The ages of the holotype and a referred molar of Elephas maximus buski described by Matsumoto in 1927, and a molar supposedly of the same subspecies described by Makiyama in 1938 from Higashi Betsuin temple in Nagoya, were investigated by AMS (Accelerator Mass Spectrometry) dating. The holotype (IGPS 7266) may date from any of four periods between 1676 and 1941 cal AD, with 1732–1777 cal AD being the most probable (40.7% likelihood). The referred specimen (IGPS 5845) most likely dates from 1784–1796 cal AD (39.4% probability), and the specimen from Higashi Betsuin from 1454–1494 cal AD (52.9% probability). The present specimens, including the holotype are, therefore, not fossils. Historical records show that Asian elephants did not inhabit Japan at these times. These molars must have been imported into Japan in some fashion during historical times and do not represent a subspecies distinct from extant Asian elephants, E. maximus. Although the nominal subspecies E. maximus buski is clearly invalid, it is not clear which of the three extant subspecies of Asian elephant is its senior synonym in this research.
Introduction
About ten species of fossil elephant have been reported from Japan. Remains of molars of the Asian elephant Elephas maximus Linnaeus, 1758 have also been reported several times (Table 1). The first such report, by Adams (1868), concerned a tooth uncovered more than 40 miles (i.e., more than 64 km) from the seacoast between Kanagawa and Edo (present-day Tokyo). While expressing doubt as to whether this specimen was indeed a fossil, Adams ended up reporting it as one. His paper included a supplemental remark by G. Busk, who indicated that this molar was identified as that of an Asian elephant, but also that it was slightly bigger than an average-sized molar of extant conspecifics.
Matsumoto (1927) later established Elephas indicus buski as a subspecies of Asian elephant, based on a molar found in Ninohe, Iwate Prefecture. Under Article 73.1 of the International Code of Zoological Nomenclature (International Commission on Zoological Nomenclature, 1999), the “type-specimen” so-called by Matsumoto (1927) is the holotype fixed by original designation. He also assigned to this subspecies 1) a left lower third molar bought in Tokyo, 2) the molar reported by Adams (1868), and 3) some molars that he had earlier reported (Matsumoto, 1924) from ? Sapporo in Hokkaido and from Gifu and Wakayama Prefectures. The left lower third molar that he called a “referred specimen” and the other specimens included in this subspecies by Matsumoto (1927) are all paratypes under Article 72.4.5 of the Code, but to avoid ambiguity, we call the former “the referred specimen” herein. Because there was no evident taphonomic modification or “fossilization” of these specimens, Matsumoto (1927) attributed them to the post-Monastirian period, i.e., after the last glacial epoch.
Shikama (1937) and Naora (1944) accepted Matsumoto's (1927) taxonomy, but Takai (1938) excluded E. maximus buski from his Japanese Neogene mammalian list because of the possibility that the specimens had been imported from India in historical times.
Although Makiyama (1938) reported a molar from Higashi Betsuin temple in Nagoya, Aichi Prefecture and another specimen from the riverside of the Yasu River in Shiga Prefecture as E. maximus, he thought that the holotype and other referred specimens of E. m. buski reported by Matsumoto (1927), as well as the specimen reported from Edobashi in Tokyo by Naumann (1881), were actually Palaeoloxodon naumanni (Makiyama, 1924). Sub-sequently, Ikebe and Chiji (1959) reported a molar of E. maximus from Fukae-higashi, Higashinari ward, Osaka, and “Otsuka (1976)” reported a mandible of the same species collected from Sendai River in Tsuruda, Satsuma County, Kagoshima Prefecture. “Otsuka (1976)” was cited by Hasegawa (1977), but its existence could not be confirmed. In his paper listing the specimens of E. maximus from Japan, Hasegawa (1977) noted, concerning those reports, that “most people are negative in recognizing those specimens as fossils”, but he also remarked, “It is strange that there are no record of the transported specimens though such rare things are discovered from the various localities of Japan.”
Table 1.
List of Asian elephants reported from Japan. Five specimens that were reported by Matsumoto (1924) and Shikama (1937) without locality data are omitted. Otsuka (1976), which was cited by in Hasegawa (1977) could not be confirmed.
Figure 1.
Photographs of Asian elephant molars from Japan subjected to redescription and AMS 14C dating in the present study (1). A–C, holotype of Elephas maximus buski (Tohoku University Museum Reg. No. IGPS 7266), left upper first molar; A, lingual side; B, buccal side; C, occlusal side; D–F, specimen from Higashi Honganji Betsuin temple in Nagoya (kept by Aichi High School, unregistered), right lower third molar; D, occlusal side; E, lingual side; F, buccal side.
