Evolution of Janthina and Recluzia (Mollusca: Gastropoda: Epitoniidae) Author Beu, Alan G. text Records of the Australian Museum 2017 Rec. Aust. Mus. 2017-08-23 69 3 119 222 http://dx.doi.org/10.3853/j.2201-4349.69.2017.1666 journal article 228171 10.3853/j.2201-4349.69.2017.1666 d63960a1-e0c5-4b97-8e7d-f8be80d376d6 2201-4349 4677010 08B086EB-8D24-4FD0-975A-E045E2596BF1 Janthina typica ( Bronn, 1861 ) Figs 24–25 Hartungia typica Bronn, 1861: 119 , pl. 19, figs 3a–d; Bronn, 1862: 32 (in part); Fleming, 1953a: 135 ; Veiga Ferreira, 1955: 6 , 8; Beu & Maxwell, 1990: 292, 411, pl. 37a–b; Tomida, 1996 : pl. 33, figs 3a–4; Maxwell in Spencer et al ., 2010: 245. Janthina hartungi Mayer, 1864a: 242 , pl. 6, figs 41a–c; 1864b: 62, pl. 6, figs 41a–c; Berkeley Cotter, 1892: 285 ; Joksimowitsch, 1911: 80 , 94; Gagel, 1911: 409 ; Berkeley Cotter, 1953: 100 ; Veiga Ferreira, 1955: 9–10 . Heligmope dennanti Tate, 1893: 329 , pl. 7, figs 5–5a; Finlay, 1931: 1 . Turbo postulatus Bartrum, 1919: 100 , pl. 7, fig. 14. Acrybia (Heligmope) dennanti (Tate) .– Cossmann, 1925: 161 , pl. 4, figs 11–12; pl. 9, fig. 3 (copy of Tate, 1893 : pl. 7, fig. 5). “ Turbo ” postulatus Bartrum. – Bartrum & Powell, 1928: 141, pl. 25, figs 6–7; Marwick, 1931: 28 , 43. Heligmope postulatus (Bartrum) . – Finlay, 1931: 1 ; Laws, 1940b: 38 ; Marwick, 1948: 6 , 8; Law, 1950: 7, faunal list. Janthina (Hartungia) typica (Bronn) . – Wenz, 1940: 815; Tomida et al ., 2013: 60 , figs 3A–D only (not figs 3E–L). Bulbus (Heligmope) denanti [sic] (Tate).–Wenz, 1941: 1036, fig. 2967. Acrybia (Hartungia) chouberti Chavan, 1951: 135 , fig. 1; Lecointre, 1952: 114 ; Choubert, 1965: 49 ; Brébion, 1973: 50 . Hartungia chouberti (Chavan) .– Fleming, 1953a: 135 ; Ludbrook, 1978: 122 , pl. 12, figs 17–19. Hartungia dennanti (Tate) . – Fleming, 1953a: 135 ; Darragh, 1970: 166 ; Ludbrook, 1973: 256 , pl. 28, figs 93–94; Darragh, 1985: 106 . Hartungia postulata (Bartrum) . – Fleming, 1953a: 135 ; Schofield, 1958: 252 ; Marwick, 1965: 10 , table 4; Fleming, 1966: 49 . Janthina typica (Bronn) . – Krejci-Graf et al ., 1958: 336 (in part); Zbyszewski et al ., 1961: 15 ; Zbyszewski & Veiga Ferreira, 1962: 219, 222, 224, 227–228, 273; Beu & Raine, 2009: BM292; Meco et al ., 2015: 61 , figs Appendix A–B. Hartungia sp.– Wilkins, 1963: 58 ; Tomida & Itoigawa, 1984: 112; Tomida, 1989: 96; Ozawa & Tomida, 1992: 428. Hartungia dennanti dennanti (Tate) . – Ludbrook, 1978: 122 , pl. 12, figs 15–16. Parajanthina japonica Tomida & Itoigawa, 1982: 60 , pl. 19, figs 1a–c (in part only). Parajanthina sp.–Tomida & Itoigawa, 1982: 62, pl. 19, figs 2–3. Hartungia sp. A.– Nakamura et al ., 1999 : pl. 2, figs 17a–b. Hartungia elegans Tomida & Nakamura, 2001: 217 , figs 2.1a–e, 2.2a–e. Eunaticina abyssalis Simone, 2014: 586 , figs 10E–K. Type material . Hartungia typica , location of any original type material unknown; extensive enquiries over more than 40 years have brought none to light. Bronn’s introduction to his descriptions of the fossils of Santa Maria Island in Hartung (1861: 116) reads: “In the spring of 1858 I received from Hartung a collection of hand specimens of Tertiary limestone with fossil shells for investigation and determination, which he had collected on Santa Maria. A year later I received another similar consignment, which Mr Drouet from Toyes had gathered at the same time as Mr Hartung, but which contained no other species than the first. I give here the results of my research” (translated from German by T. A. Darragh, Museum Victoria , pers. comm. 10 Nov 2015 ). Bronn had only one specimen of Hartungia typica ( Bronn, 1861: 129 , table 3) so the “contained no other species” statement makes it certain