There is increasing interest in the nature and biological significance of romantic love but few quantitative data are available for testing specific hypotheses. This paper describes the use of a survey instrument to assess incidence and duration of romantic attractions over a 2-year period amongst students (121 male; 162 female) progressing from school to university education. The results for males and females were similar and schooling – single-sex or co-educational – had little effect. Students averaged 1·45 romantic episodes per year and 93% of students reported at least one episode over the survey period. Duration of attraction was around 9 weeks if never reciprocated and around 12 weeks if reciprocated. There was seasonal variation of onset of episodes with peak incidence over the summer or early autumn seasons. Collectively the results accord with the view that frequent, short-duration romantic episodes could have a role in selection of appropriate long-term reproductive partnerships.