Local immunological defense mechanisms in the cervicovaginal mucosa currently remain incompletely defined, especially from a quantitative point of view. Addition of an inert substance, lithium chloride (LiCl), into the washing buffer used to carry out the vaginal washing for collecting cervicovaginal secretions and measurement of its concentration with a flame absorption spectrophotometer, before and after the specimen is sampled, permits the quantification of the volume of cervicovaginal secretions collected and the approximation of the dilution factor of a soluble component introduced by the washing. Lithium, at a concentration of 10 mM, gives the best precision of measurement and has no effect on the results of the immunoassays. In a population of 27 nonpregnant women (age range, 18 to 45 years), the volume of cervicovaginal secretions collected by vaginal washing with 3 ml of LiCl-phosphate-buffered saline was 12% +/- 3.2% (mean +/- standard deviation) of the total volume and showed large interindividual variations (range, 5.6 to 18.8%); the mean dilution factor of a soluble component from the vaginal secretions was 9.9% +/- 2.8% (range, 6.3 to 18.8%). According to the date of the menstrual cycle, the mean volume of collected cervicovaginal secretions was significantly increased in the luteal phase in comparison with the follicular phase; conversely, the mean dilution factor of a soluble component was more important in the follicular than in the luteal phase. These features strengthen the need to quantify accurately the dilution factor introduced by vaginal washing when studying cervicovaginal immunity.