Vaginal administration of contraceptive steroids has several advantages over oral administration including avoidance of the digestive tract and the 1st passage effect continuous release of the active principle allowing lower doses and fewer side effects independence of the method from sexual relations and possibility of application directly by the user. An early vaginal ring containing medroxyprogesterone acetate in silastic inhibited ovulation but released the progesterone in large initial bursts followed by rapid decline offering no advantage over parenteral administration. The problem was resolved through development of a vaginal ring with an inert core covered with a fine layer of a mixture of silastic and progesterone in turn covered by a fine layer of silastic. The speed of release of the new ring was proportional to t he surface of the ring and inversely proportional to the thickness of the outer layer. This method was used to release variable quantities of levonorgestrel over 3-6 cycles which inhibited ovulation in 85% of users but provided only mediocre cycle control. The World Health Organization at about the same time was sponsoring research focussed on development of a vaginal ring releasing small doses of progestins with a minimal blocking of ovulation but a major contraceptive effect at the level of the cervical mucus and endometrium. Several progestins were evaluated and dose-response tests were conducted on a number of them. A ring releasing 20 mcg/day of levonorgestrel was found to cause fewer cycle problems and to have a moderate affect on ovarian function. The Population Council in New York developed a method adding estrogen to the progestin to resolve the problem of irregular bleeding and obtain better cycle control. Development was halted despite promising results because of difficulties of mass production and the results of metabolic studies that indicated the method had undesirable effects on lipoprotein levels. But large scale multicountry trails with the levonorgestrel-releasing ring continued to give positive results. Bleeding problems when they arose were less serious than those with low-dose oral progestins. The efficacy was at least equal. A detailed study of social acceptability found the level of acceptability of the vaginal ring to be high with women appreciating their ability to remove the ring themselves. Despite the favorable results marketing of the levonorgestrel vaginal ring was delayed by problems in mass production and in procuring adequate supplies of silastic. Recent developments have included new lightweight plastics and a model that allows the ring to be divided into sections. A multicenter study is underway of a ring releasing 120 mcg of desogestrel and 30 mcg of ethinyl estradiol. Results have been very favorable. A vaginal ring is likely to be commercially available by 1995.