RESEARCH ARTICLE
A new species of Eubostrychoceras (Ammonoidea, Nostoceratidae) from the Santonian (Upper Cretaceous) of Hokkaido, Japan
2025 年 29 巻 p. 182-198
詳細
2025 年 29 巻 p. 182-198
A new species of nostoceratid ammonoid, Eubostrychoceras perplexum sp. nov., is described from the Santonian (Upper Cretaceous) in the Tappu, Kotanbetsu, and Haboro areas of Hokkaido, northern Japan. The diagnostic features of E. perplexum sp. nov. are tightly coiled, slowly enlarging whorls with a small apical angle, a small umbilicus, and a rounded rectangular whorl section. In addition to Hokkaido, the new species is also recorded from Wakayama, southwestern Japan. Given that the occurrence of this new species might be limited to the uppermost Santonian, it may thus serve as a useful indicator of stratigraphic correlation in the northwestern Pacific region. Although the new species cannot be assumed to have a direct phylogenetic relationship with any of the described Eubostrychoceras species, Hyphantoceras orientale (Yabe) is established to have multiple characters in common with E. perplexum sp. nov., and both species occur sympatrically in the upper Santonian. However, as E. perplexum sp. nov. lacks rows of tubercles, it does not satisfy the definition of the genus Hyphantoceras and accordingly cannot be placed within this genus. Nonetheless, the plausibility of a phylogenetic relationship between the two species cannot be ruled out. This contradiction between taxonomic evaluation and phylogenetic estimation raises questions regarding the current taxonomic criteria used to classify genera within the family Nostoceratidae.
ZooBank registration: urn:lsid:zoobank.org:pub:A4DCCCF7-3AD9-4B8F-800D-9BCB95F48788
The Late Cretaceous heteromorph ammonoid family Nostoceratidae Hyatt, 1894 is characterized by helically coiled whorls in the main part of the shell. Other forms (e.g. straight, planispirally coiled, and hamitoid) are present during the early stages of growth, and in some cases, modifications can occur in the later stages of growth (Matsumoto, 1967; Wright et al., 1996; Schaffert and Larson, 2021). In addition to these basic definitions, derived forms with three-dimensional meandering shells lacking a helically coiled whorl in the main part are also included (e.g. Nipponites Yabe, 1904 and Ryuella Klinger and Kennedy, 1997). Within the Nostoceratidae, Eubostrychoceras Matsumoto, 1967 is characterized by a helically coiled shell ornamented with ribs, sometimes constrictions, and no tubercles (Matsumoto, 1967, 1977). Species of Eubostrychoceras have been recorded from the Turonian–Campanian in different regions worldwide (e.g. Usher, 1952; Kaplan and Schmid, 1988; Klinger and Kennedy, 2003), and in the northwestern Pacific region, including the Japanese islands, numerous species of Eubostrychoceras have been established to occur continuously in the marine strata of different areas (Yabe, 1904; Matsumoto, 1967, 1977; Obata et al., 1991; Misaki and Maeda, 2010; Aiba et al., 2017, 2022; Muramiya et al., 2025).
To date, the systematic classification and estimation of phylogenetic relationships of taxa within Nostoceratidae have generally placed more emphasis on ornamentation rather than shell shape (e.g. Yabe, 1904; Matsumoto, 1967, 1977; Matsumoto and Muramoto, 1967). In this regard, innovative advances in the theoretical morphology of Nostoceratidae, notably those based on the work of Okamoto (1984, 1988a, b, c, 1989), have provided further support for this view of classification and phylogenetic reconstruction. For example, Okamoto (1989) theoretically demonstrated that the two genera Eubostrychoceras and Nipponites, which are characterized by similar ornamentation but different coiling patterns, are phylogenetically closely related.
Numerous macrofossils collected within the framework of biostratigraphic studies are housed in the Mikasa City Museum in Mikasa, Hokkaido, Japan (e.g. Kurihara and Hirano, 2003; Funaki and Hirano, 2004; Oizumi et al., 2005; Kurishima et al., 2013). In 2016, among the molluscan fossils in this repository collected by Oizumi et al. (2005), the author identified one specimen that was likely to be referrable to Eubostrychoceras from the Upper Cretaceous strata in the Tappu area, northwestern Hokkaido. Subsequently, similar heteromorph ammonoid specimens have been collected by Akio Tomita, Masatoshi Iwata, Harunobu Yamato, Koichi Fukuoka, and Yusuke Muramiya and the author. In this paper, these specimens have been described, their relationships with other species have been discussed, revealing new aspects of the Late Cretaceous ammonoid fauna of the Northwest Pacific region.
