Up to 1976 an integrated malaria control programme has been carried out in Mayotte island. By 1980, there was a drastic reduction of the number of the malaria cases and of the plasmodic index (below 1%). But in 1984, an uprise of cases and a consistent increase of the plasmodic index was observed throughout the island (up to 10%). It is supposed to be due to the laxity in malaria control measures. Immediately a reinforcement of all aspects of malaria control measures was undertaken. Every house is sprayed four times a year with fenitrothion (2 g/m2). Chemical and biological ("guppy" fishes) larviciding measures were extensively applied throughout the island together with simple environmental measures like filling peridomestic breeding sites. Mass chemoprophylaxis was gradually decreased and is now only applied for pregnant women according to the W.H.O. recommendations. Presumptive treatment of febrile cases is generalized before laboratory confirmation. The surveillance of malaria prevalence is a continuous process both active and passive but, every year, a randomized parasitical and serological survey is carried out in a sample of representative villages. In 1986 there was only 3 indigenous cases observed despite the introduction of numerous carriers from neighbouring infected countries. It appears that pertinent control can reduced or may be eliminate malaria transmission in a short period of time. But the maintenance of the malaria clearance in the local, ecological and geographical context need a continuation of the control measures at least for the near future, as the population has lost most of its immunity and has become highly vulnerable to malaria.