Classified among systems using low-pressure membranes, microfiltration and ultrafiltration are characterized as new water purification systems. They have served as a barrier against mainly Cryptosporidium for the past decade. At present, polymeric hollow fiber membranes are the main type used worldwide. However, in Japan, a water purification system using a monolith-type ceramic membrane with unique material and performance characteristics was developed in the early 1990s. The numbers of such systems in use and the total capacity of ceramic membrane systems are steadily increasing. They now account for more than 10% of the market share in Japan. This system was developed as the next-generation water purification system to take the place of the conventional system that had been applied to river surface water treatment in Japan with coagulation, sedimentation and rapid sand filtration. With this development, the authors selected and introduced their combination with coagulation as pretreatment and dead-end microfiltration. The ceramic membrane system affords excellent energy saving, reduction of long-term maintenance work, and stable operation without degradation of removal efficiency within water quality parameters such as total organic carbon (TOC). Operating experience at commercial plants and many pilot tests have confirmed that the ceramic membrane system can be operated with no breakage and with a CIP frequency of once every 1-3 years under the condition of 1-2.5 m3/m2/day (25-62.5 gfd) filtration flux. Moreover, more than 95% of the initial specific flux can be restored by simple CIP methods, and a water recovery ratio greater than 98% has been continuously maintained. This paper discusses the performance of the ceramic membrane system and presents operating data, including data from actual operations at municipal water treatment plants. The authors are continuing their efforts to improve the membrane and system, and to verify their performance at the pilot plant. Verification tests led to the expectation that a higher flux operation at 4-10 m3/m2/day (100-250 gfd) would be achieved in 2005. The higher flux operation and mass production of membranes should realize further cost reductions of the system. There is also the possibility that even the per water production volume cost of installing ceramic membrane systems can be less than that of existing polymeric membrane systems. Includes 4 references, tables, figures.