that the holotype belonged to the German geologist Hartung rather than to Drouet. Bronn’s fossil collection was originally in the University of Heidelberg, but was purchased by Louis Agassiz ( Cleevely, 1983: 68 ) and is now in Paleontology, Museum of Comparative Zoology, Harvard University. However, it does not include any Azores material (K. Boss, Museum of Comparative Zoology, pers. comm. 18 May 1983 ). Possibly the Azores material described by Bronn was returned to the collectors, Hartung and Drouet. In view of the acquisition by Mayer of the material in Heidelberg described by Bronn (see below under Remarks ) it is also possible that all Bronn’s Azores material was sent to Mayer and never returned, but again, it is not present in MayerEymar’s collection in NMB . The location of any Azores material formerly belonging to Hartung also is unknown. Janthina hartungi , again no original material known. Mayer-Eymar’s collection (in NMB ) includes only NMB Po.6227, two poor, small (shell fragment c. 11.5 mm wide) impressions in modelling clay from the same cavity in the rock at Santa Maria Island, identified by Mayer-Eymar as Janthina hartungi , and labelled “Ponta dos Matos, S. Maria”. This impression of a partial spire bears weak axial ridges and vague, low spiral folds ( Fig. 24M ), indicating that it is indeed a very poor partial mould of a Janthina species similar to J. typica , but Mayer’s (1864a, b: pl. 6, fig. 41) illustrations of a complete shell ( Figs 25D–F ) cannot have been based on it. Mayer (1864a, b: 63) recorded a specimen from “Ponta dos Mattos” (sic), but the modelling clay impression is not identifiable to species and is not considered to be type material. Again, the location of the specimen Mayer’s illustration was based on is unknown. Therefore, a neotype is required for both Hartungia typica and Janthina hartungi , unequivocally to establish the application of these names to the present species. To find any genuine well-preserved Santa Maria Island specimens of Janthina typica in modern collections or to re-collect specimens at Santa Maria Island proved very difficult. The writer visited Santa Maria Island with Bernard Landau in February 1998 , but localities near sea-level in the Touril Complex were all inaccessible (during winter) through either severe wave action or their location at the foot of inaccessible cliffs. None was found in the overgrown remnants of the limestone quarry at Figueiral, although pectinids were common. The one well-preserved specimen from Santa Maria Island referable to Janthina typica that the writer is aware of is a very small one, MIGM 1312. This specimen ( Figs 24A–C ) is here designated the neotype of both Hartungia typica Bronn, 1861 and Janthina hartungi Mayer, 1864 . It is from “Farolim da Ponta do Norte” (North Point Lighthouse), Santa Maria Island. Ponta do Norte is now clearly understood to be in early Zanclean Touril Complex, following the reinterpretation by Sibrant et al. (2015) . Unfortunately, the apertural side of the neotype is poorly preserved, the specimen evidently having been attached to the outcrop by this side, but the specimen definitely has spiral folds and fine axial ridges over the entire teleoconch, and has a low spire as in all other small specimens of J. typica . Figure 24. Specimens of Janthina typica (Bronn) , all except A–C enlarged uniformly. (A–C) neotype designated here of Hartungia typica Bronn, 1861 and Janthina hartungi Mayer, 1864 , MIGM 1312, Farolim da Ponta do Norte (North Point Lighthouse), Santa Maria I., Azores, Zanclean (photos by C. Marques da Silva). (D, E) neotype designated here of Turbo postulatus Bartrum, AUGD G 5721, Kaawa Creek, SW