Cryptopecten is an extensively distributed genus, but the occurrences of fossils so far recorded seem to be restricted almost entirely to the Neogene and Quaternary of Japan except for some fossils of C. nux in a few regions of Southeast Asia and east Africa. Reflecting the influence of the warm paleocurrents, the occurrences of fossil Cryptopecten are concentrated in the Pacific coastal region of southwest Japan and the Ryukyu Islands except for a few cases on the Japan Sea coast (Fig. 4).

Fossil Cryptopecten occur dominantly in coarse-grained, sometimes conglomeratic, sandy sediments. Warm open-sea elements of molluscs commonly occur together. Conjoined valves are quite rare, and at some localities many valves suffered strong abrasion before burial. I have not seen any strictly autochthonous occurrence of fossil Cryptopecten, but judging from the lithology of fossil beds and associated molluscan fossils, the habitat of Cryptopecten in geological ages after Middle Miocene does not seem to be much different from that of the present time.

Two Early Miocene (formerly regarded as Late Oligocene) species, Pecten kyushuensis Nagao, 1928, from the Waita Formation in Fukuoka Prefecture (north Kyushu) and Aequipecten hataii Kanno, 1958, from the Nenokami Formation in Saitama Prefecture (central Honshu), are possibly early representatives of Cryptopecten, but their phy-logenetic relation to Middle Miocene and later species is by no means certain owing to the poorly preserved nature of the material.

Pecten (Aequipecten?) yanagawaensis Nomura and Zinbo, 1936, from the early Middle Miocene (15-16 Ma) Yanagawa Formation in Fukushima Prefecture (north Honshu) is an extinct species safely assignable to Cryptopecten because of the many essential characteristics it shares with C. bullatus and C. vesiculosus. C. yanagawaensis is also known from the Moniwa Formation in Miyagi Prefecture (Nomura, 1940; Masuda, 1958, 1962), the Shukunohora Formation in Gifu Prefecture (Itoigawa, Shibata and Nishimoto, 1974) and the Sunakozaka Formation in Ishikawa Prefecture (Ogasawara, 1976), all in north-central Honshu and early Middle Miocene in age. During early Middle Miocene times (approximately 15-16 Ma) a tropical or subtropical climate predominated, and a characteristic molluscan fauna of warm-current type spread over the Japanese Islands except for the northern part of Hokkaido (Chinzei, 1978). C. yanagawaensis is undoubtedly an element of the warm-water fauna of this age.

After middle Middle Miocene, tropical-subtropical faunas seem to have retreated from the main part of the Japanese Islands, and there are few fossil records of Cryptopecten in the Late Miocene and Early Pliocene sediments. Shikama (1973, pp. 190-191) recorded the occurrence of several specimens of C. yanagawaensis in the basal part of the Zushi Formation, which is probably Late Miocene in age. This occurrence may be important for the present study, but the specimens were not described and unfortunately cannot be found in the collection he bequeathed to the Yokohama National University.

The earliest occurrence of Cryptopecten vesiculosus is found in the Middle Pliocene (approximately 3.5 Ma) calcareous sandstone of the Harada Formation of the Shirahama Group in Shizuoka Prefecture (central Honshu). This species seems to have flourished from Late Pliocene onwards, often constituting a dominant element in the molluscan faunas of Kuroshio type. There are numerous fossil localities in the Upper Pliocene and Pleistocene along the Pacific coast of southwest Japan, and a few small samples were also obtained from some Plio-Pleistocene deposits on the Japan Sea coast. In certain period of Pliocene this species also distributed on the Pacific coast of north Honshu beyond the northern limit of distribution at the present time.