Okazaki (1982b) suggested that a molar from Hishiikecho, Nishio City, Aichi Prefecture, which had been identified as Euelaphas trogontherii (Pohlig, 1885) by Matsumoto (1924) and as Paraelephas trogontherii (Pohlig, 1885) by Takai (1938), was really E. m. buski. In addition, Okazaki (1982b, 1984) renamed Hasegawa's (1977) specimen that had earlier been reported from the Yasu River (actually collected from the Seri River; Okazaki, 1982b) in Shiga Prefecture as Parelephas sp. Kamei (1991) believed that all specimens identified as E. m. buski should be classified as either the extant E. maximus or extinct P. naumanni.
The general view now seems to be that some of the Asian elephant specimens reported from Japan are not E. maximus at all, while other specimens are indeed from Asian elephants but in some way represent recent artificial transport from foreign lands. However, scientific evidence based on modern techniques has been lacking to confirm this latter hypothesis. In an effort to correct this, we determined the age of the type specimen of E. m. buski, the other specimen from the same locality that was referred to this subspecies by Matsumoto (1927), and also Makiyama's (1938) specimen from Nagoya, by AMS (Accelerator Mass Spectrometry) dating.
Table 2.
Measurements of specimens. Measurements were carried out following Takahashi (1991). The symbol “+” represents more plates at the mesial or distal part of the molar, now lost.
Material
Holotype of Elephas maximus buski Matsumoto, 1927
Tooth class.—Left upper first molar (Figure 1A–C).
Depository.—Tohoku University Museum (Reg. no. IGPS 7266).
Locality.—Ninohe, Iwate Prefecture.
Horizon.—Unknown.
Reference.—Matsumoto (1927).
Description.—Complete molar with abrasion of roughly mesial half of occlusal surface. Surface color dark brown, but part of enamel exposed on occlusal surface white. Coronal cement not complete, but much remaining in situ. Enamel rings of occlusal surface oval and small; crowded enamel folding also seen. Lateral fissure not visible. Five enamel tubercles on each lamella in distal region. Measurements shown in Table 2.
Referred specimen (paratype) of Elephas maximus
buski Matsumoto, 1927
Tooth class.—Left lower third molar (Figure 2A–D).
Depository.—Tohoku University Museum (Reg. no. IGPS 5845).
Locality.—Unknown (purchased in Tokyo).
Horizon. —Unknown.
Reference.—Matsumoto (1927).
Description. — Complete molar with abrasion of mesial part, cut artificially at midpoint of mesio-distal length. Surface color dark brown. Coronal cement almost complete. Dentin and cementum milky-white, and enamel white, as seen in section, indistinguishable from those of extant elephant molars. In lateral view, lamellae appearing somewhat compressed in mesial-distal direction. Facet for anterior molar present on mesial side. Folding relatively strong in enamel rings of occlusal surface. In section, enamel rings oval with strong folding. Measurements shown in Table 2.
Figure 2.
Photographs of Asian elephant molars from Japan subjected to the redescription and AMS 14C dating in the present study (2). A–D, referred specimen (paratype) of Elephas maximus buski, left lower third molar (Tohoku University Museum Reg. No. IGPS 5845); A, occlusal side; B, cutting plane; C, buccal side; D, lingual side.
Specimen from Higashi Betsuin temple in Nagoya
Tooth class.—Right lower third molar (Figure 1D–F).
Depository.—Ai chi High School (no registration number).
Locality.—Shinshu Otani-ha Nagoya Betsuin (also known as Higashi Betsuin) temple, Tachibana, Naka ward, Nagoya, Aichi Prefecture.
Horizon.—Unknown (2 m below ground).
Reference.—Makiyama (1938).
Description.—Complete, rather narrow molar with abrasion of mesial part. Surface color dark brown, but dentin and enamel seen in a broken lamella surface white and milky-white, respectively, indistinguishable from those of extant Asian elephant molars. Coronal cement absent. Enamel rings on occlusal surface oval in most mesial part of tooth, where abrasion most advanced, but divided into three equal sections by constrictions in third lamella, where abrasion less advanced. Loxodont sinus absent from all enamel rings. Several enamel tubercles present on distal parts of non-abraded lamellae. Measurements shown in Table 2.
Table 3.
Results of AMS 14C dating of three molars attributed to Elephas maximus buski by Matsumoto (1927) or to E. maximus by Makiyama (1938). PLD is an analysis number of Paleo Lab Co., Ltd. The measured radiocarbon ages are unadjusted values before correction by the extent of isotopic fractionation on the 13C/12C ratio. Calendar-calibrated dates are provided along with the conventional 14C ages.
