A New Specimen of Palaeoloxodon Naumanni from Hokkaido and Its

第四紀研究(The Quaternary Research) 43 (3) p. 169-180 June 2004

A New Specimen of Palaeoloxodon naumanni from

Hokkaido and its Significance

Keiichi Takahashi*1, Yuji Soeda*2, Masami Izuho*3, Kaori Aoki*4, Goro Yamada*2 and Mono Akamatsu*2

This paper describes a new-discovered upper right second molar of Palaeoloxodon naumanni from Yubetsu, Hokkaido, that was found in August 1998, and suggests that alternating migration of two kinds of proboscidean, Mammuthus primigenius and Palaeoloxodon naumanni, took place there in relation to climate change. 14C dating of the root of the molar gives an age of 30,480±220yrs BP(measured 14C age). Although the fossil molar was found loose, geological investigations suggest strongly that it derives from a peaty silt bed distributed around ravine in which the fossil was found. This bed includes the Ds-Oh (Daisetsu-Ohachidaira) volcanic ash of 30ka. Judging from the ages and vegetations of the formations from which P. naumanni or M. primigenius remains have been found in Hokkaido, vegetation change controlled by global climate change seems to have affected the migration of the two kinds of proboscidean into Hokkaido. The discovery of P. naumanni remains of 30ka in Hokkaido suggests the possibility of a northward re-migration of P. naumanni from Honshu during the MIS 3.

Keywords: Mammuthus primigenius, Palaeoloxodon naumanni, climate change, MIS 3, Hokkaido, Late Pleistocene

thus primigenius. I. Introduction In this paper, the Yubetsu specimen is de- In August, 1998, after heavy rains, brothers scribed, and alternating replacement of two Hiroshi and Yasushi Yokoyama were walking kinds of proboscidean, MMprimigenius and P. in a forest in Higashi-baro, Yubetsu, Hokkaido, naumanni, in the northern part of the Japanese looking for edible wild plants. During their Islands is discussed in relation to climatic fluc- search, they found a proboscidean molar in a tuations in the Late Pleistocene. brook (Fig. 1). Thereafter, H. Yokoyama do- II. A note on stratigraphy nated the specimen to the Yubetsu-cho Board of Education. Subsequent studies of the speci- Stratigraphical description of the site of men and a geological survey at the site have occurrence: revealed that the molar fossil is assignable to The Yubetsu specimen was found in a small Palaeoloxodon naumanni and its age is about 30 lying in a small ravine about 150m long (Fig. 1, ka. This specimen is much younger than finds Loc. a). Almost no rock outcrop is present in or of P. naumanni previously reported in Hok- along side the brook, where the land surface is kaido, being within the age range of Mammu- covered with vegetation and talus deposits. On

Received July 9, 2003. Accepted February 21, 2004. *1 Lake Biwa Museum . 1091 Oroshimo, Kusatsu, 525-0001, Japan. E-mail: *2 Historical Museum of Hokkaido . 53-2 Konopporo, Atsubetsu, Atsubetsu-ku, Sapporo, 004-0006, Japan. *3 Sapporo Buried Cultural Property Center . Minami 22, Nishi 13, Chuo-ku, Sapporo, 064-0922, Japan. *4 Geological Survey of Japan , AIST. Tsukuba Central 7, 1-1-1 Higashi, Tsukuba, 305-8567, Japan. 170 Takahashi, K., Soeda, Y., Izuho, M., Aoki, K., Yamada, G. and Akamatsu, M. June 2004 the sidewall of the ravine through which the gravel layer, a primary fall unit of fine volcan- brook runs, about 10m downstream from the ic ash layer (Ash 1), a peaty silt layer, and a fossil locality (Fig. 1, Loc. b), the strata are par- rounded gravel layer (Fig. 2, Column b). Par- tially visible. The outcrop section at Loc. b ticularly, the volcanic ash layer is an important (Fig. 1) reveals several sedimentary units, marker for understanding the stratigraphic listed in ascending order: a subangular small- correlation and for determining the age determination of the deposits in and around the occurrence point of this fossil molar.

Fig. 1 Locality map of the Yubetsu specimen Topographic maps are from 1:50,000 maps of "Engaru" by Geographical Survey Institute of Japan.