Auckland , New Zealand , Opoitian (Zanclean). (F–I) 2 specimens , Kaawa Creek, SW Auckland , Opoitian (Zanclean); F, I, GNS TM4670; G–H, GNS TM4671. (J, K , O) holotype of Acrybia (Hartungia) chouberti Chavan, Service Géologique du Maroc , Rabat, Morocco , P7064; unwhitened photos from Rabat sent by N. H. Ludbrook (see also Ludbrook 1978: 122 , pl. 12, figs 17–19, at half this scale); Fouarat, E of Casablanca, Morocco , Zanclean. (L) abraded specimen, DBUA-F-428, University of the Azores, Ponta Delgado, Azores; from Ponta do Castelo, SE tip of Santa Maria I., Azores, Zanclean, coll. S. Ávila, 15 Sep 2006 (photo by S. Ávila). (M) NMB Po.6227, ex Mayer-Eymar collection; possible type material of Janthina hartungi Mayer , modelling clay impression from natural cast in rock, Ponta dos Matos, Santa Maria I., Azores, Zanclean. (N) GNS GS12289, E branch Greeks Creek,Arahura Valley, Westland, New Zealand , Waipipian (early Piacenzian). Scale bars 10 mm ; upper bar applies to Figs 24A–C, lower bar applies to all other figures. Figure 25.Specimens of Janthina typica (Bronn) . (A–C) GNS WM7656,Grange Burn Fm,Kalimnan (Zanclean),Muddy Creek,near Hamilton, Victoria , Australia ; i.e., from type locality of Heligmope dennanti Tate. (D–F) copy of Mayer’s (1864 : pl. 6, figs 41a–c) illustrations of Janthina hartungi, Santa Maria I. , Azores, Zanclean; slightly enlarged from published size, H 19, D 22 mm (photos by D. L. Homer). (G , I–K , M) weakly sculptured specimen, MNHN IM.2000-27158, holotype of Eunaticina abyssalis Simone, Marion Dufresne cruise MD55 stn 45-CB79, 1500–1575 m , off Itaúnas, Espiritu Santo State, Brazil , SW Atlantic, probably Zanclean;G, SEM, apical 2.5 teleoconch whorls, showing outer calcitic layer removed; I–K, M, apertural,ventral,apical and posterior views. (H) GNS GS12164, fragment, Whenuataru Tuff (Waipipian, early Piacenzian), Tarawhenua Peninsula, Pitt I., Chatham Islands; oblique latero-dorsal view. (L) SAMA T1494A, lectotype designated here of Heligmope dennanti Tate,Grange Burn Fm, Kalimnan (Zanclean) ,Muddy Creek,near Hamilton, Victoria , Australia (photo by D. L. Homer). (N , Q, R , T) ULPGC LE20151–2, Universidad de Las Palmas de Gran Canaria, Canary Islands; largest two of six specimens from La Esfinge, Gran Canaria, Canary Islands, Zanclean; N, R, LE20152; Q, T, LE20151, broken outer lip partially retained in sediment. (O, P , S) copies of Bronn’s (1861 : pl. 19, figs 3a–d) illustrations of the holotype of Hartungia typica ; Santa Maria I., Azores, Zanclean; at published size, H 18, D 22 mm (photos by D. L. Homer). Scale bars:Fig.25G = 500 µm ; Figs 25D–F and 25O–P, S are copies of published figures, approx.natural size; lower scale bar, 10 mm ,applies to all other figures. Heligmope dennanti , five syntypes in SAMA Tate type collection; SAMA T1494A–B, two syntypes , “Miocene”, Muddy Creek, near Hamilton, western Victoria (Grange Burn Formation, Kalimnan Australian Stage, Zanclean, early Pliocene; Beu & Darragh, 2001: fig. 6); SAMA T1515A–C, three syntypes , Hallett Cove Sandstone (Piacenzian), labelled “Miocene, Hallett’s Cove, St Vincent Gulf”, coast south of Adelaide, SouthAustralia. The specimens from Grange Burn Formation (T1494A–B) are both well-preserved specimens of Janthina typica . In contrast, one of those from Hallett Cove Sandstone (T1515B) is a small, incomplete shell probably assignable to J. typica , with a low spire and obvious spiral folds all over, whereas the other two (T1515A, C) are more complete specimens with no spiral folds on the sutural ramp. They are identified as J. chavani , despite Ludbrook’s (1978: 122) misgivings about their identity. As noted above under “Biostratigraphy”, the