The link between Pacific rainfall and weather patterns around the world is well known. It is therefore important that changes in Pacific rainfall patterns be identified. It has been suggested that global warming may play a role in altering the tropical Pacific hydrologic cycle. This study uses an unsupervised, competitive Neural Network to explore patterns and trends in Pacific rainfall from The Comprehensive Pacific Rainfall Data Base (PACRAIN) and Taylor's Atlas of Pacific Islands Rainfall. Both datasets were obtained from raingauge measurements on Pacific islands and atolls. The PACRAIN dataset comprises daily rainfall measurements from the period 1971-present from over 350 stations. Rainfall data prior to 1971 has been obtained from Taylor's Atlas which contains monthly rainfall for 118 stations. Rainfall data in Taylor's Atlas goes back in most instances prior to 1950, and prior to 1900 for 12 of the 118 stations. Results from the Neural Network analysis are discussed, and changes in Pacific rainfall patterns are identified. The results are compared to recent studies which have shown evidence for an enhanced hydrologic cycle in the tropical Pacific over the last two decades