Fig. 2 Columnar sections of Locs. a and b (Fig. 1) 2004年6月 Palaeoloxodon naumanni from Hokkaido and Its Significance 171

An artificial outcrop was excavated near the as Aoki and Arai (2000). fossil molar locality (Fig. 1, Loc. a) to clarify Characteristics of tephra: the strata from which it might have been de- Ash 1 is a well-sorted, crystal-rich ash. The rived. The outcrop is about 2m wide and about maximal length of pumice particles contained 3m high, and the strata were similar to those at in it is about 2-3mm, and that of heavy min- Loc. b (Fig. 1),10m downstream. The sequence erals is 1-2mm. The mineral content is as fol- starts from a basement rock of shale corre- lows: light minerals 69%, orthopyroxene 14 sponding to the Yubetsu Group of Cretaceous %, and amphibole 7%. Volcanic glass shards to Paleogene strata (Tajika and Yahata, 1991), comprising 10% of the content are composed of through a subangular small-gravel bed with white and well-bubbled pumice. The refractive thin, peaty silt layers and a peaty silt bed, to an indices of the minerals are as follows: glass (n) angular gravel bed at the top (Fig. 2, Column a). 1.494-1.505, orthopyroxene (γ) 1.705-1.711, and A thin lens of fine volcanic ash, which has a amphibole (n2) 1.673-1.680. close resemblance to the tephra at Loc. b (Fig. Ash 2 is present as a thin lens. It consists of 1), is recognized in the peaty silt bed (Ash 2). well sorted, fine and crystal-rich ash; volcanic Analysis method of tephra: glass shards are very rare because of weather- The procedure for preparation of samples of ing. The refractive indices of its component tephra was as follows: 1) washing by ultra- minerals are as follows: glass (n) 1.498-1.501. sonic cleaner, 2) grain separation by sieving, orthopyroxene (γ) 1.705-1.710, and amphibole and 3) observation of minerals under a stereo- (n2) 1.672-1.679. These petrographic charac- microscope. Volcanic glass type, glass color, ters resemble those of Ash 1. Therefore, both and mineral composition of Ash 1 were identi- ash layers seem to belong to the same tephra. fied under the stereomicroscope. Correlation of tephra: The refractive indices of volcanic glass There are four marker-tephra layers of Late shards and heavy minerals were measured by Pleistocene age in the study area: Daisetsu- RIMS-2000 as described by Yokoyama and Ya- Ohachidaira (Ds-Oh: Nakamura and Hirakawa, mashita (1986) and Danhara et al. (1992). Major 2000), Shikotsu pumice fall 1 (Spfa-1: Katsui, chemical components of the volcanic glass was 1958), Kussharo-Shoro (Kc-sr: Arai et al., 1986) measured with an electron probe micro-ana- and Toetoko(Tsk: Okumura, 1991). The major lyzer (EPMA), following the same procedures volcanic glass of Ash 1 is characterized by high

Table 1 Major element composition of glass shard from Loc. b and Pleistocene marker-tephra samples in the coastal area of the Sea of Okhotsk, eastern Hokkaido 172 Takahashi, K., Soeda, Y., Izuho, M., Aoki, K., Yamada, G. and Akamatsu, M. June 2004

