DOIs explained: how to find the paper behind a citation

What a DOI is

A DOI, or Digital Object Identifier, is a persistent identifier for a piece of content, usually a research article, book chapter or dataset. It looks like this: 10.1530/rep.1.00001. The part before the slash (the prefix, which always starts with 10.) identifies the organisation that registered it; the part after the slash (the suffix) is chosen by that organisation.

The DOI system is governed by the DOI Foundation, a not-for-profit organisation, and its syntax is standardised as ISO 26324 (DOI Foundation, The Identifier). Registration agencies such as Crossref, which serves scholarly publishers, assign DOIs and store metadata about each article: title, authors, journal, volume, pages and publication date.

Why DOIs keep working

The trick is a level of indirection. A DOI does not contain a web address. Instead, the resolver at https://doi.org/ looks up the DOI and sends the reader to wherever the content currently lives. When a journal moves to a new publisher, the publisher updates the record, and every DOI link in the world follows automatically.

The journal Reproduction is a good example. It was published for many years by Bioscientifica for the Society for Reproduction and Fertility, and moved to Oxford University Press at the start of 2026. Its articles kept their DOIs, so a DOI link written in 2005 still works today.

Why old web addresses break

Web addresses are not persistent. When a journal leaves a website, links to that site stop working unless someone keeps them alive. For many years this domain served Reproduction and its predecessors, and thousands of papers, textbooks and web pages still link to addresses here, such as /content/152/6/R263.full or /cgi/doi/10.1530/....

How the reference archive works

This site is independent and is not the journal or its publisher. To keep those old citations useful, it maintains a reference archive: for each of more than 13,000 articles, a reference page at the address the journal used, with the citation details obtained from Crossref and a link to the article's DOI. The older address forms, such as abstract, full text and PDF variants, lead to the same record. The full text itself is available from the publisher through the DOI; this site does not host or reproduce articles.

For example, a much-cited 2016 review, "The effects of radiofrequency electromagnetic radiation on sperm function", has its reference page at /content/152/6/R263/, which links to the publisher's version through its DOI. We discuss that topic in are sperm counts really falling?

Finding a paper from a citation

1. If the citation includes a DOI, add https://doi.org/ in front of it and open the link.
2. If it gives volume, issue and page, a search on the journal's site or in a bibliographic database usually finds it. For the journals once hosted here, the archive by year lists every volume and issue.
3. If it points to an old address on this domain, follow it: the reference page shows the citation and the DOI.

Access to the full text

Some articles are open access; others require a subscription. Many universities and public libraries provide access. Authors are often allowed to share their accepted manuscripts in institutional repositories, and a search for the article's title can find such versions.

For how to judge what a paper actually shows once you have found it, see how to read a fertility study.

Other identifiers you will meet

Besides the DOI, two other identifiers are common in biomedical research. A PMID is the number given to an article indexed in PubMed, the database of biomedical literature maintained by the US National Library of Medicine; it is useful for finding abstracts and links to full text. An ORCID identifies a researcher rather than a paper, so that their work can be found even if they change name or institution.

Reading a citation

A typical citation lists authors, year, title, journal, volume, issue and pages, and ends with a DOI. For example, a citation to Reproduction volume 152, issue 6, starting at page R263, corresponds to the review on radiofrequency radiation and sperm mentioned above. The letter "R" in the page number marks review articles in that journal. Old citations to the Journal of Reproduction and Fertility often give only volume and page, which is enough to find the article in the archive by year.

Why keep old addresses alive?

Broken links are a real problem for science. Studies of scholarly references have found that a large share of web links cited in papers stop working over time, a phenomenon sometimes called link rot. Keeping old addresses resolving, and pointing them to the persistent DOI, is a small contribution to keeping the scientific record usable. The history of the research those addresses lead to is the subject of what animal research taught us about reproduction.

How the reference pages are built

The reference pages on this site are generated from bibliographic metadata registered with Crossref: title, authors, journal, volume, issue, pages, year and DOI. They do not include the text of the articles, which belongs to the publishers and authors. If a reference page contains an error, or an old address on this domain does not lead to the right record, readers can report it by email and it will be corrected.

Persistent identifiers are unglamorous infrastructure, but they are what allow readers today to check the claims made in a paper published decades ago.