specimen of J. typica presumably is from the lower, early Piacenzian part of the formation, whereas the specimens of J. chavani presumably are from the overlying late Piacenzian–Gelasian part of the formation. Tate’s two Muddy Creek syntypes were illustrated by Ludbrook (1973 : pl. 28, figs 93–94) in an unusual oblique dorso-lateral view necessitated by their still being glued to Tate’s tablet. A lectotype designation is necessary because two species are present among Tate’s syntypes . Tate’s figured syntype , SAMA T1494A ( Fig. 25L ) from Grange Burn Formation at Muddy Creek, Victoria , is much the larger of the Muddy Creek syntypes and is here designated the lectotype of Heligmope dennanti . The paralectotype T1494B is a juvenile specimen. The lectotype has the spire tip missing, but both Muddy Creek specimens have obvious, prominent spiral folds over the entire teleoconch and are conspecific with the neotype of Hartungia typica . Turbo postulatus , holotype basal fragment not found in AUGD (N. Hudson, AUGD pers. comm. 24 Sep 2012 ). The specimen selected by Bartrum & Powell (1928: 141, pl. 25, figs 6–7) as the “ neotype ”, AUGD G5721 ( Figs 24D–E ) is here again designated the neotype of Turbo postulatus , although it had no status as a neotype in 1928, as the original type material was still available ( ICZN Article 75.1). Both specimens are from Kaawa Creek, coast south of Waikato Heads, southwest Auckland , North Island, New Zealand (Opoitian New Zealand Stage, Zanclean, early Pliocene; Cooper, 2004 : fig. 13.1). Material of Janthina typica from this site is rather fragile and tends to disintegrate through the calcite outer layer flaking off the aragonite inner layer, and the neotype has been reassembled recently by N. Hudson ( AUGD ), so presumably the holotype fragment disintegrated many years ago. A neotype is required to establish that the name applies to J. typica rather than supplanting J. chavani , which occurs at several other New Zealand localities. The neotype is an unusual, highly inflated, subspherical specimen with a short spire, relatively weak spiral folds, a strongly and evenly inflated last whorl, and a particularly prominent major fold generated by the sinus in the outer lip, but agrees with other material of J. typica in having spiral folds over the entire surface. Other specimens from Kaawa Creek (e.g., Figs 24F–I ) are closely similar to specimens from Grange Burn Formation at Muddy Creek, Victoria ( Figs 25A–C, L ) and the neotype of Hartungia typica in shape and sculpture, and there is no doubt that the Kaawa Creek population falls within the variation of Janthina typica . Acrybia (Hartungia) chouberti , holotype in G. Lecointre collection, Service Géologique du Maroc , Rabat, Morocco , P7064 ( Ludbrook, 1978: 122 , pl. 12, figs 17–19; Figs 24J–K, O ), not seen; plaster casts in Paléontologie collection, MNHN , and GNS WM7327. The type locality in Morocco was described by Chavan (1951) as “Aïn Sebaa, cuttings from well 10 … in sandstone, with Semicassis cf. laevigata (Defr.), Gryphaea forskali (Chemn.) , Balanus perforatus Brug. Holotype …; two fragments; one internal mould” (translation from Chavan, 1951: 136 ). However, the cast in GNS ( WM 7327) bears the locality label “Dar bel Hamri”. The locality was described by Chavan (1951: 135) as “from the series of l’Oued Fouarat”, and is known in most works on Moroccan Plio-Pleistocene stratigraphy (e.g., Arambourg, 1969 ) as Fouarat. Fouarat was located on the map by Lecointre (1963: 22) a few kilometres east of Casablanca. The holotype was said by Chavan (1951: 135) to be “a little taller than wide”, but his illustration of the holotype ( Chavan, 1951 : fig. 