In the Boso Peninsula in Chiba Prefecture (near Tokyo) thick marine deposits of Late Pliocene and Pleistocene are almost successively developed. The molluscan faunas seem to have changed in harmony with the oscillations of warm and cold currents, and there are at least seven horizons yielding C. vesiculosus; namely, Kurotaki Formation (Sample Iy), Umegase Formation (Sample Tm 1, 2), Ichijuku Formation (Sample Ij 1, 2), Sanuki Formation (Sample Sn 1-3), "Higashiyatsu Formation" (Sample Hg), Jiz ô d ô Formation (Sample Ny 1-4, Jz, Ic, Ab 1, 2), Yabu Formation (Sample Sm), Kioroshi Formation (Sample Tn) in upward sequence. The sequence and Stratigraphic position of samples are collectively shown in Fig. 5. Though such a good sequence of marine sediments is rather exceptional, Late Pliocene and Early Pleistocene samples of C. vesiculosus were obtained from the Miura Group (Sample Nj) in Kanagawa Prefecture (central Honshu), the Kakegawa Group (Sample Kg 1, 2) in Shizuoka Prefecture (central Honshu), the Miyazaki Group (Sample Mz 1, 2) in Miyazaki Prefecture (south Kyushu), the Shimajiri Group (Sample Ik, Ob) on Okinawa Island, and a few other areas.

Cryptopecten-bearing fossil beds of Late Pleistocene and younger age seem to be rare in southwest Japan, probably because outer sublittoral sediments have not been raised on land over such a short period. In the inner part of Kagoshima Bay and its surrounding volcanic region (south Kyushu), however, local uplift of an unusual rate must have occurred, and a striking Holocene fossil bed (ca. 2,720 years B. P.) with abundant valves of C. vesiculosus is exposed at Moeshima (=Niijima) Islet (Sample Ms).

The fossil record of Cryptopecten nux in Japan also dates back to the Middle Pliocene. It occurs, though rarely, in the Yonahama Formation of the Shimajiri Group (approximately 3.0 Ma) at Miyako Island, Okinawa Prefecture (Sample Mk (N)), and also in the Late Pliocene Shinzato Formation (Sample Ik (N)) at Ikei Islet and the Early Pleistocene Chinen Formation (Sample Ob (N)) in the southern part of Okinawa Island. At the second and third localities C. nux is found together with C. vesiculosus, C. nux seems in fact to be a long-lived species, the occurrence of some Early Miocene fossils having been confirmed in Tanzania (Eames and Cox, 1956).

Since Nomura and Zinbo (1934) described the rich molluscan fauna of the Ryukyu (=Riukiu) Limestone at the north of Kamikatetsu on Kikai Island, Kagoshima Prefecture, many investigators have visited this locality for various purposes of stratigraphy and paleontology. According to Sakanoue et al. (1967) and Omura (1983), this fossil bed is about 80,000 years old. Nomura and Zinbo reported Pecten (Aequipecten) kikaiensis together with Pecten (Aequipecten) vesiculosus and Pecten (Aequipecten) nux. As a result of my survey on their original specimens in the Tohoku University and newly collected material, I have reached somewhat different taxonomic conclusions. The holotype of P. (A.) kikaiensis (IGPS coll. cat. no. 50357) is evidently conspecific with the numerous newly collected specimens from the same locality (Sample Kk (N)) which are safely assignable to Cryptopecten nux. At the same locality there are many specimens belonging to two other species of Cryptopecten. One is, as Nomura and Zinbo (1934) judged, certainly related to C. vesiailosus but clearly distinguishable from it by the fewer and more spinose radial ribs and weaker convexity of left valve, as described in this paper under the name of Cryptopecten spinosus sp. nov. (Sample Kk (S)). The other species, represented by weakly inflated and thin-shelled specimens (Sample Kk (A)), is identical with Cryptopecten bullatus, though all the specimens are regarded as immature. Incidentally, no fossils of C. bullatus have been recorded from any other locality in Japan (or probably, anywhere else in the world) except for a few comparable specimens from the Late Pliocene Shinzato Formation of Okinawa Island, described by Noda (1980) under the name of "Aequipecten sp."

Yokoyama (1922) described two strongly inflated valves of a small pectinid from "Shitô " [Late Pleistocene Yabu Formation at Ochishimoshinden of Semata, Chiba Prefecture] under the name of Pecten tissoti [should be tissotii]. In the revisory monographs of Yokoyama's illustrated specimens, Taki and Oyama (1954) and Oyama (1973) called it Aequipecten (Cryptopecten) sematensis Oyama. This form is, as described later, here treated as a subspecies of Cryptopecten nux.