K2O content and low Na2O, TiO2, and FeO* con- remains. Age determination of the Ds-Oh tent (Table 1). These chemical characters most tephra will be a subject for future research, but closely resemble Ds-Oh among the four mark- we can safely conclude that there are some er-tephras. Furthermore, Ash 1 differs from strata of the latest Pleistocene sediments there, both Kc-Sr and Spfa-1 in mineral composition roughly corresponding to MIS 3. and the refractive indices of its minerals. The age of the fossil molar was measured to Discussion of age of sediments: be 30, 520±220yrs BP (Table 2). This age was The Ds-Oh tephra, a marker of late Pleisto- measured by the AMS method on the basis of cene age, was confirmed around the fossil local- collagen extracted from the tooth root, and we ity. In addition, wood fragments and seeds believe that contamination from modern car- from above and below the Ds-Oh bed at both bon dioxide was avoided. Because the fossil loci were measured by AIMS or by a radiomet- molar was found floating on the surface of the ric method (Table 2), and all materials indicate brook, it is difficult to identify correctly the more than 44,000yrs BP, beyond the measur- original strata in which the fossil was pre- able boundary. served; nonetheless, we assume that the fossil Although the Ds-Oh tephra has been re- molar derived from the peaty silt bed or angu- ported to be as old as about 30ka (Katsui et al., lar gravel bed in the vicinity of the fossil local- 1979; Nakamura et al., 1999; Nakamura and ity. Hirakawa, 2000), all ages of wood fragments III. Description of molar and seeds just above and below both ashes (Ds-Oh tephra) are much older than this. The Proboscidea Illiger, 1881 identification of the Ds-Oh tephra seems to be Elephantidae Gray, 1821 correct, and we cannot explain well this dis- Elephantinae Gray, 1821 cordance of age. It may be that the wood Palaeoloxodon Matsumoto, 1924 fragments were derived from older sediments, Palaeoloxodon naumanni (Makiyama, 1924) such as the lower subangular small-gravel bed Material: Right upper second molar, kept at Loc. a (Fig. 1). by Yubetsu-cho Board of Education, Abashiri- Recently, Machida and Arai (2003) revised shicho, Hokkaido (Fig. 3). the age of Ds-Oh as >30ka, and the age of the Locality: Retrieved from a brook in Hi- just underlying marker tephra, Spfa-1, as 40 to gashi-baro, Yubetsu-cho, Abashiri-shicho, Hok- 45ka. Both had been dated as 30ka and 31 to 34 kaido (143°39'39"E, 44°02'06"N). ka, respectively, by Machida and Arai (1992). Horizon: Unknown. This suggests that the age of Ds-Oh is a little Age: AIMS 14C dating of collagen extracted older than 30ka, but the age discordance still from the root of the molar indicates 30, 480±

Table 2 Measured radiocarbon ages from Locs. a, b (Fig. 1) and a fossil molar in Yubetsu, northeastern Hokkaido 2004年6月 Palaeoloxodon naumanni from Hokkaido and Its Significance 173

Fig. 3 Photographs of the Yubetsu specimen, Palaeoloxodon naumanni (right upper M2)

220yrs BP (measured 14C age). Sample number: Table 3 Measurements of the Yubetsu specimen, HB-1 (Laboratory Number: Beta-134800). Palaeoloxodon naumanni The medial portion of the original molar has been lost to attrition, and only the distal portion of the molar remains. The roots are almost lost. Measurements of specimen are given in Table 3, following the methodology of Taka- hashi (1991), slightly modified in part. The size of the whole molar is slightly smaller than the average of previously recorded second molars of P. naumanni. The pericoronal cementum and lingual portion of the first plate are absent. The whole enamel loops are narrow in mesial- distal section. The first and second enamel loops are expanded in the center. In the third loop, the mesial margin projects slightly me- distal surface is flat and slightly depressed on sially, whereas in the fourth and fifth loops, the the root side. mesial margin of the loop runs parallel with Based on its size, this specimen could be the distal one. The sixth loop is constricted on mistaken for a first molar, but the enamel both sides, whereas the seventh loop is con- thickness and the width of the molar indicate stricted on the buccal side only. The eighth that it is actually a second molar. loop is a half plate. IV. Discussion The enamel is relatively thick, and enamel folding is clearly visible in the center part of 1. Ages and habitats of P. naumanni and the enamel loop in each plate. In side view, M. primigenius in Hokkaido plate height is seen to be low, even in the distal Four specimens of P. naumanni hitherto plates. Furthermore, the lateral fissure is not have been found in Hokkaido (Matsumoto, visible in any plate. These features together 1929; Ishikari Lowland Research Group, 1963; indicate that this molar is well worn. The Kamei, 1970, 1978; Kamei et al., 1971, Table 4). 174 Takahashi, K., Soeda, Y., Izuho, M., Aoki, K., Yamada, G. and Akamatsu, M. June 2004

