An observational study of 144 (mean age 50 [range 45-57] years) perimenopausal women randomly selected from the National Register was conducted in order, to study long term effects of vaginal deliveries on anorectal function. The mean number of vaginal deliveries was two (range 0-6). Measurements were perineal position at rest and during straining, anal mucosa electrosensitivity, maximum resting pressure, maximum squeeze pressure of the anal sphincters, and pudendal nerve terminal motor latency. All tests were performed by a single investigator (AMR) who had no knowledge of the subject's parity. Increasing parity correlated with a lowered perineal position at rest (correlation coefficient (r) = 0.26, p = 0.003) and during straining (r = 0.24, p = 0.006), an increased threshold of anal sensibility (r = 0.22, p = 0.008), and an increased pudendal nerve terminal motor latency (r = 0.27, p = 0.002). No effect of parity on the maximum resting pressure (r = 0.06, p = 0.70) and maximum squeeze pressure (r = 0.06, p = 0.36) was found. The number of vaginal deliveries only accounted for a minor fraction of the total variability (between 3.6-5.7%). It is concluded that repeated vaginal deliveries have a long term adverse effect on anorectal function in a population of randomly selected healthy perimenopausal women.