Surveys were made in the 500 km river section between Hukou and Zhengjiang, which spent about 660 hours and covered a distance of about 4700 km in the past 3 years. The breathing intervals and behavior patterns of the baiji were observed. Ten baiji were counted in the surveys in 1989, 10 and 12 were counted in the two trips respectively in 1990. Because of unfavorable weather conditions, only 2 baiji were sighted in 1991. Six baiji were recognized by the photographs taken in 1989. Of which, the no. 1 and 3 were photographed again in 1990, one additional baiji was identified by 1990 photos. The movement of the animal no. 1, 3 and 7 was recorded by the photographs. The maximum resight distance recorded is up to 204 km. This indicates that the baiji migrated within a range of up to 200 km or more. Photographic identification technique was proved very important in the study of individual movement. It is also a useful tool in population estimate. Some baiji sighted on different dates and at different localities were proved by the photos to be of the same group. With these facts in mind, the previous published data on the number of groups and individuals should be reexamined very carefully. The results of our studies suggest that the population density is about 1 per 14 km in the lower Yangtze. This is about half of the estimate made by Zhou et al. by the end of 1980s. If this estimate is also appropriate for the baiji population in the middle reaches, at least we can assume that the dolphin's number has fallen to less than 200. The baiji is close to extinction. In response to the serious decline and threat to the baiji, we have made great efforts to help save the baiji, including initiate the semi-natural reserve project at Tongling and promote public awareness by prepare and distribute education materials.