Table 4 Specimen list of Palaeoloxodon naumanni in Hokkaido

One additional specimen from Hokkaido was about 120ka. This was based on a study of the reported by Naora (1954), but it is unclear pollen assemblage and the discovery of the whether it actually originated from there. Kutcharo-Haboro tephra (Kc-Hb), which was Only two of these specimens, the Kuriyama deposited about 110ka (Nakamura and Hi- and Churui specimens, have had their horizons rakawa, 2000), in a horizon 5m above the Chu- reported in detail. rui specimen. This age (120ka) of the Churui The Kuriyama specimen was described first specimen corresponds to MIS 5e, which is gen- by the Ishikari Lowland Research Group (1963). erally considered to have been the warmest It was reported from the lower part of the period during the Last Interglacial Age. Stud- Kakuta Formation which was estimated to ies of pollen assemblages and macro plant fos- consist of sediments of the Wurm Glacial or sils have revealed the presence of deciduous the Riss-Wurm Interglacial age. Subsequently, broad-leafed trees such as Fagus crenata, Styrax Uma-oi Collaborative Research Group (1987) japonicus, and Corylus heterophylla var. thun- examined this area. They concluded that the bergii together with boreal conifers such as Kuriyama specimen had originated from the Abies sachalinensis and Picea flehnii (Yano, late Late Pleistocene Misato terrace deposit, 1972, 1978). These features indicate the vegeta- based on evidence from the stratigraphic and tion around the Tokachi plain at that time to tectonic setting of the formation. Recently, have included mixed boreal coniferous and de- Nojo et al. (2002) reexamined the geological fea- ciduous broad-leafed woodland. This shows tures of this area and reported that the speci- that the temperature of this period was almost men had certainly originated from the Lower the same or a little cooler than at present in Kakuta Formation. Thus, discussions concern- Hokkaido. ing the horizon of the Kuriyama specimen are The Yubetsu specimen reported here was ongoing, and it has not yet been determined dated at about 30ka (Table 2). It is well known clearly. that from about 34ka to 26ka there was a The Churui specimen was discovered in the slight amelioration of MIS 3. During this pe- Horokayantoh Formation. There have been riod, the vegetation of the Ishikari Lowland two different opinions as to the age of this area in western Hokkaido was dominated by formation : the Riss-Wurm Interglacial age deciduous broadleaf trees such as Betula, Al- (Tokachi Collaboration Research Group, 1971; nus, Ulmus, and Quercus, together with Tsuga Matsui et al., 1978) and the Mindel-Riss Inter- and boreal conifers (Igarashi and Kumano, glacial age (Osanai et al., 1971). Akamatsu et al. 1981). This shows that the temperature of this (1990) demonstrated that the specimen's age is period was warmer than that of the preceding 2004年6月 Palaeoloxodon naumanni from Hokkaido and Its Significance 175

or following period. A similar result has been kaido, which is fairly near the locality of the reported based on pollen analysis in the east- Yubetsu specimen. These two examples sug- ern margin of the Konsen Plateau, eastern Hok- gest that an environment of mixed boreal co-

Fig. 4 Locality map of Mammuthus primigenius and Palaeoloxodon naumanni in Hokkaido The locality numbers in the figure are the same as in Tables 4 and 5.

Table 5 Specimen list of Mammuthus primigenius in Hokkaido 176 Takahashi, K., Soeda, Y., Izuho, M., Aoki, K., Yamada, G. and Akamatsu, M. June 2004 niferous and deciduous broad-leafed wood- yama, 1938; NMinato, 1955; Kamei, 1987; Naka- lands existed at about 120ka and about 30ka. ya et al., 1992; Yamada et al., 1996, Fig. 4). On the other hand, nine specimens of M. Among these specimens, six were associated primigenius have also been reported from al- with a horizon or age; however, the ages of the most the same area of Hokkaido where the first and second Ogoshi specimens are uncer- specimens of P. naumanni were found (Maki- tain. The ages of the other four specimens