The Cretaceous fore-arc basin deposits of the Yezo Group are widely distributed over an span of more than 1,000 km in a north-south direction from Sakhalin in the Russian Far East to Hokkaido in northern Japan (e.g. Matsumoto, 1954; Vereshchagin, 1977; Takashima et al., 2004; Shigeta and Maeda, 2005). This group is well exposed in the Haboro, Kotanbetsu, and Tappu areas of northwestern Hokkaido, where the stratigraphy has been studied in detail (Igi et al., 1958; Tsushima et al., 1958; Tanaka, 1963; Yamaguchi and Matsuno, 1963; Hashimoto et al., 1965; Tanabe et al., 1977; Sekine et al., 1985; Toshimitsu, 1985, 1988; Toshimitsu and Maiya, 1986; Toshimitsu and Kikawa, 1997; Toshimitsu et al., 1998; Wani and Hirano, 2000; Moriya and Hirano, 2001; Okamoto et al., 2003; Funaki and Hirano, 2004; Oizumi et al., 2005; Tsujino, 2009; Kawabe and Okamoto, 2012; Honda and Hirano, 2014). Although different divisions of the Yezo Group have been proposed for the Haboro, Kotanbetsu, and Tappu areas, for the purposes of this study, following Takashima et al. (2004), the Yezo Group in these areas is divided into the Shuparogawa, Maruyama, Hikagenosawa, Saku, Haborogawa, and Hakobuchi formations in ascending order.
The 1,750–2,250-m-thick Haborogawa Formation is composed of mudstone and sandstone and characterized by at least three cycles of upward coarsening deposits, which Okada and Matsumoto (1969) referred to as the ‘Haboro cycle’ (Okamoto et al., 2003; Wani, 2003; Takashima et al., 2004). On the basis of the biostratigraphy of inoceramid bivalves and ammonoids, the depositional age is estimated to be the late Turonian–early Campanian (Takashima et al., 2004). The Haborogawa Formation corresponds to the lithostratigraphic units Ua–Uk defined by Tsushima et al. (1958) and Igi et al. (1958), among which the 200–450-m-thick Uh Unit consists of sandy siltstone and very fine- to coarse-grained sandstone containing green minerals and mushroom- or irregular-shaped concretions, which yield abundant molluscan macrofossils, particularly Plesiotexanites kawasakii, Hyphantoceras orientale, and Heteroptychoceras obatai (Figure 1; Wani and Hirano, 2000; Okamoto et al., 2003; Wani, 2003; Oizumi et al., 2005; Aiba, 2019). In addition, the Uh Unit is correlated with the Santonian Inoceramus amakusensis Zone (Toshimitsu et al., 1995, 1998, 2007). Given that the Uf and Ug units are correlated with the Santonian Inoceramus amakusensis Zone, whereas the Ui-j Unit is correlated with the lower Campanian Platyceramus japonicus Zone (Toshimitsu et al., 1995, 1998, 2007), the Uh Unit probably represents the upper part of the Santonian (e.g. Maeda, 1993; Okamoto et al., 2003; Aiba, 2019).
The 13 specimens described herein were collected from the Haborogawa Formation in the Haboro, Kotanbetsu, and Tappu areas (Figure 1; Table 1). Three of these specimens (MCM-M0263, MCM-M0298, and MCM-W0329-1) were retrieved from in situ calcareous concretions in the Uh Unit of the Kotanbetsu and Tappu areas, and eight specimens (MCM-A1984, MCM-A1995, MCM-A2026, MCM-A2032, MCM-A2254, MCM-M0277, MCM-M0299, and MCM-M0310) were collected from float calcareous concretions in the Haboro, Kotanbetsu, and Tappu areas, which almost certainly originated from the Uh Unit based on the stratigraphic distribution upstream and the lithology of the matrix. Fukuoka (2000) described the locality of MCM-A1995 as the Haboro River in the Haboro area, but this is incorrect. The correct locality is the Kotanbetsu River in the Kotanbetsu area. Precise sampling information is lacking for the remaining two specimens; however, one of them (KYUM GKH 20025) is recorded as having been collected from the Santonian in Akanosawa Creek in the Tappu area. It is, nevertheless, uncertain as to whether it was retrieved from float or in situ concretions, although based on the stratigraphic distribution of the river and the lithology of the matrix, it is probable that this specimen was derived from the Uh Unit. The other specimen (SMAC3853) is recorded as having been collected from the Santonian in the Kotanbetsu area (Hayakawa, 2003), and on the basis of the lithology of its matrix, it is probable that this specimen was also derived from the Uh Unit. Plaster casts of KYUM GKH 20025 and SMAC3853 are housed in the Mikasa City Museum (catalogued as MCM-M0296 and MCM-M0297, respectively).
