notes for referees

The task of a referee involves three main components: speed, confidentiality and thoroughness. These are, to some extent, conflicting, and the following notes may be of assistance.

If you feel there is any conflict of interest, either real or perceived, in carrying out your review, please refer to the Ethical considerations below.

1. Speed

Unless the article is of exceptional length or complexity, your report should be despatched within one week of receipt of the article. If you are unable to act, please let the Co-editor know immediately. You are welcome to suggest an alternative referee.

2. Confidentiality

Articles should be treated as confidential. In exceptional cases you may find it desirable to discuss the article with one of your associates; if you wish to do so, please let the journal know first. It is your responsibility to ensure that the article is only seen by that individual and is not read by others. The names of any individuals who have helped with the review should be declared so that they are associated with the article in our records. Under no circumstances should the article or any copy of it be retained.

The Editors do not reveal the identity of referees to authors. If, nevertheless, you wish to correspond directly with the author, the first contact should be made through the Editor or Co-editor.

3. Thoroughness

It will be helpful if you keep the following points in mind both in reading the article and in writing your report:

• (a) Is the work original and new?
• (b) Is the work sound and technically up-to-date? Are the conclusions firmly based on the observations or reasoning given? The journal accepts reports of difficult or challenging structures not meeting all validation criteria, provided the presented structures are correct and the difficulties and strategies used to treat them are scientifically discussed and properly documented.
• (c) Are there any aspects of the article for which you would like to suggest another referee?
• (d) Is the Abstract an adequate summary of the article? Abstracts should, as far as practicable, be 'informative' rather than 'indicative'; that is, they should contain the most important numerical and other results, and not merely state that such results are given in the article. The Notes for Authors specify some of the items that should be mentioned in the Abstract, including chemical formulae and special features like crystallographically imposed symmetry, Z' > 1, disorder, twinning.
• (e) The Results and discussion section should contain significant discussion beyond just a mere statement of the geometric parameters already visible in the CIF. Any non-crystallographic results reported should be discussed, not just listed, and significantly enhance the understanding of the system being studied. Added material that says little is not warranted.
• (f) Has all numerical data in the text sections been reproduced correctly? Geometric parameters should be exactly as in the CIF lists, not be rounded if the s.u. is less than 19, and given with s.u.values where such are derivable. Weak interactions should not be overanalysed; the distances and/or angles involved should be within reasonable ranges. Symmetry codes must be given in the text and figure captions when referring to atoms not in the 'home' asymmetric unit. Symmetry codes should be checked if possible. Any additional non-standard tables should be constructed appropriately, the parameters therein reported correctly and s.u. values included when available.
• (g) The experimental refinement does not need to state standard procedures, but should clearly and adequately document all non-standard procedures employed, including the treatment of disorder or twinning. Any constraints and restraints should be described with target and s.u. values, where appropriate. Details of how the H-atom positions were located or calculated should be included. The X–H distances and U(iso) definitions used in calculated values should be stated. Authors frequently use standard text here without adapting it to the case in hand, so this needs to be checked. The prose should avoid jargon, like program instruction names.
• (h) Is the literature adequately and correctly referenced? The consistency of the references and their citations in the text can be checked in publCIF.
• (i) Is the journal for which the article has been submitted the most appropriate? Consider particularly if the Results and discussion section contains sufficient discussion of the structures, as mentioned in (e).
• (j) If the article has been offered to the appropriate journal, and the results are worthy of publication, is the presentation the most efficient possible? The 'information density' should be high and uniform throughout. Duplication of tabulated parameters should be kept to a minimum in the Results and discussion section.

The following additional points should also be considered with care:

• (k) Does the article contain any tabular, illustrative or other lengthy matter which is, while worth preserving, likely to be of interest only to a very few readers? The IUCr has a Supplementary Publication Scheme which provides for the preservation and future availability of such material. The geometry tables to be printed should only contain parameters that are discussed.
• (l) Is the language usage satisfactory? Suggestions for minor corrections may be made when returning your review report. Obscure passages, if any, should be identified for the author to improve.

Finally, if you have suggestions that would make the article you are reviewing more accessible to a wide readership, please include these in your report.

4. Ethical considerations

IUCr publications are members of the Committee on Publication Ethics (COPE) and follow COPE guidelines on publication ethics. Referees are asked to follow the COPE Ethical Guidelines for Peer Reviewers when handling articles for IUCr publications.