Mya arenaria, AND M. JAPONICA

Zhang, Junlong, Yurchenko, Olga V, Lutaenko, Konstantin A, Kalachev, Alexander V, Nekhaev, Ivan O, Aguilar, Robert, Zhan, Zifeng & Ogburn, Matthew B, 2018, A tale of two soft-shell clams: an integrative taxonomic analysis confirms Mya japonica as a valid species distinct from Mya arenaria (Bivalvia: Myidae), Zoological Journal of the Linnean Society 184 (3), pp. 605-622 : 617-619

DISTRIBUTION OF M. ARENARIA AND M. JAPONICA

The soft-shell clam M. arenaria is widely accepted to have a circumboreal distribution ( Bernard, 1979, 1983; Strasser, 1999; Coan et al., 2000; Huber, 2010; Zhang et al., 2012). Previous molecular studies indicated a complex colonization history between Europe and North America, but these analyses lacked samples from the northwest Pacific – the putative range of M. japonica ( Strasser & Barber, 2009; Layton et al., 2014; Barco et al., 2016; Cross et al., 2016; Lasota et al., 2016). Our genetic and spermatozoan ultramorphological analyses strongly suggest that soft-shell clams in the northwest Pacific represent M. japonica . Two publicly available sequences identified as ‘ M. arenaria ’ (the valid species name at the time of identification) from China (QWEAS092-15 + QWEAS093-15) were highly similar (<2% difference COI; Supporting Information, Table S2) to western Pacific M. japonica . These individuals are probably M. japonica and do not reflect the presence of M. arenaria in the western Pacific. MacNeil (1965) reported that M. arenaria occurred in Japan only in the middle Miocene and that later and recent specimens from the Okhotsk Sea, west coast of Japan and China are M. japonica , which was supported by Lutaenko & Noseworthy (2012) and Scarlato (1981). Numerous other authors have also reported M. japonica from China, Japan, Korea and the Russian Far Eastern seas ( Jay, 1856; Makiyama, 1935; Habe, 1955, 1977; Tchang, Qi & Li, 1955; Fujie, 1957; Scarlato, 1981; Okutani, 2000). While both M. arenaria and M. japonica were present along western North America at one time, there is no palaeontological or archaeological evidence that either species survived the Pleistocene glaciation, with the possible exception of relict populations in the Bering Sea ( Carlton, 1979). Mya arenaria was first introduced into the northeastern Pacific via Eastern Oyster Crassostrea virginica Gmelin, 1791 seed plantings in San Francisco Bay, California, during the 1860–1870s ( Carlton, 1979). Since that time, it has spread northward, by either intentional plantings or natural dispersal, to southern Alaska ( Carlton, 1979; Strasser, 1999; Powers et al., 2006). In contrast, until the present study, there were no recent eastern Pacific records of M. japonica or any evidence of intentional or accidental introductions of western Pacific Mya into the eastern Pacific. Two publicly available sequences identified as ‘ M. arenaria ’ from the British Columbia, one collected from Quascilla Bay (DFO097-11) and the other from Graham Island (RBCMI219-14), were highly similar (<2% difference COI; Supporting Information, Table S2) to all western Pacific M. japonica . These individuals, both collected in June 2011, are probably M. japonica and went unnoticed because of morphological similarities to M. arenaria , the expected species in British Columbia. Thus, these records represent the first modern documented occurrence of M. japonica in the northeast Pacific. In addition, after examining photographs of these two clams, the authors are confident in classifying them as M. japonica based on the morphological criteria used in this study.

We are uncertain of the source of introduction or status of possible eastern Pacific M. japonica populations. However, similar to M. arenaria , M. japonica was probably introduced via oyster plantings, but in this case, the Pacific Oyster Crassostrea gigas Thunberg, 1793 . These plantings, which occurred from the early 1910s until after the Second World War ( Carlton, 1979; Lavoie, 2005), are also thought to have led to other bivalve introductions, such as the Japanese Littleneck Ruditapes philippinarum Adams & Reeve, 1850 and Quadrate Trapezium Neotrapezium liratum Reeve, 1843 . Alternatively, M. japonica may have been introduced by ballast water, as is speculated with the Asian Semele Theora lubrica Gould, 1861 and Asian brackish-water clam P. amurensis in California ( Fofonoff et al., 2003). In either case, M. japonica was probably introduced decades after M. arenaria became widespread and well established in the eastern Pacific ( Carlton, 1979), which reduced the likelihood of detection of M. japonica in areas where M. arenaria was already present or in nearby localities. Furthermore, both individuals were larger-sized adults collected nearly 400 km apart, which suggests there may be viable populations of M. japonica in at least British Columbia or possibly other Pacific Northeast locations. Albeit unlikely, the presence of individuals with high COI similarly to M. japonica could represent F1 hybrids with maternal M. japonica contribution or possibly reflect historical or cryptic hybridization ( Pfenninger, Reinhardt & Streit, 2002; Rees, Dioli & Kirkendall, 2003) between M. arenaria and M. japonica . While there were significant differences in sperm morphology between M. arenaria and M. japonica , we are uncertain if these differences (or ecological/behavioural differences) would prohibit hybridization.