| specimen number | type | locality | collector | number of whorls | a (mm) | b (mm) | c (mm) | d (°) |
|---|---|---|---|---|---|---|---|---|
| MCM-A1984 | P | Haborogawa C., Haboro area (f) | H. Yamato | 3 | 27.6 | 4.9 | 7.6 | 14 |
| MCM-A1995 | P | Kotanbetsugawa R., Kotanbetsu area (f) | K. Fukuoka | 6 | 48.2 | 4.6 | 8.5 | 13 |
| MCM-A2026 | P | Akanosawa C., Tappu area (f) | M. Iwata | 3 ½ | 22.2 | 2.8 | 6.0 | 15 |
| MCM-A2032 | H | Akanosawa C., Tappu area (f) | A. Tomita | 3 ½ | 55.2 | 8.2 | 14.4 | 10 |
| MCM-A2254 | P | Branch of Kaminosawa C., Kotanbetsu area (f) | Y. Muramiya | 5 | 33.7 | 2.1 | 6.4 | 12 |
| MCM-M0263 | P | Akanosawa C., Tappu area (in situ) | Author | 1 ½ | 9.6 | 2.5 | 3.6 | – |
| MCM-M0277 | P | Kaminosawa C., Kotanbetsu area (f) | Author | 5 | 35.5 | 2.5 | 7.4 | 14 |
| MCM-M0298 | P | Kaminosawa C., Kotanbetsu area (in situ) | Author | 2 ½ | 15.7 | 2.5 | 5.3 | 15 |
| MCM-M0299 | P | Branch of Kaminosawa C., Kotanbetsu area (f) | Author | 6 | 46.1 | 2.8 | 9.7 | 10 |
| MCM-M0310 | P | Akanosawa C., Tappu area (f) | Author | 3 ½ | 23.9 | 2 | 5.7 | 12 |
| MCM-W0329-1 | P | Akanosawa C., Tappu area (in situ) | M. Oizumi | 1 | 11.5 | 5.3 | 6.5 | – |
| KYUM GKH 20025 | P | Akanosawa C., Tappu area (–) | M. Yamashita | 4 ½ | 37.5 | 4.9 | 10.0 | 11 |
| SMAC3853 | P | Kotanbetsu area (–) | H. Hayakawa | 3+shaft | 31.1 | 0.9 | 4.8 | 15 |
| WMNHGe-1140210361 | – | T6, Aridagawa area | A. Misaki | 2 | 58.2 | 20.3 | 24.0 | 10 |
In addition, one specimen (WMNHGe-1140210361) from the Santonian–Campanian interval of the Futakawa Formation, Sotoizumi Group in the Aridagawa area, Wakayama, southwestern Japan previously reported by Misaki (2016) was examined (for a detailed stratigraphy, refer to that paper).
This systematic description follows the classification established by Wright et al. (1996), with the exception of the treatment of Eubostrychoceras (see below). The measurement points and some morphological terms are shown in Figure 2.
Institutional abbreviations.—KYUM GKH: The Kyushu University Museum, Fukuoka, Japan; MCM: Mikasa City Museum, Hokkaido, Japan; SMAC: Sapporo Museum Activity Center, Hokkaido, Japan; WMNH: Wakayama Prefectural Museum of Natural History, Wakayama, Japan.
Suborder Ancyloceratina Wiedmann, 1966
Superfamily Turrilitoidea Gill, 1871
Family Nostoceratidae Hyatt, 1894
Genus Eubostrychoceras Matsumoto, 1967
Type species.—Eubostrychoceras indopacificum Matsumoto, 1967.
Diagnosis (modified after Matsumoto, 1967, 1977).—The shell is mainly characterized by helically coiled whorls. The tightness of the coiling varies, ranging from tightly coiled forms, in which the whorls are in contact (torticonic), to loosely coiled forms, in which the whorls are completely separated, similar to a corkscrew. Some of the whorls are extremely elongated, approaching a nearly straight form. During the early stages of growth, straight or planispiral coiled parts may be present, whereas during the later stages of growth, the direction of coiling changes, becoming hook-shaped in some cases. The ribs are typically slightly oblique and gently sinuous, occasionally branched or inserted. Some species are characterized by periodic constrictions during growth and flared ribs in the later stages. At all growth stages, however, rows of tubercles are absent. The siphuncle is positioned approximately at the center of the whorl, although may deviate slightly from this position, and the suture line is of a deeply and distinctly incised lytoceratid type.
Comparisons.—Several genera within the family Nostoceratidae exhibit helically coiled whorls similar to those of Eubostrychoceras, but each can be distinguished from Eubostrychoceras by the following characters. Nostoceras Hyatt, 1894: two or more rows of tubercles are present throughout growth; Didymoceras Hyatt, 1894 and Bostrychoceras Hyatt, 1900: two rows of tubercles develop to varying degrees; Hyphantoceras Hyatt, 1900: three to four rows of tubercles develop; and Yezoceras Matsumoto, 1977: two rows of tubercles develop in the main part of growth and the siphuncle is located at the base of the whorls (Matsumoto, 1967, 1977; Wright et al., 1996; Misaki and Tsujino, 2021).
