Abstract: The Japanese Working Group of IGCP Project 61 (Holocene sea-level change) has published“Atlas of Holocene Sea-level Records in Japan”in 1981, which is chiefly based on already published data in the last twenty years up to March, 1981. It is written in English in order to introduce Japanese works internationally. The atlas consists of the following three parts: Part I (p. 1-100); All sea-level curves or relative sea-level curves known to the compilers are shown together with important data, diagrams and tables for constructing the curves. Part II (p. 101-165); Some papers, which seem to be significant to understand the geologic or geomorphologic evolution of the coastal areas related to Holocene sea-level changes, are introduced. Part III (p. 166-186); List of 14C results which are regarded to have close relation to Holocene sea-level records. In addition, bibliography of Holocene sea-level study is listed (p. 187-195).Based on the collected data in the atlas, we compilers tried to summarize the results of recent researches and some problems on the Holocene sea-level change in Japan. The number of papers has increased as a result of the increase in 14C data. Especially many papers were published at the time of publication of the three special issues on this subject (Fig. 1). The studies are chiefly concentrated on the areas of large coastal plains such as Kanto, Nobi and Osaka, where sufficient bore-hole data for constructing the sea-level curves were available. Uplifting areas fringed by the Holocene marine terraces, such as South Kanto, the Ryukyu Islands and others where sea-level records were found on the land, have been also intensively studied (Fig. 2).Holocene sea-level records are summarized in Table 1. Fig. 3 shows examples of relative sea-level curves from selected areas. Most Japanese sea-level curves are characterized by higher sea-level of ca. 6, 000y.B.P. (Jomon transgression), which followed rapid sea-level rise accompanying the deglaciation. Since then, the sea-level has lowered to the present level with minor fluctuations. Sea-level lowering is found three times during the last ca. 10, 000 years (1, 2 and 3 in Table 1). Number and time of sea-level lowering are, however, sligthly different place to place. The magnitude of fluctuations in sea-level curves is also different locally (Table 1 and Fig. 5). For instance, the maximum stage of the Jomon transgression occurred at ca. 6, 500y.B.P. in the areas which are characterized by high rate of uplift. In contrast, it is ca. 5, 000y.B.P. or younger in rather stable or subsiding areas (Fig. 4).Examples of heights and ages of the Holocene marine terraces are listed in Table 2. They are usually subdivided into several steps in the areas of high rate of uplift. Subdivision of the terraces was probably caused by co-seismic uplift in association with large earthquakes, judging from the differences in the number and age of each step, although there are no enough 14C age data on the lower steps of the terraces. However, some features of small steps, such as width of step or thickness of deposits overlying steps, seem to be attributed to minor sea-level fluctuation.