An experimental rabbit model was developed to study the microbial colonization of intrauterine contraceptive devices. Tailed and tailless devices were surgically inserted into into the uterus by two different routes: surgically, directly into the uterine horn, thus avoiding contact with the vaginal and cervical microfloras, or via the vagina and cervix. After 1 to 8 weeks the devices were recovered and prepared for scanning electron microscopy. The surfaces of surgically inserted devices remained uncolonized all through the experiment whereas in those inserted via the cervix microorganisms colonized the core surface as early as 2 weeks after insertion. Our data suggest that in our experimental conditions the mode of insertion appears to be the major factor influencing the microbial colonization of intrauterine contraceptive devices and that the presence of a tail does not seem to play a significant role.