1) shows a specimen that is slightly wider than it is tall, although with the spire apex missing. He stated the dimensions as “height: 32 to 34 mm ; width: 30 mm ” ( Chavan, 1951: 136 ), but his drawing, stated to be enlarged x 1.75, provides dimensions of H 30, D 32 mm , so possibly Chavan accidentally reversed the dimensions. The unwhitened photographs sent to Ludbrook (1978 : pl. 12, figs 17–19) from Rabat, Morocco , were sent in turn to the writer by N. Ludbrook ( Figs 24J–K, O ) and show again that it is a specimen of Janthina typica with unusually numerous, prominent, narrow spiral folds on the sutural ramp, similar to those of the lectotype of Heligmope dennanti ( Fig. 25L ). Parajanthina japonica is included below under Janthina chavani ( Ludbrook, 1978 ) . Hartungia elegans , holotype in Department of Earth Sciences, Nagoya University, ESN2687, from Tano Formation (late Miocene, late Tortonian–early Messinian, planktonic foraminiferal zone N17) at Tano, Miyazaki Prefecture , near the east coast of Kyushu, Japan (Tomida & Nakamura, 2001: 217, fig. 1a); one paratype MFM 111029, from Senhata Formation, Miura Group (late Miocene, also zone N17) at Motona, Chiba Prefecture , east side of Tokyo Bay, Boso Peninsula, Honshu, Japan (Tomida & Nakamura, 2001: 217–218, fig. 1b); not seen. These specimens have obvious spiral folds all over and are interpreted here as severely dorsoventrally to obliquely compressed internal moulds of Janthina typica ; several similar specimens have been collected in New Zealand . Eunaticina abyssalis , holotype MNHN IM.2000-27158 (only known specimen; Figs 25G, I–K, M ) from Marion Dufresne cruise MD55 ( BRESIL ) station 45-CB79, 19°01'59.9916"S 37°47'59.9964"W , 1500–1575 m , off Itaúnas, Espiritu Santo State, Brazil, SW Atlantic, collected by P. Bouchet, B. Métivier and J. Leal, 25 May 1987 . The holotype was said by Simone (2014: 587) to be in “compact blocks (no living specimen)”, and is a Pliocene fossil specimen of Janthina typica enclosed in weakly lithified cream limestone (“globigerina ooze”) dredged from the sea floor. Examination of the holotype , loaned by P. Bouchet and V. Héros ( MNHN 16 Sep 2015 ) left no doubt that this specimen is a weakly sculptured specimen of J. typica with a low spire and low spiral folds and fine, closely spaced axial ridges all over. It has now been cleaned further and whitened before photography, revealing its sculpture more clearly. The smooth, polished “protoconch” illustrated by Simone (2014 : figs 10I–J) is actually the apical spire whorls of the Janthina teleoconch ( Fig. 25G ); the protoconch is missing. The narrow spiral gap between whorls on this smooth apex demonstrates that the 25 µm-thick calcitic outer layer has been dissolved from the first c. 2.5 teleoconch whorls, revealing the smooth, lightly polished aragonite infilling of the original calcitic shell and leaving a polished surface lower than the outer surface of the rest of the teleoconch. Weak impressions of the spiral and, to a lesser extent, axial ridges on the exterior of the calcitic teleoconch remain on the aragonitic shell. Matrix from within the aperture of the holotype was examined for calcareous nannofossils by Denise Kulhanek (International Ocean Drilling Program Office, College Station, Texas) and Claire Shepherd ( GNS ; pers. comm. 15 Oct 2015 ). They reported that nannofossils are difficult to interpret because of overgrowth or dissolution of many specimens; the remobilized carbonate presumably caused the weak lithification of the sample. Reticulofenestra is dominant and small Gephyrocapsa specimens are present, but not large ones, indicating an age greater than 1.73 Ma. This is