According to Laursen (1966), historical reports of M. arenaria from the Arctic Ocean are erroneous and should represent M.truncata ovata Jensen, 1900 , which is currently recognized as a junior synonym of M. pseudoarenaria (see Huber, 2010). However, we identified M. arenaria from the Russian waters of the Barents Sea, a marginal sea of the Arctic Ocean, and it has also been reported further west near Forsøl, Norway ( Crocetta & Turolla, 2011). A rigorous genetic analysis (incorporating both nuclear and mitochondrial markers) of M. arenaria and M. japonica collected from across the eastern and western Pacific, Bering Sea, Arctic Ocean and greater Atlantic is required to fully resolve the zoogeography of both species, determine the extent of M. japonica introductions in the eastern Pacific, identify any possible cryptic introductions of M. arenaria in the western Pacific, determine the identity of soft-shell clams in the Bering Sea and examine possible hybridization of M. arenaria and M. japonica .

DIVERGENCE TIME AND EVOLUTIONARY / MIGRATION HISTORY

We estimate that M. japonica diverged from M. arenaria 4.1–12.5 Myr; this agrees with the evolution of M. arenaria , which evolved during early Pliocene to late Miocene from its ancestor Mya fujiei MacNeil, 1965 , which arose during middle Miocene ( MacNeil, 1965; Strauch, 1972). Mya arenaria first appeared in the fossil record from Japan in late Miocene formations and almost at the same time from the eastern Pacific in California ( Strauch, 1972). Mya japonica was also probably present in the eastern Pacific at least during the Pleistocene ( MacNeil, 1965; Carlton, 1979). Although the formation of the Bering Strait during the Pliocene overlapped with the divergence of M. arenaria and M. japonica , it is likely that only M. arenaria crossed the Arctic to reach the western and eastern Atlantic ( MacNeil, 1965; Bernard, 1979; Vermeij, 1989). Strauch (1972) suggested M. arenaria may also have migrated through the Central American Passage from the Pacific to the Atlantic, where it then spread to Europe in the late Pliocene ( Strasser, 1999), but fossil evidence for this migratory route is lacking ( Bernard, 1979). During the Pleistocene glaciation, M. arenaria was extirpated from all areas except the western Atlantic and M. japonica was extirpated from all areas except the western Pacific ( Strasser, 1999). However, the status of fossil, archaeological and recent soft-shell clams in the Bering Sea requires further investigation, which could represent relict populations of M. arenaria or M. japonica , adventitious individuals of M. japonica from the western Pacific or an undescribed cryptic species ( MacNeil, 1965; Bernard, 1979; Carlton, 1979). The evolutionary history of Mya is complex and there is much uncertainty regarding the identification and evolutionary relationships of fossil material ( MacNeil, 1965) and their connection with extant forms, particularly as it relates to the synonymy of extant species and species identified solely from fossil material ( Petersen, 1999). Due to the morphological similarities between M. arenaria and M. japonica and levels of phenotypic plasticity, it may be difficult to fully resolve their evolutionary and migration history.

MacNeil (1965) and Strauch (1972) reported that the oldest records of M. truncata and Mya cuneiformis Böhm 1915 were from the middle Miocene, both of which were direct descendants of the late Oligocene or early Miocene species, Mya salmonensis Clark, 1932 , while M. pseudoarenaria and Mya priapus Tilesius, 1822 derived from M. cuneiformis in the late Miocene. Nakashima (1999) determined that the oldest records of M. pseudoarenaria , M. cuneiformis and M. truncata were from the early Miocene and M. uzenensis may have directly evolved from M. salmonensis . Our results indicate that the separation time of M. uzenensis and M. truncata (3.9–21.0 Myr) is nearly the same as M. japonica and M. arenaria and that M. uzenensis and M. truncata share a common ancestor, but it is unclear if M. uzenensis directly evolved from M. salmonensis or M. cuneiformis .