Discussion.—The tightly coiled form of some Eubostrychoceras species closely resembles the shell morphology of Turrilitoides Spath, 1923, a non-tuberculated genus within the family Turrilitidae Gill, 1871, and it has been suggested that Eubostrychoceras may have originated from Turrilitoides (Matsumoto, 1967, 1977; Wright et al., 1996). Although there are relatively few characters that can be used to distinguish the two genera (Matsumoto, 1977), they can be differentiated based on following features. Turrilitoides lacks an initial straight shaft, whereas Eubostrychoceras includes species with such a morphology. Furthermore, in Turrilitoides, a constriction develops only at the matured aperture, whereas Eubostrychoceras includes species in which constrictions develop during growth.
Remarks.—Although Wright et al. (1996) treated Eubostrychoceras as a subgenus of Nostoceras, the latter is clearly distinguished from the former by the presence of two or more rows of continuous tubercles. Indeed, since its publication, numerous authors have treated Eubostrychoceras as an independent genus (e.g. Klinger and Kennedy, 2003; Misaki and Maeda, 2010; Aiba et al., 2017). Accordingly, in the present study, I treat Eubostrychoceras as an independent genus in the family Nostoceratidae.
Eubostrychoceras perplexum sp. nov.
Figures 2, 3, 4, 5, 6, 7, 8, 9
ZooBank lsid: urn:lsid:zoobank.org:act:0C947F01-23D8-48ED-8C88-1918834D5F3C
Nostoceras (Eubostrychoceras) sp. e. Fukuoka, 2000, p. 214, upper row, the second from right.
“undescribed heteromorph ammonoid” Hayakawa, 2003, p. 137, fig. 3-20.
Nostoceratidae gen. et sp. indet. Misaki, 2016, p. 129, fig. 3G.
Type specimens.—MCM-A2032 (Figures 3A–D, 8I) is designated as the holotype. MCM-A1984 (Figure 6J–M), MCM-A1995 (Figures 5E–H, 8E), MCM-A2026 (Figures 4J–M, 8C, D), MCM-A2254 (Figure 6E–H), MCM-M0263 (Figure 4F–I), MCM-M0277 (Figure 6A–D), MCM-M0298 (Figure 5I–L), MCM-M0299 (Figures 5A–D, 8G–H), MCM-M0310 (Figure 4A–D), MCM-W0329-1 (Figure 4E), KYUM GKH 20025 (Figures 4N–O, 8F), and SMAC3853 (Figures 6I, 8A–B) are designated as paratypes. Table 1 summarizes locality information and specimen sizes.
Type locality.—Float in Akanosawa Creek in the Tappu area of northwestern Hokkaido, northern Japan.
Material examined.—In addition to the type specimens derived from the Yezo Group, one specimen from the uppermost Santonian–lower Campanian in the Futakawa Formation, Sotoizumi Group in the Aridagawa area, Wakayama, southwestern Japan that was identified as Nostoceratidae gen. et sp. indet. (WMNH-Ge-1140210361) by Misaki (2016, fig. 3G) was examined (Figures 7, 8J; Table 1).
Diagnosis.—Eubostrychoceras with tightly coiled slowly enlarging torticonic whorls, small apical angle, small umbilicus, and rounded rectangular whorl section in the main part. In the earlier part, whorls are straight, followed by loosely coiled whorls, with a circular whorl section. Ornamentation consists of slightly curved sigmoidal normal ribs, with modest periodic constrictions appearing as the shell grows. The siphuncle is located at the upper part of the whorls.
Etymology.—The specific epithet is derived from the Latin word perplexus, meaning confused, and indicates that this species shares characteristics with species of another genus, resulting in a contradiction between taxonomic evaluation and phylogenetic estimation.
Description.—The holotype MCM-A2032 consists of three and half volutions of tightly coiled torticonic whorls with a small umbilicus and an apical angle of 10° (Figure 3). The whorl section is a rounded rectangular shape (Figure 2B). The ornamentation consists of slightly curved sigmoidal normal ribs and modest periodic constrictions in the later part of the preserved whorls (Figures 3, 7I, 8). The constriction grooves are asymmetrical, with a moderate slope on the adapical side and a steeper slope on the adoral side, resulting in slight serration of the succeeding ribs. The periodic constrictions are observed MCM-A1995 and WMNH-Ge-1140210361 (Figures 5E–G, 7, 8). The siphuncle is located in the upper part of the whorls (Figure 2B). The suture line comprises a deeply incised bifid lateral lobe and shallowly incised, bifid, asymmetrical saddles (Figure 2C).
The earliest whorl, including the initial chamber and ammonitella, is not preserved in any examined specimen, although parts of earlier whorls are preserved in SMAC3853, MCM-M0310, MCM-M0298, and MCM-A2254 (Figures 4A–D, 5I–L, 6E–I). These earlier whorls are straight, followed by loosely, helically coiled sinistral whorls with a small umbilicus. As the shell grew, the helically coiled sinistral whorls became more tightly coiled, touching each other, the umbilicus became smaller (becoming torticonic), and the whorl section gradually changed from circular to a rounded rectangular shape. In the examined specimens, all of which are sinistral, the apical angle of the tightly coiled whorls ranges from 10° to 15°.