Fig. 5 Relationship between vegetation and proboscides fossils in the southwestern part of Hokkaido after 60ka Arrows show the horizon of M. primigenius and P. naumanni, respectively. Modified from Igarashi (1993). 2004年6月 Palaeoloxodon naumanni from Hokkaido and Its Significance 177 have been reported as 50-40ka and about 20 however, the newly-colonized area must pos- ka (Table 5). sess a suitable climate and vegetation for the At 50-40ka, the vegetation in Hokkaido in- animals. Even if an uncolonized area becomes cluded taiga forest composed mainly of Larix accessible, the presence of unsuitable vegeta- gmelinii, Picea pumila, and Picea jezoensis. In tion will halt or reverse any migratory trends. the Last Glacial Maximum (about 20ka), the In this context, the Yubetsu specimen may northeastern half of Hokkaido, where the spe- illustrate the converse, a case where a strait is cimens of M. primigenius occurred, was covered narrow and suitable vegetation is continuous, with an open forest of taiga with grassy plains so animals that have a good swimming ability, (Igarashi et al., 1989, 1990; Igarashi, 1993). such as elephants, can migrate over the strait. 2. Climate change and migration of two Such a possibility is supported by the exam- kinds of proboscidean ples the proboscideans populations of the Before the Yubetsu specimen was discov- Channel Islands in California (Roth, 1996) and ered, it had been presumed that P. naumanni of Sulawesi in Indonesia (van den Bergh, 1999). occurred in Hokkaido in the warm period This topic will be discussed in more detail in around 120ka, but that only M. primigenius oc- the future, when additional evidence will have curred there from 50ka to 20ka. Following the been accumulated. discovery at Yubetsu, it became clear that the Acknowledgements ages and the habitats of the two species in Hokkaido were different (Fig. 5). This suggests We thank Dr. Y. Satoguchi and Ms. C. Yama- that these two species were adapted to their kawa, curators of the Lake Biwa Museum, for preferred environments as mentioned above, useful advice on tephrochronology and plant and they migrated either north or south with fossils. Mr. H. Takeuchi, Executive Director of northward expansion of the cool-temperate the Yubetsu Board of Education, and his subor- forests or southward expansion of the open dinates supported us in the geological investi- forest taiga in MIS 3. These fluctuations in gation around the fossil locality. Two land- vegetation zone were a consequence of global owners, Mr. H. Yokoyama and Mr. T. Hase- climate changes. gawa, gave us ready consent to survey in their The presence of P. naumanni in Hokkaido at hills. Dr. T. Soda, Director of the Maebashi 30ka raises a question about the linkage be- Institute, Paleoenvironmental Research Insti- tween Honshu and Hokkaido. The role of the tute Co., Ltd. supported us in the tephra analy- Tsugaru Strait between Hokkaido and Honshu sis. We also thank Dr. Y. Igarashi for informa- Islands as a biogeographical barrier has been tion about vegetation in Hokkaido during the discussed in relation to the maximum drop of Late Pleistocene, and Dr. M. J. Grygier for im- sea level during the last glaciation (Minato, proving English expression. The refractive in- 1966; Oba,1988; Ohshima, 1990). For example, dices of volcanic glass shards and heavy miner- Kawamura (1985, 1989, 1998) proposed the exis- als were measured at Tokyo Metropolitan Uni- tence of an "icebridge" instead of a landbridge versity, Department of Geography. during this period, and insisted that the Tsu- This study has been financially supported by garu Strait has not formed a barrier for large Lake Biwa Museum Joint Research Project mammals with a migratory habit in recent (project no. Joint Research 01-03). times. We ourselves do not have any new data References concerning the landbridge between Hokkaido and Honshu; the discovery of P. naumanni re- Akamatsu, M., Yamada, G., Watabe, M., Ego, M. and mains from about 30ka in Hokkaido, however, Okumura, K. (1990) The age of the Naumann's elephant from Churui, Hokkaido and its vegetation. suggests the possibility of a northward migra- Proc. Hokkaido Branch, Geol. Soc. Japan, no. 1, 37- tion from Honshu during MIS 3. 40. (J) Many arguments have been predicated main- Akiyama, M., Kamei, T, and Nakai, N. (1989) 14C age of ly on the existence of a landbridge that enabled elephant fossils from sea bottom. Jour. Fossil Res., animals to migrate. For successful migration, 22, 22-23. (J) 178 Takahashi, K., Soeda, Y., Izuho, M., Aoki, K., Yamada, G. and Akamatsu, M. June 2004

