STUDY QUESTION Do seminal plasma transforming growth factor-β (TGFB) cytokines vary within individuals over time, and does this relate to sperm parameters, age or prior abstinence? SUMMARY ANSWER Activin A and follistatin, and to a lesser extent TGFB1, TGFB2 and TGFB3, vary within individuals over time, in association with duration of abstinence. WHAT IS ALREADY KNOWN Seminal plasma TGFB cytokines can influence sperm function and reproductive success through interactions with the female reproductive tract after coitus. Over time, individual sperm parameters fluctuate considerably. Whether seminal fluid TGFB cytokines vary similarly, and the determinants of any variance, is unknown. STUDY DESIGN, SIZE, DURATION Between two and seven semen samples were collected from each of 14 fertile donors at 6-10 week intervals over the course of 12 months, then seminal plasma cytokines and sperm parameters were measured. PARTICIPANTS/MATERIALS, SETTING AND METHOD The concentrations and total amounts per ejaculate of TGFB1, TGFB2, TGFB3, activin A and follistatin were determined using commercial assays. Sperm parameters were assessed according to WHO IV standards. Mixed model analysis was utilised to determine the relative contribution of between- and within-individual factors to the variance. Relationships between cytokines and sperm parameters, as well as effect of age and duration of abstinence, were investigated by correlation analysis. MAIN RESULTS AND THE ROLE OF CHANCE Within-individual variability contributed to the total variance for all cytokines and sperm parameters, and was a stronger determinant than between-individual variability for activin A and follistatin as well as for total sperm concentration and sperm motility. Positive correlations between each of the three TGFB isoforms, and activin and follistatin, suggest co-regulation of synthesis. Duration of abstinence influenced total content of TGFB1, TGFB2, activin A and follistatin. TGFB1 correlated inversely with age. LIMITATIONS, REASONS FOR CAUTION A limited number of donors from a single clinic were investigated. Clinical information on BMI, nutrition, smoking and other lifestyle factors was unavailable. Further studies are required to determine whether the findings can be generalised to larger populations and different ethnicities. WIDER IMPLICATIONS OF THE FINDINGS These data reveal substantial variation over time in seminal fluid cytokines and indicate that repeated analyses are required to gain precise representative data on an individual's status. Within-individual variation in seminal fluid components should be taken into account when investigating seminal fluid cytokines. STUDY FUNDING/COMPETING INTERESTS This study was supported by grants from the National Health and Medical Research Council of Australia, ID453556 and APP1041332. The authors have no competing interests to disclose.