Measurements.—The measured shell morphologies are shown in Figure 2A, with the corresponding values being summarized in Table 1.
Comparison.—The tightly coiled whorls of Eubostrychoceras perplexum sp. nov. are clearly distinct from those in species that have more loosely coiled whorls: E. japonicum (Yabe, 1904, p. 17, pl. 3, fig. 8), E. otsukai (Yabe, 1904, p. 14, pl. 3, fig. 9, pl. 4, figs. 1–3), E. protractum (Collignon, 1969, p. 31, pl. 29, figs. 2069, 2070), E. zulu Klinger and Kennedy (2003, p. 229, figs. 2A, 63F), and E. valdelaxum Aiba et al. (2017, p. 257, figs. 3–6). Eubostrychoceras perplexum sp. nov. can also be distinguished from E. japonicum and E. valdelaxum by its rounded rectangular whorl section.
Eubostrychoceras saxonicum (Schlüter, 1875, p. 30) (see also, Schlüter, 1876, p. 135, taf. 35, fig. 10; Kaplan and Schmid, 1988, taf. 1–3; Breitkreutz and Metzdorf, 1991, abb. 3–9), E. muramotoi Matsumoto (1967, p. 335, pl. 19, figs. 1, 2), and E. matsumotoi Cobban (1987, p. A2, pl. 1, figs. 1–26) all have relatively small torticonic shells, including an earlier straight part. However, the early straight part in these three species differs from that in the new species in being reversed in growth direction and involved in the subsequent helically coiled whorls. Additionally, these species can be distinguished from E. perplexum sp. nov. by a wider umbilicus and a larger apical angle (20°–30°, 50°–60°, and 65° respectively).
Eubostrychoceras indopacificum Matsumoto (1967, p. 333, pl. 18, fig. 1) is similar to E. perplexum sp. nov. in that it has torticonic shells. However, the holotype of E. indopacificum is a poorly preserved specimen. Additional specimens from an area nearby the type locality (Fukushima, northeastern Japan), South Africa, and Madagascar have been reported (Klinger and Kennedy, 2003, fig. 1 D–G; Klinger et al., 2007, fig. 10A–C; Muramiya et al., 2025, fig. 4), and these include specimens of a size comparable to that of the holotype of the new species. Comparing the holotype of E. perplexum sp. nov. with these E. indopacificum specimens reveals that the latter are characterized by a slightly less tight coiling and a wider umbilicus than seen in the former. In addition, during ontogeny, the apical angle of E. indopacificum changes from approximately 40° to 15°, whereas in E. perplexum sp. nov. there is comparatively little change in the apical angle throughout growth, remaining between 10° and 15°.
Eubostrychoceras nibelae Klinger and Kennedy (2003, p. 227, fig. 1H, I) is also similar to E. perplexum sp. nov. in that it is characterized by torticonic shells. However, E. nibelae can be distinguished from the new species on the basis of a slightly less tight coiling, a wider umbilicus, and a larger apical angle (approximately 65°). Notably, however, the description of E. nibelae was based on a single specimen, and it is conceivable that this individual might represent the juvenile shell of a different species, such as E. indopacificum.
Eubostrychoceras elongatum (Whiteaves, 1903, p. 331, pl. 44, fig. 2) can be distinguished from the new species by its sharp, coarse ribs. E. elongatum exhibits intraspecific variation and ontogenetic changes in coiling, including both torticonic and helically coiled shells with separated whorls, and includes both sinistral and dextral shells (Collignon, 1969; Ward, 1976; Haggart, 1989; Misaki and Maeda, 2010), which also differ from those of the new species.
Eubostrychoceras densicostatum Matsumoto (1977, p. 332, pl. 52, fig. 2), which has slightly looser coiled whorls and a larger umbilicus, differs from the new species in these characteristics. Eubostrychoceras medinai (Olivero, 1988, p. 264, fig. 4) has relatively tightly coiled whorls, although in contrast to E. perplexum sp. nov., these do not come into contact with each other, and the rib density is lower in this species.
Compared with the described specimens of E. perplexum sp. nov. from the Yezo Group, the specimen figured as Nostoceratidae gen. et sp. indet., WMNH-Ge-1140210361 by Misaki (2016, fig. 3G) derived from the uppermost Santonian–lower Campanian in the Futakawa Formation, Sotoizumi Group in the Aridagawa area, Wakayama, southwestern Japan (Figure 7), is relatively large, although is similar with respect to its torticonic shell and slightly curved sigmoidal ribs. However, given the poor preservation of this specimen, detailed observations of the ornamentation are not possible. However, no obvious tubercles are visible, and some grooves between the ribs are modestly deeper, resembling periodic constrictions. On the basis of these comparisons, this specimen is accordingly assigned to the new species.
