Although the concept of persistence in chlamydial infections has been recognized for about 80 years, there is still very little known about the mechanism by which this occurs. In this review, we revisit an old paradigm, long known to chlamydiologists and veterinarians, that in virtually all hosts of chlamydiae, including mammals and birds, chlamydiae reside in the gastrointestinal tract for long periods of time in the absence of clinical disease. Thus, if gastrointestinal infection occurs in most hosts, then it is very likely that gastrointestinal infection occurs in humans as well. We demonstrate that gastrointestinal infection does indeed occur in humans and propose that this anatomical site is the source of persistent infection in humans. The data in ruminants and animal models demonstrate that the immune system is unable to clear chlamydiae from the gut, so they can remain indefinitely, with continual shedding in feces. Clearly, many women become reinfected from an untreated partner; however, we propose that women, cured of genital infection, remain at risk for autoinoculation from the lower gastrointestinal tract. Moreover, there are substantial data demonstrating treatment failure of chlamydial infections, particularly with azithromycin. New data in the mouse model have shown that azithromycin is far less effective against chlamydial gastrointestinal infection than against genital infections. Therefore, it is possible that women cured of genital infection by antibiotics remain infected in the gastrointestinal tract and can become reinfected by autoinoculation from that site.