Defining the Jurassic-Cretaceous boundary is a controversy in stratigraphic study of the world. It has been widely accepted that this boundary can be defined at the bottom of Berriasian in Tethys, with the appearance of the ammonite Berriasella jacobi dating to ca. 145 Ma. However, it is difficult for the widespread terrestrial deposits in China to correlate with the international standard of marine facies. The Somanakamura Group in Japan is represented by a succession of marinecontinental transitional strata. It provides a bridge of marine and nonmarine stratigraphic correlation. The ammonite and radiolarian fossils preserved in this group suggest an age from Bajocian to early Valanginian. The J-K boundary was defined in or atop the Tomizawa Formation of the group according to the ammonite data. The present authors study the fossil spores and pollen newly found from the Tomizawa and Koyamada formations. Three assemblages have been recognized. They are Assemblage 1(Cyathidites-Classopollis) from the upper part of the Tomizawa Formation, Assemblage 2(Cyathidites-Jiaohepollis) from the lower part of the Koyamada Formation, and Assemblage 3(Cyathidites-Spheripollenites-Ephedripites) from the middle to upper part of the Koyamada Formation. With the reference of ammonite evidence, the J-K boundary can be defined between Assemblage 1 and Assemblage 2. This palynological J-K boundary can be correlated with that of terrestrial sequence in China. However, local biostratigraphy imply that the continental J-K boundary in China is of 135 or 137 Ma age. It has a considerable discrepancy from the marine standard. Biogeographically, the distribution pattern of spores and pollen in southern China is in accordance with that in the Somanakamura Group, which parallels the Tuchengzi Formation in northeastern China. By the palynological correlation between the Somanakamura Group and the strata in southern China, and then with the sequence in northeastern China, it is suggested that the continental J-K boundary is located in the Tuchengzi Formation. Defining the Jurassic-Cretaceous boundary is a controversy in stratigraphic study of the world. It has been widely accepted that this boundary can be defined at the bottom of Berriasian in Tethys, with the appearance of the ammonite Berriasella jacobi dating to ca. 145 Ma. However, it is difficult for the widespread terrestrial deposits in China to correlate with the international standard of marine facies. The Somanakamura Group in Japan is represented by a succession of marinecontinental transitional strata. It provides a bridge of marine and nonmarine stratigraphic correlation. The ammonite and radiolarian fossils preserved in this group suggest an age from Bajocian to early Valanginian. The JK boundary was defined in or at the the Tomizawa Formation of the group according to the ammonite data. The present authors study the fossil spores and pollen newly found from the Tomizawa They are Assemblage 1 (Cyathidites-Classopo llis) from the upper part of the Tomizawa Formation, Assemblage 2 (Cyathidites-Jiaohepollis) from the lower part of the Koyamada Formation, and Assemblage 3 (Cyathidites-Spheripollenites-Ephedripites) from the middle to upper part of the Koyamada Formation. With the reference of ammonite evidence, the JK boundary can be defined between Assemblage 1 and Assemblage 2. This palynological JK boundary can be correlated with that of terrestrial sequence in China. However, local biostratigraphy imply that the continental JK boundary in China is of 135 or 137 Mage. It has a decent discrepancy from the marine standard. Biogeographically, the distribution pattern of spores and pollen in southern China is in accordance with that in the Somanakamura Group, which parallels the Tuchengzi Formation in northeastern China. By the palynological correlation between the Somanakamura Group and the strata in southern China, and then with the sequence in northeastern China, it is suggested that thecontinental J-K boundary is located in the Tuchengzi Formation.