Discussion.—Eubostrychoceras perplexum sp. nov. has common features with some species that are not assigned to the genus Eubostrychoceras. Turrilitoides hugardianus (d’Orbigny, 1842, p. 588, pl. 147, figs. 9–11) is similar to E. perplexum sp. nov. in having a torticonic shell, but it differs in having a circular whorl section (e.g. Jattiot et al., 2021, fig. 22B, AI, AO), whereas E. perplexum sp. nov. has a slightly elongated, rounded rectangular whorl section. The slightly larger umbilicus of T. hugardianus (d’Orbigny, 1842) further distinguishes it from E. perplexum sp. nov., which has a smaller umbilicus. Additionally, an initial straight part has not been observed in T. hugardianus, and a constriction develops only at the matured aperture. These might also be a distinguishing characteristic. Eubostrychoceras perplexum sp. nov. has the following characteristics in common with Hyphantoceras orientale (Yabe, 1904, p. 19, pl. 3, fig. 7): a rounded rectangular whorl section, mostly sinistral whorls, slightly curved sigmoidal ribs, and a straight part in the earlier shell (Okamoto, 1988a, pl. 7, fig. 8). However, H. orientale can be distinguished from the newly described species by the three to four rows of tubercles present through its growth, with no development of a constriction. Eubostrychoceras perplexum sp. nov. is also similar to Yezoceras miotuberculatum Matsumoto (1977, p. 320, pl. 46, fig. 1) with respect to the following features: a rounded rectangular whorl section, slightly curved sigmoidal ribs, and tightly coiled torticonic shell, which is recognized as an intraspecific variation within Y. miotuberculatum. Nonetheless, Y. miotuberculatum can be distinguished from E. perplexum sp. nov. by the presence of two rows of tubercles and a siphuncle at the base of the whorls. The presence of three to four rows of tubercles and lack of constrictions in the genus Hyphantoceras Hyatt, 1900, and the presence of two rows of tubercles in the main part and the siphuncle positioned at the base of the whorls in the genus Yezoceras Matsumoto, 1977, are accordingly considered important characters that define each of these genera (Matsumoto, 1977; Wright et al., 1996). Consequently, even though E. perplexum sp. nov. has a number of characters commonly observed in species within the aforementioned genera, it would not be valid to assign E. perplexum sp. nov. to any these genera. In conclusion, on the basis of the current classification scheme, E. perplexum sp. nov. is assigned to the genus Eubostrychoceras.
Remarks.—Multiple modifications of ornamentation are observed in several specimens of Eubostrychoceras perplexum sp. nov. (Figures 8B–H, 9), among which are disturbed ribs, interrupted ribs, localized absence of ribs, longitudinal linear structures, distorted flare-like ribs, minute random projections, and unclear flat-topped elevations. Constriction, which is considered one of the diagnostic features of the new species, is excluded here. The most commonly identified of these features are the disturbed and interrupted ribs of differing sizes, which occur throughout growth (Figures 8B–D, 9). In the paratype MCM-A2026, disturbance begins with a small scar and gradually diminishes (Figures 4J–M, 8D). The paratype MCM-A1995 shows localized absence of ribs in addition to disturbances in the later part (Figures 5E–H, 8E). The paratype KYUM GKH 20025, has a longitudinal linear structure originating from a small scar (Figures 4M–O, 8F). It is possible that these features arose as a consequence of exogenous injury (= forma verticata, Hölder, 1956). The paratype MCM-M0299 exhibits overall disturbance in the later one whorl (Figures 5A–D, 8G), followed by distorted flare-like ribs, minute random projections, and a slightly deformed whorl (Figures 5A–D, 8H). Unfortunately, due to poor preservation and damage to the shell during preparation, these characters are difficult to examine in more detail. The holotype MCM-A2032 has an unclear flat-topped elevation on several ribs immediately after the constrictions in the later one whorl (Figure 8I). The single elevation is located at the center of each rib and positioned near the center, although given the poor state of preservation, the details cannot be observed. Modification of the ornamentation in the later stage of growth is a common feature in many ammonoids, including those in the family Nostoceratidae (e.g. Eubostrychoceras, Didymoceras, Muramotoceras, Nipponites, and Yezoceras: Yabe, 1904; Matsumoto, 1977; Okamoto, 1989; Shigeta et al., 2016; Aiba et al., 2021). Moreover, it has been established that the types of ornamentation occurring in the later stage of growth are generally stable within species (e.g. Nipponites mirabilis and Yezoceras spp.: Matsumoto, 1977; Okamoto, 1989; Aiba et al., 2021). Contrastingly, however, modifications in the ornamentation observed in some specimens of E. perplexum sp. nov. are accompanied by distortions in the whorl shape, randomness, and obvious abnormalities, with large intraspecific variation. Notably, the largest specimen WMNH-Ge-1140210361 lacks such modifications (Figures 7, 8J), thereby indicating that these modifications are unrelated to a cessation of growth or maturation. Consequently, changes in ornamentation of this species are not considered valid diagnostic features in later stages of growth.