Aoki, K. and Arai, F. (2000) Late Quaternary tephro- Katsui, Y. (1958) On the fossil forest embedded in the stratigraphy of marine core KH94-3, LM-8 off San- Shikotsu pumice fall deposits, Hokkaido. Jour. riku, Japan. The Quat. Res. (Daiyonki-Kenkyu), 39, Geol. Soc. Japan, 64, 379-385. (J+E) 107-120. (J+E) Katsui, Y., Yokoyama, I. and Ito, T. (1979) Asahi-lake, Arai, F., Machida, H., Okumura, K., Miyauchi, T., Soda, its volcanic geology, history of eruption, present T. and Yamagata, K. (1986) Catalog for late Quater- sates of activity and prevention of disaster. 82p, nary marker-tephras in Japan II-tephras occur- Committee for Prevention of Disasters of Hok- ring in northeast Honshu and Hokkaido-. Geo- kaido. (J) graphical Reports of Tokyo Metropolitan Univ., 21, Kawamura, Y. (1985) Succsession of the mammalian 223-250. fauna in Japan since the Last Glacial Period. The van den Bergh, G. D. (1999) The Late Neogene ele- Earth Monthly, 7, 349-353. (J) phantoid-bearing faunas of Indonesia and their pa- Kawamura, Y. (1989) Quaternary rodent faunas in the laeozoogeographic implications. 419p, Scripta Geo- Japanese Islands (Part 2). Mem. Fac. Sci. Kyoto logica, 117. Univ., Ser. Geol. Min., 54, 1-235. Danhara, T., Yamashita, H. and Kasuya, M. (1992) An Kawamura, Y. (1998) Immigration of mammals into improved system for measuring refractive index the Japanese Islands during the Quaternary. The using the thermal immersion method. Quat. Inter- Quat. Res. (Daiyonki-Kenkyu), 37, 251-257. (J+E) nat., 13/14, 89-91. Machida, H. and Arai, F. (1992) Atlas of tephra in and Hokkaido Group (1969) Quaternary system of Hok- around Japan. 276p, Univ. Tokyo Press. (J) kaido. Assoc. Geol. Coll. Japan Monograph, 15, 1-35. Machida, H. and Arai, F. (2003) Atlas of tephra in and (J+E) around Japan [revised edition]. 336p, Univ. Tokyo Igarashi, Y. (1993) History of environmental changes Press. (J) in Hokkaido from the viewpoint of palynological Makiyama, J. (1938) Japonic Proboscidea. Mem. Coll. research. Higashi, S., Osawa, A. and Kanagawa, K. Sci. Kyoto Imp. Univ. [B], 14, 1-59. (eds.) Biodiversity and ecology in the northern- Matsui, M., Satoh, H. and Kosaka, T. (1978) The age of most Japan: 1-19, Hokkaido Univ. Press. the fossil Naumann's elephant bearing bed. Assoc. Igarashi, Y. and Kumano, S. (1981) Vegetational Geol. Coll. Japan Monograph, 22, 399-408. (J+E) change during the Last Glacial Age in Hokkaido. Matsumoto, H. (1929) On Loxodonta (Palaeoloxodon) The Quat. Res. (Daiyonki-Kenkyu), 20, 129-141. (J+ namadica (Falconer and Cautlay) in Japan. Sci. Rep. E) Tohoku Imp. Univ., 2nd Ser. (Geol.), 13, 1-5. Igarashi, Y., Miyata, Y., Noi, H. and Yamada, O. (1990) Matsuzawa, I. and Kosaka, T. (1987) On the Quater- Fossil pollen and spore assemblages of the Last nary system in the Cape Erimo district-with Glacial Age from the eastern part of the Konsen special reference to the horizon of Mammonteus Plateau, Eastern Hokkaido. The Quat. Res. (Dai- primigenius (M2)-. Prof. Masaru Matsui Memorial yonki-Kenkyu), 29, 131-138. (J+E) Volume Editorial Board (ed.) Professor Masaru Igarashi, Y., Yamada, O. and Matsushita, K. (1989)14C Matsui Memorial Volume: 71-78. (J) age of buried peat from Shinko-cho, northern part Minato, M. (1955) Zu den Mammonteus faunen Ho- of Sapporo, Japan-14C age of the Quaternary de- kkaidos. Japan Jour. Geol. Geogr., 26, 105-113. (G) posits in Japan (171)-. Earth Sc., 43, 186-188. (J) Minato, M. (1966) The final stage of land bridges in Ishikari Lowland Research Group (1963) A new fin- the Japanese Islands. Earth Sci., no. 85 86, 2-11. (J ding of a molar tooth of Palaeoloxodon namadicus +E) naumanni. Earth Sc., no. 69, 42. (J) Minato, M. (1967) On the age of Mammoths in Japan Kamei, T. (1970) Preliminary report on the excavation and Siberia. Earth Sci., 21 (2), 13-17. of Naumann's elephant (Elephas namadicus nau- Nakai, N., Arita, Y., Nakamura, T., Kamei, T., Aki- manni MAK.) at Chuhrui, Hokkaido. Fossil Club yama, M. and Sawada, K. (1991) AMS radiocarbon Bull., no. 3, 1-10. (J+E) ages of mammal fossils from Lake Nojiriko, Na- Kamei, T. (1978) On Naumann's elephant, Palaeoloxo- gano Pref. and environmental changes during the don naumanni (Makiyama) from Churui, Hokkaido. last glacial age. Summaries Researches Using Assoc. Geol. Coll. Japan Monograph, 22, 345-355. AMS at Nagoya Univ., 2, 26-38. (J+E) (J+E) Nakamura, Y. and Hirakawa, K. (2000) Petrographic Kamei, T. (1987) New finds of the wooly mammoths properties of Daisetsu-Ohachidaira tephra, Hok- teeth from the sea bed of the Nemuro strait, Hok- kaido, Japan. Bull. Vol. Soc. Japan, 45, 281-288. (J) kaido, Japan. Prof. Masaru Matsui Memorial Vol- Nakamura, Y., Hirakawa, K. and Naganuma, T. (1999) ume Editorial Board (ed.) Professor Masaru Matsui Tephra at and around a Paleolithic site in the Memorial Volume: 1-12. (J) Shirataki Basin, eastern Hokkaido, Japan. Jour. Kamei, T., Taruno, H. and Kobayashi, I. (1971) On Geogr., 108, 616-628. (J+E) Naumann's elephant from Chuhrui, Hiroo-gun, Nakaya, H., Akamatsu, M. and Yamada, G. (1992) New Hokkaido (preliminary report). Mem. Hist. Mus. late Pleistocene Mammuthus primigenius (Probos- Hokkaido, 1, 27-38. (J) cidea, Mammalia) and Sinomegaceros yabei (Artio- 2004年6月 Palaeoloxodon naumanni from Hokkaido and Its Significance 179