Methods and results of AMS 14C dating
In order to take samples for dating, we visited the specimen depositories and ground the molars using a rotary micromotor for dental laboratories. The pretreatment and AMS 14C dating were performed by Paleo Lab Co., Ltd., as follows: To remove contamination of organic matter from the sample, surfaces were cleaned using an ultra-sonic cleaning device. The washed samples were leached in 0.2 M NaOH at 4°C for 12 h. After neutralization and drying the reacted samples were pulverized using a mortar and pestle. The sample powders were decalcified in 1.2 M HC1 at 4°C for 12 h in a cellulose tube, and were subsequently gelatinized in distilled water acidified to pH 3.0 at 90°C for 12 h. The gelatinized samples were cleaned using a glass filter, and the dissolved gelatin was recovered by freeze-drying. To measure the atomic ratios of carbon and nitrogen and δ13C and δ15N values, extracted collagen was analyzed using an elemental analyzer coupled to an isotope ratio mass spectrometer. The isotope ratios were calculated based on the difference from reference materials of carbon and nitrogen in per mille units. To synthesize graphite for radiocarbon dating, CO2 gas from the gelatin was purified, then reduced to graphite with hydrogen over iron powder in a vacuum line (Minagawa et al., 1984; Kitagawa et al., 1993). The 14C content in the synthesized graphite was measured by AMS using a compact accelerator mass spectrometer (NEC 1.5SDH). The calender age is obtained in OxCal 4.2 with IntCal 13 (Ramsey, 2009; Reimer et al., 2013). A further measurement of the C/N profile of each sample was also done, with values falling within the range (2.9–3.6) typically exhibited by bone collagen (cf. DeNiro, 1985) (Table 2). This indicates a low probability of exogenous contamination. The holotype of E. m. buski (IGPS 7266) may date from any of four periods between 1676 and 1941 cal AD, with 1732–1777 cal AD having the highest probability (40.7%). The referred specimen (IGPS 5845) similarly had a most probable date range of 1784–1796 cal AD (39.4% likelihood), and that of the molar from Higashi Betsuin temple in Nagoya was 1454–1494 cal AD (52.9% likelihood) (Table 3).
Discussion
The three examined specimens of Elephas maximus buski that are stored at the Tohoku University Museum and at Aichi High School are indistinguishable from the extant Asian elephant, E. maximus, in size and morphological characters such as the extent of enamel folding. In addition, while the surface coloration of these specimens is dark brown, inside the enamel is white and the dentine and cementum are milky white as mentioned above, just as in extant Asian elephants. These observations indicate at least that the three examined molars are not fossils; instead, they may belong to or be very closely related to the extant Asian elephant.
AMS 14C dating of these three specimens indicates various ranges of possible age, but the holotype most likely dates from 1732–1777 cal AD, the referred specimen most likely from 1454–1494 cal AD, and the specimen from Nagoya most likely from 1454–1494 cal AD. Such dates indicate that the elephants lived in historical times, specifically from the Muromachi to the middle of the Edo Periods of the Japanese history (the beginning the 14th century to the middle of the 17th century). However, there is no mention of wild Asian elephants inhabiting Japan in any historical document: These three specimens, either as living elephants or as isolated teeth, must have been imported artificially into Japan from foreign lands. According to Isono (2007), there are six records of living Asian elephants brought into Japan from overseas before the Meiji Period, i.e., before 1868. The many discoveries of Asian elephant molars listed in Table 1 suggest that unrecorded Asian elephant molars were imported as well, perhaps as medicinal goods or gifts.
Yamada (1966) reported that the Asian elephant molar was discovered underground in the foundation of Higashi Betsuin temple in Nagoya by digging near a wall that had collapsed in the Great Ansei Earthquake in 1854. However, that temple, officially known as Shinshu Otaniha Nagoya Betsuin, was established in 1690, considerably later than the AMS 14C date of the molar there. Earlier, a castle belonging to Nobuhide Oda (father of the powerful 16th century samurai warlord Nobunaga Oda) stood at the same site. This castle is thought to have been built in 1534, so the molar evidently predates it, too.
The origin of molars assigned to Elephas maximus buski has been unclear. The results of this study show that the “type” and referred specimen that Matsumoto (1927) considered to be a new fossil subspecies of elephant actually belonged to Asian elephants that lived in the 18th century. The nominal subspecies E. m. buski is not based on fossils after all, but on artificially transported specimens of extant Asian elephant. This subspecies is, therefore, invalid. Three extant subspecies are known in E. maximus. As the specimens studied in this research are all isolated teeth, it is difficult to identity to which subspecies the specimens belong. As previously pointed out, AMS dating also supports the inference that Asian elephant fossils from Japan may be artificial transports from foreign lands.
Acknowledgements
We thank Masataka Matsumoto, headmaster of Aichi Junior High School and High School, Haruyuki Kono, office manager of Aichi High School, and Jun Nemoto, Division of Geoenvironmental Science, Faculty of Science, Tohoku University who gave us the opportunity to examine specimens and provided facilities for our work. We also thank Mark J. Grygier, Lake Biwa Museum for revising the English of an earlier draft of this manuscript.