Occurrence.—Three of the specimens described herein were retrieved from in situ calcareous concretions in the Uh Unit of the Haborogawa Formation, whereas the remainder were collected from float calcareous concretions or concretions lacking precise sampling information in the Tappu, Kotanbetsu, and Haboro areas. Although the exact horizons from which the concretions were derived are uncertain, it is almost certain that they originated from the Uh Unit, which is correlated with the uppermost part of the Santonian Inoceramus amakusensis Zone (Toshimitsu et al., 1995, 1998, 2007). This species is also known from the uppermost Santonian–lower Campanian in the Futakawa Formation, Sotoizumi Group in the Aridagawa area, Wakayama, southwestern Japan (Misaki, 2016).
The specimens of Eubostrychoceras perplexum sp. nov. from the Haborogawa Formation, Yezo Group in Hokkaido, northern Japan, are all considered to have originated from the Uh unit (the uppermost Santonian), with three of the 13 specimens retrieved in situ from outcrops. In addition, one specimen reported from the Futakawa Formation, Sotoizumi Group in Wakayama, southwestern Japan, was identified as a new species. The Futakawa Formation is estimated to be the uppermost Santonian–lower Campanian (Misaki, 2016), and locality T6, from which this specimen was found, could possibly be constrained to the uppermost Santonian by correlation with the Yezo Group. Thus, E. perplexum sp. nov., with its potentially limited stratigraphic occurrence, may have utility in determining stratigraphic correlations between distant areas.
The classification and estimation of phylogenetic relationships among genera within Nostoceratidae have traditionally been based on shell morphology and ornamentation, with particular emphasis on the latter in previous studies (e.g. Matsumoto, 1967, 1977; Wright et al., 1996; Klinger and Kennedy, 2003). This perspective is further supported by theoretical elucidation of the mechanisms underlying the regulation of shell formation in the families Nostoceratidae and Diplomoceratidae (Okamoto, 1984, 1988a, b, c, 1989). For example, on the basis of computer simulations, Okamoto (1988c) demonstrated that the three coiling patterns of heteromorph ammonoids—helicoid-type coiling (Eubostrychoceras), crioceratoid-type coiling (Scalarites), and meandering coiling (Nipponites)—are realized via interactive regulation between the orientation of the aperture and the growth direction of the whorls. Furthermore, it has been established that even a slight adjustment in the orientation of the aperture can lead to a change in the overall coiling pattern and that “saltation” can occur between helicoid-type coiling and meandering coiling, which is consistent with patterns observed in the actual fossil record (Okamoto, 1988c, 1989).
In this study, specimens of a formerly undescribed species were assigned to the genus Eubostrychoceras based on the classification scheme established in previous research. Within Eubostrychoceras, E. muramotoi Matsumoto (1967) and E. indopacificum Matsumoto (1967) are characterized by torticonic whorls and appear to be relatively similar to E. perplexum sp. nov. Eubostrychoceras muramotoi and E. perplexum sp. nov. are also similar with respect to having a straight part, although these species differ regarding their contrasting directions of growth. Both E. muramotoi and E. indopacificum are recorded from the upper Turonian to the middle Coniacian (Toshimitsu and Hirano, 2000), whereas the new species is known from the upper Santonian. This slight temporal gap might be due to the incompleteness of the fossil record. However, at present, it is difficult to infer a direct phylogenetic relationship between E. perplexum sp. nov. and these species. Furthermore, no other Eubostrychoceras species that are morphologically similar to E. perplexum sp. nov. and occur sympatrically are known.
