It is interesting that the principal histological features of acute H. pylori gastritis, surface epithelial degeneration and neutrophil polymorph infiltration, remain as the most sensitive indicators of the 'activity' of infection in the chronic phase. It is not surprising therefore that these are the first features to resolve after successful H. pylori eradication therapy. In one of the earliest studies of histological response to eradication, McNulty et al (1986) endoscoped patients immediately after a three-week treatment regime and found a highly significant decline in polymorph scores. The response was even more striking four weeks after the end of treatment, as at that time biopsies from responders were virtually devoid of polymorphs (Valle et al, 1991). Indeed the disappearance of polymorphs from a post-treatment biopsy is a useful indicator of successful eradication. Less attention has been paid to the recovery of the surface epithelium yet this is an impressive feature when comparing pre- and post-treatment biopsies. Using subjective grading of surface epithelial lesions, Solcia et al (1994) found a dramatic and highly significant improvement in mean grade immediately after anti-H. pylori treatment. Recently a morphometric approach was used to demonstrate a significant increase in surface epithelial cell height corresponding to the recovery that accompanies successful H. pylori eradication (Hassan et al, 1993). Chronic inflammatory cell infiltrate resolution is much slower. There is only a gradual reduction in cell density so that even 6 months after eradication treatment the mean score had only fallen by 50% of pre-treatment values (Solcia et al, 1994). In the author's experience, a minor increase in such cells persists for many months and may never completely resolve, in that more lymphocytes and plasma cells are seen than in a truly normal (pre-infection) stomach. Valle et al (1991) found resolution of chronic inflammation in only 15% of subjects at 6 months and in 51% at one year after eradication. The long delay in disappearance of lymphocytes and plasma cells poses a question over continuing antigenic stimulation in the absence of infection. Possible answers could involve persistence of antigenic moieties in dendritic cells of the lamina propria, or sensitization to host antigens brought about by infection so that an element of autoimmunity persists after eradication. Unravelling these mechanisms will add important new elements to our understanding of the long-term consequences of this fascinating infection.