking up plates with images. Here we find that the very first image is wrongly labelled as K 7277. It should be K 7244. Looking at plate xiv, the photograph of the papyrological verso includes a piece of Coptic to the right, while the correct piece of text is given below adjacent to the verso, but here the main photograph is of a piece of Coptic text and should not be present (there is only Greek on one side). These matters are minor, but they are distractions from using the edition efficiently.

Another danger area is in providing two transcriptions of each manuscript. Depending on the way in which the diplomatic transcription is expanded into the reading text, there is plenty of opportunity for making accidental changes (use of electronic methods would be comparatively secure, but typing or writing them out afresh is hazardous). Here I did not find many mistakes in the eight or nine transcriptions I checked. Note that line 2 of the second column of the recto of no. 45 (0218) reads τιτo not πιτo (correct in the reading text, wrong in the diplomatic transcription), and see the notes on P45 below.

Next, one has to compare transcription with image to assess the results. Inevitably, there will be points of disagreement. Sometimes I think the editors have treated the darkening one often gets on the broken edge of a parchment as ink when it is not. For example, we saw no trace of any letter before the upsilon in line 12 of the first column of the recto of no. 45 (0218). Checking this transcription against that I completed with Bill Elliott, also by autopsy, I find that we were more confident about partial letters, working on the traditional principle that if part of a letter is clear enough to be the letter demanded by the context, it should be included with a dot, and if enough is present that it could not be any other letter, then even a dot may quite often not be necessary. Even allowing for extra permissible caution, I find the reconstruction a little conservative. For example, in line 9 of the first column of the recto, a microscope reveals enough of the nu for it to be allowed (with a dot). On the other hand, we were not able to confirm any of the χ in line 7 of column 2, which we find here (with a dot). Enough has been said to show the reader that these are not always easy matters to resolve. But users often want the benefit of the interpretation of an experienced papyrologist, whose recognition of a partial letter, derived from wide experience and careful study of this manuscript, is worth more than the judgement of a brief acquaintance.

I turn to the leaf of what is arguably the most important manuscript in the collection, no. 2 (P45). In one respect the edition is extremely misleading. Part of this leaf is in Dublin, and the reading text presents as reconstruction wording which is actually extant. This is particularly grievous when the reconstruction differs from what we know the manuscript to read, on line 6 of the verso (Μatthew xxvi.23) where the Dublin fragment has [χειρα μετ εμoυ εν τ], but the Porters' reconstruction follows the Nestle-Aland text (which reads μετ εμoυ την χειρα εν τ etc). The editors do refer to the existence of the Dublin fragment in their introduction, so why have they not used it?

Turning to the diplomatic transcription, on the codicological recto, I note the following problems:

• I see enough of the mu at the end of line 2 for it to be certain, but it is not recorded at all.

• In line 14 the editors read the nomen sacrum as ρνα, criticising Gerstinger's edition for reading τρνα. In fact the shape of the character and its position points to its being a staurogram, and thus Gerstinger is correct.

• In line 20 the letter partially missing at the end could only be upsilon.

• In line 24 the second part of the transcription reads oυc . In fact what is visible is π[αρ]εχετε followed by the very beginning of the crossbar of tau.

• In line 28, the first letter is definitely upsilon.

• In line 31, there is no trace in the transcription of the clear initial chi (it is present in the reading text).

• Rather more is visible at the bottom of the page than is recorded: the alpha after mu in line 34 is very clear, and some other traces of ink are not even noted as traces. The microscope could surely yield more here.

A little experience shows that the editors are not more in line with the general practice of reading letters than they claim, for example in the dotted eta of line 15.

The maximum permitted space for this review precludes further similar checks. I leave it to its readers to decide whether this is the level of accuracy which they desire in such an edition. But there are questions to be asked. Most notably, the regularising of errors out of the reading text is highly problematical.