dactyla, Mammalia) from Yuni-cho, Hokkaido, Tajika, J. and Yahata, M. (1991) Explanatory text of Northern Japan. Abstracts 1992 Annual Meeting the geological map of Japan. Scale 1: 50,000 "En- Palaeont. Soc. Japan, 61. (J) garu", 104p, Geological Survey of Hokkaido. Naora, N. (1954) Old stone age in Japan. 298 p, Rep. Takahashi, K. (1991) Measurement method for molar. Dept. Archaeol. Waseda Univ., 2. (J) Kamei, T. (ed.) Japanese Proboscidean fossils: 127- Nojo, A., Takahashi, Y., Tanaka, N., Takahashi, S., 131, Tsukiji-shokan. (J) Hashimoto, C. and Iwamizawa Research Group Tokachi Collaboration Research Group (1971) 14Cage (2002) Stratigraphy and paleoenvironmental changes of the peat bed containing with Elephas namadicus of the Upper Pleistocene in the eastern end area naumanni at Bansei in Tokachi, Hokkaido-14C age of the Ishikari Low Land, Hokkaido, Japan. Earth of Quaternary deposits in Japan (67)-. Earth Sci., Sci., 56, 253-268. (J+E) 25, 187-188. (J) Oba, T. (1988) Comment for sea level change. The Uma-oi Collaborative Research Group (1987) Late Quat. Res. (Daiyonki-Kenkyu), 26, 243-250. (J+E) Pleistocene stratigraphy and paleogeography of Ohshima, K. (1990) The history of straits around the the eastern marginal area of the Ishikari Lowland, Japanese Islands in the Late Quaternary. The Quat. Central Hokkaido, Japan. Earth Sc., 41, 303-319. (J Res. (Daiyonki-Kenkyu), 29, 193-208. (J+E) +E) Okumura, K. (1991) Quaternary tephra studies in the Yamada, G., Akamatsu, M., Nakaya, H. and Kuma- Hokkaido districts, northern Japan. The Quat. Res. saki, N. (1996) AMS-14C date of Mammoth molar (Daiyonki-Kenkyu), 30, 379-390. (J+E) found on off the coast of Rausu, eastern part of Okumura, K. (1993) Wave length dispersive X-ray Hokkaido. Bull. Hist. Mus. Hokkaido, 24, 1-8. (J+E) spectrometry EPMA method. Japan Association Yano, M. (1972) On the plant remains from the fossil for Quaternary Research (ed.) A handbook of Qua- elephant bearing bed in Tokachi Plain, Hokkaido. ternary research 2: 177-189, Tokyo Univ. Press. (J) Earth Sci., 26, 12-19. (J+E). Ono, Y. (1991) The northern landbridge of Japan. Yano, M. (1978) Paleoenvironment of the fossil Nau- Mongoloid, 10, 37-44. (J) mann's elephant bearing bed, inferred from plant Osanai, H., Mitani, K., Uozumi, S., Matsushita, K., remains. Assoc. Geol. Coll. Japan Monograph, 22, Matsunami, T., Nakamura, S. and Shigeyama, T. 389-392. (J+E). (1971) Geology of the horizon containing elephant Yokoyama, T. and Yamashita, T. (1986) Preliminary fossil at Churui. Mem. Hist. Mus. Hokkaido, 1, 53- measurement of refractive index for orthorhombic 62. (J) pyroxene and hornblends by a new refractometer Roth, V. L. (1996) Pleistocene dwarf elephants of the (RIMS-86). Tsukumo Chigaku, 21, 30-36. (J) California Islands. Shoshani, J. H. and Tassy, P. (eds.) The Proboscidea: 249-253, Oxford Univ. (J+E) in Japanese with English abstract, (J) in Ja- Press. panse, (G) in German. 180 Takahashi, K., Soeda, Y., Izuho, M., Aoki, K., Yamada, G. and Akamatsu, M. June 2004