Interestingly, however, several species within other genera have certain similarities to Eubostrychoceras perplexum sp. nov. with respect to characters other than ornamentation. In particular, Hyphantoceras orientale (Yabe, 1904) has multiple features in common, including a straight early part extending along the coiling axis, a small apical angle, a rounded rectangular whorl section, predominantly sinistral whorls, slightly curved sigmoidal ribs, and a siphuncle positioned slightly above the center of the whorls. Additionally, although H. orientale has an earlier first appearance in the geological record than E. perplexum sp. nov., it occurs sympatrically in the upper Santonian. On the other hand, these species differ with respect the coiling patterns, with the whorls in E. perplexum sp. nov. being in contact with each other, whereas they are distinctly separate in H. orientale. Notably, however, there are species in both genera Eubostrychoceras and Hyphantoceras that show intraspecific variation and ontogenetic changes in the degree of whorl separation (e.g. Hyphantoceras transitorium: Aiba, 2019; Eubostrychoceras elongatum: Misaki and Maeda, 2010). Given that an initial straight shell is observed in various lineages within Nostoceratidae (e.g. Tanabe et al., 1981; Okamoto, 1988a, b), the occurrence of this character alone is not necessarily indicative of a direct phylogenetic relationship. However, in the case of Hyphantoceras, at least two species that are likely to be directly related share an initial straight shell extending along the coiling axis (Okamoto, 1988a, pl. 7, fig. 8; Aiba, 2022, figs. 2, 3). Regarding ornamentation, E. perplexum sp. nov. lacks tubercles, whereas H. orientale typically has three or four rows of tubercles. However, specimens of H. orientale with underdeveloped tubercles have recently been discovered alongside typical specimens (Aiba and Karasawa, 2025), and, notably, in one of the two described specimens, there was an almost complete loss of tubercles (Aiba and Karasawa, 2025, fig. 2). Similarly, in numerous other ammonoids, intraspecific variations in the occurrence of tubercles have been well documented (e.g. Neogastroplites: Reeside and Cobban, 1960; Yokoyamaoceras: Maeda, 1993; Acanthoscaphites: Machalski, 2010). Within the family Nostoceratidae, Didymoceras Hyatt, 1894 has been reported to exhibit variations in the development of tubercles; however, although it has been suggested that the tubercles of Didymoceras are regular and those of Bostrychoceras Hyatt, 1900 are irregular (Błaszkiewicz, 1980), there are species assigned to Didymoceras that exhibit irregular tubercles or the loss of tubercles during growth (Kennedy and Cobban, 1997; Kennedy et al., 2000; Niebuhr and Jagt, 2016; Shigeta et al., 2016, 2019). On the basis of these findings, it has thus been proposed that Bostrychoceras might be a junior synonym of Didymoceras (Misaki and Tsujino, 2021). Contrastingly, it would appear that the position of the siphuncle is a relatively stable character on which to base classification. For example, although the three species of Yezoceras are characterized by different coiling patterns, the siphuncle is consistently located in the lower part of the whorls (Matsumoto, 1977; Aiba et al., 2021).
By comprehensively considering the aforementioned taxonomic characters, it is plausible to hypothesize a phylogenetic framework in which E. perplexum sp. nov. is derived from H. orientale. However, given that the divergence of Hyphantoceras and Eubostrychoceras (or their ancestral taxa) is estimated to have occurred prior to the Turonian age (Matsumoto, 1967; Tanabe et al., 1981), assuming a phylogenetic relationship between these two species in the Santonian would imply that Eubostrychoceras is paraphyletic, which contradicts the current systematics. This disparity between taxonomic evaluation and phylogenetic estimation might indicate that there are certain issues with the current classification scheme for the genera within Nostoceratidae. In the current scheme, ornamentation is treated as an absolute criterion; however, the presence and number of tubercles, the coiling patterns, and other characteristics may be equally important diagnostic features, and it might be necessary to estimate evolutionary series based on comprehensive comparisons. More specifically, for instance, the aforementioned contradiction could potentially be resolved by expanding the definition of the genus Hyphantoceras to include species without rows of tubercles. However, this would require a careful re-examination of all species currently assigned to Hyphantoceras. Such a revision, which would fundamentally overturn the traditional definition of Hyphantoceras that has long considered the presence of rows of tubercles as a primary diagnostic feature, would also necessitate a reevaluation of the distinctions between Hyphantoceras and other genera within the Nostoceratidae. In any case, the specimens described in this study alone are insufficient for such an evaluation. A more complete understanding of the taxonomy and phylogeny of the Nostoceratidae will require continuous and careful stratigraphic investigations, along with a thorough comparison of the shell morphologies of individual species.
I thank Harunobu Yamato and Masatoshi Iwata (Volunteer’s Association of Mikasa City Museum), Koichi Fukuoka (Ishikari, Hokkaido), Akio Tomita (Sapporo, Hokkaido), and Yusuke Muramiya (Fukada Geological Institute) for donating specimens and providing information regarding localities; the late Hitoshi Furusawa (Sapporo Museum Activity Center), Haruyoshi Maeda and Yasuhiro Ito (both, the Kyushu University Museum), and Masaaki Ohara (Wakayama Prefectural Museum of Natural History) for kindly permitting examination of the museum collection; Kenji Ikuno (The Museum of Nature and Human Activities, Hyogo) for valuable discussions; Yasunari Shigeta (National Museum of Nature and Science), Akihiro Misaki (Kitakyushu Museum of Natural History and Human History), Ryoji Wani (Yokohama National University), and two anonymous reviewers for their valuable comments on improving the manuscript. I also thank the late Minoru Yamashita and late Hiroshi Hayakawa, who collected the KYUM GKH and SMAC specimens, respectively. Thanks are also extended to Fumiko Murakami and the late Mamoru Murakami (Murakami Lodging House, Haboro Town), and the Haboro, Kotanbetsu, and Tappu forestry offices.
