Citing sources that do not support the assertion being made can misinform readers, perpetuate mistakes and deny credit to the researchers who should have been acknowledged. To quantify citation fidelity in marine biology, we retrieved 198 papers from 2 recent issues of 33 marine biology journals. From each paper we randomly selected 1 citation, recovered the source material, and evaluated its appropriateness. We discovered that the assertion was ‘clearly supported’ by the citation in only 75.8% of cases, the support was ‘ambiguous’ in 10.6% of cases and the citation offered ‘no support’ to the original statement in 6.0% of cases. The remaining 7.6% of cases were classified as ‘empty’ (citations to secondary sources). We found no relationship between citation appropriateness and the position of the assertion in the paper, number of authors, number of references, article length and Journal Impact Factor. That 1 in 4 citations in marine biology should be viewed with scepticism is alarming and has important ramifications for both scholarship and bibliometrics.