北海道から新しく発見されたナウマンゾウ標本 (こPalaeoloxodon naumanni)とその意義

高橋啓一*1・添田雄二*2・出穂雅実*3・青木かおり*4・

山田悟郎*2・赤松守雄*2

〔要旨〕

1998年8月に北海道網走支庁湧別町の林道脇の沢か定した. ら発見されたナウマンゾウ右上顎第2大臼歯化石の記載今回の標本も含め,これまで北海道で発見されていると,気候変化に伴ってマンモスゾウとナウマンゾウの棲ナウマンゾウとマンモスゾウの産出年代およびその当時み分けが北海道で入れ替わった可能性を報告した.臼歯の植生を考えると,地球規模の気候変動とそれに伴う植化石の年代測定結果は30,480±220yrs BP(未補正14C 生の変化に合わせて,2種類の長鼻類が時期を変えて棲年代値)であった.臼歯が発見された沢には,臼歯化石のみ分けていたことが推定された.同時に,約3万年前の年代とほぼ同じ時代に噴出した大雪御鉢平テフラ(DS- ナウマンゾウ化石の発見は,MIS 3の頃の北海道にナウ Oh)を含む地層が分布していることから,この臼歯はこマンゾウが津軽海峡を渡って来ることができたか,どうの沢に堆積する地層から洗いだされた可能性が高いと推かという議論の一材料を提供することとなった.

*1滋賀県立琵琶湖博物館研究部〒525-0001 草津市下物町1091番地 .E-mailh *2北海道開拓記念館〒004-0006 札幌市厚別区厚別町小野幌53-2 . *3札幌市埋蔵文化財センター〒064-0922 札幌市中央区南22条西13丁目 . *4産業技術総合研究所深部地質環境研究センター〒305-8567 つくば市東1-1-1つくば中央第7 .