consistent with the holotype being a Pliocene fossil rather than a present-day specimen. Other material examined . Santa Maria Island : Touril Complex (Zanclean), Ponta do Castelo, SE tip of island, Zanclean (DBUA-F-428, University of the Azores , Ponta Delgado, Azores Islands, Portugal; 1 specimen , collected by S. Ávila, 15 Sep 2006 ; photographs sent by S. Ávila, 02 Oct 2012 ; Fig. 24L ). This specimen is abraded and incomplete, but nevertheless is a relatively large specimen confirming the occurrence of Janthina typica at more than one locality on Santa Maria Island. Gran Canaria Island: La Esfinge, a short distance north of Las Palmas de Gran Canaria, east side of La Isleta, NE Gran Canaria, Canary Islands , material reported by Meco et al . (2015 , 2016 ) ( ULPGC LE20151–LE20156, 6 specimens loaned by Joaquín Meco, ULPGC ; Figs 25N, Q–R, T ). As noted above, nearby 40 Ar/ 39 Ar dates on an underlying lava flow provide a maximum age for the Janthina specimens of 4.20±0.18 Ma, late in Zanclean time ( Meco et al ., 2015 ). About 120 specimens from localities in Australia and New Zealand: Australia : Victoria : Grange Burn Formation (Kalimnan, Zanclean), Muddy Creek , near Hamilton, W Victoria ( NMV P26906 , 1 ; P26908–9 , 2 ; P40662 , 1 ; P40663 , P40667 , 3 ; P316445 , 1 ; P316449 , 1 ; P406640 , P406645 , 5 ; GNS WM 7656 , 1 ); Lower Jemmys Point Shellbed (Kalimnan, Zanclean), road cutting SW side of Bunga Creek , Princes Highway, near Lakes Entrance , E Victoria ( Wilkins, 1963: 58 ; NMV P 22612 , Wilkins’s spec., 1 fragment; P26905 , 1 ; P40666 , 1 ). South Australia : Hallett Cove Sandstone, Hallett Cove , coast south of Adelaide , one of Tate’s syntypes ( SAMA T1515C , with spiral folds all over, here early Piacenzian) . New Zealand : Kapitean ( Messinian ): East Cape : first cutting E of Awatere River mouth, East Cape ( NMNZ M043211 , 1 incomplete); 250 m NE of lighthouse, East Cape ( AUGD 10194 , Z14/f0127, grid ref. Z14/994775; 1 ). Gisborne district : Waimata Rd , 16 km W of Tolaga Bay ( GS1357 , Y17/f7477, grid ref. Y17/604021; 2 poor moulds, severely compressed). Opoitian (Zanclean): SW Auckland : Kaawa Creek , coast S of Waikato Heads ( GS996 , GS5513 , R12/f8518, 11 ; R13/f7020, R13/f7022, R13/ f7027, R13/f7033, R13/f7051, R13/f7056, R13/f7057, R13/f7059, R14/ f7062, grid ref.R13/646085, GNS and mainly in AUGD , many; Figs 24F–I ; OUGD unnumbered, pres. C. R. Laws, 1 ) . Gisborne district : Waihora Valley , Te Karaka , Gisborne ( Marwick, 1931: 43 ) ( GS870 , Y17/f7462, grid ref.Y17/364961; 1 poor mould) . Hawke’s Bay : Mangawhero Stream ( GS1543 , W19/f7462, grid ref. W19/831408; 6 ); Waikaremoana Rd , road cut 30 km from Wairoa ( GS1544 , W19/f7566, grid ref. W19/810464; 1 ); Waikaretaheke River 1 km S of junction with Waiau River ( GS1555 , W19/ f7472, grid ref. W19/793441; 2 ); Waikaremoana Rd , road cut 27 km from Wairoa ( GS1556 , W19/f7473, grid ref. W19/709446; 1 ); Cricklewood Rd , 0‒1200 m E of Waiau Rd ( GS1557 , W19/f7474, grid ref. W19/661402; 1 ); beneath white tephra, Waiau River , 200 m from Mangaone Stream ( GS1561 , W19/f7477, grid ref. W19/665418; 3) ; junction Waiau River and Pakihiwi Stream ( GS1567 , W19/f7516, grid ref. W19/708424; 1 ); Cricklewood Rd , 2.4 km E of Mohaka –Putere Rd ( GS1580 , W19/f7486, grid ref. W19/618401; 1 ); Waihi Stream 400–800 m upstream from Waihi homestead (possibly Kapitean ; GS2063,W19/f7492,grid ref.W19/694499; 1 ); Hangaroa–Tiniroto Rd 5 km SSW of Hangaroa ( GS 2852 , X18/f7488, grid ref. X18 /099650; 1 ); Parikanapa Rd ( GS8030 , X18 /f9624, grid ref. X18 /104593; 3 ); Opoiti Limestone , Mangapiopio Stream ( GS8041 , X18 / f9627, grid ref. X18 /034645; 1 ); Putere Rd ( GS8154 , W19/f7617, grid ref. W19/615403; 1 ); Mohaka River below Willow Flat bridge ( GS8182 , W19/f8573, grid ref. W19/520370; 1 poor mould); road SW of Willow Flat ( GS8213 , W19/f8582, grid ref. W19/506352; 1 ); 1 km E of Ardkeen hall, Frasertown – Waikaremoana Rd ( GS8273 , W19/f7675, grid ref. W19/812434; 1 ); Ruakituri River road ( GS11313 , X18 /f081, grid ref. X18 /951592; 1 ); Hangaroa Bluffs , Tiniroto – Gisborne Rd ( GS11466 , X18 / f7553, grid ref. X18 /100659; 1 fragment); siltstone between Kidnappers shelter hut and Cape Kidnappers ( GS10855 , W21/f8584, grid ref. W21/607657; 16 poor); bay S side of Cape Kidnappers ( GS10856 , W21/ f8585, grid ref. W21/611648; 2 poor). Westland : Kapitea Creek , base of Opoitian section ( GS12487 , J32/f9774, grid ref. J32/562409; 1 poor); road cut, Greenstone–Kumara Rd ( GS3046 , J32/f9146, grid ref. J32/634410; 1 ; GS11557 , J32/f9809, grid ref. J32/634410; c. 15 small specimens in one block); Greeks Ck., S side Arahura Valley ( GS2875 , J33/f7075, grid ref. J33/511258; 2 ); C. S. Almond’s (1980) collections from Arahura–Kaniere district in OUGD (topographically lower localities along S side of Arahura Valley are Opoitian; higher ones in the same streams are Waipipian): first left bank tributary Arahura River (J33/f051, grid ref.J33/549382; 1 ); Fraser Creek (J33/f008, grid ref. J33/540286; 1 ; J33/f092, grid ref. J33/540287; 1 ); McKay’s Ck. , Kaniere (J33/f021B, grid ref. J33/515261; 2 ; J33/ f024C, grid ref. J33/513261; 1 ); Greeks Ck. , Arahura (J33/f035, grid ref. J33/546284; 1 , crushed; J33/f036D, grid ref. J33/545283; 1 ; J33/f047, grid ref. J33/542283; 1 ) . Waipipian (early Piacenzian ): Auckland : Otahuhu well (GS 3528, R11/ f7014, grid ref. R11/755698; 1 , in C. R. Laws collection; Marwick , 1948: 6 ; Laws , 1950: 7 , listed from 3 beds in Otahuhu well faunal list); Mouldy’s farm, Pukekohe ( GS 3611, R12/f7001, grid ref. R12/752493; 1; Marwick , 1948: 8 ). Hawke’s Bay: limestone, mouth of Kopuni Stream, W coast of Mahia Peninsula ( GS 11476, X20/f7559,grid ref.X20/277142; 1); Te Reinga Falls, Wairoa River ( GS 1541, X18/f9477, grid ref. X18/021539; 1); near base of Waipipian section, Mohaka River ( GS 13930, W19/f062, grid ref. W19/545366; 1); Mohaka River, Mesopeplum crawfordi ( Hutton, 1873b ) locality at “second flat” (GS13931, W19/f066, grid ref. W19/599369; 1 poor); Esk Valley ( GS 683, V20/f8474, grid ref. V20/423949; 1 ); Black Reef, Cape Kidnappers ( OUGD 8242, W21/f8544, grid ref. W21/604657; 1 fragment). South Taranaki: Manaia Beach, end of Rainie Rd ( GS 875, Q21/f6492, grid ref. Q21/117792; 2 ); Ngamatapouri, Waitotara Valley ( GS 1166, R21/f8497, grid ref.R21/670793; 1); Waingongoro River mouth ( GS 1172, Q21/f6494, grid ref. Q21/123792; 2). Westland: conglomerate at top of E branch, Greek’s Creek, Arahura Valley ( GS 12289, J33/f066, grid ref. J33/547283; 1 ; Fig. 24N ); C. S.Almond’s collections in OUGD :E branch Greeks Ck. (J33/f066, grid ref. J33/547283; 1 ). Chatham Islands: Whenuataru Tuff, near NW end of Tarawhenua Peninsula, Pitt I., Chatham Islands ( GS 12164, CH/f025B, 1 fragment; Fig. 25H ). The only other specimens observed in world museums are the neotype of Janthina typica , the type material of Heligmope dennanti , the neotype of Turbo postulatus , and the holotype of Eunaticina abyssalis , all listed above under “Type material”. Distribution . The type material of Janthina typica was recorded by Zbyszewski & Veiga Ferreira (1962b: 273) from Pinheiros and Feiteirinhas (i.e., Ponta das Salinas), Zanclean localities on Santa Maria Island. Specimens also were recorded from “Feiteirinhas, Ponta dos Matos, Pinheiros, Praia” by early authors, summarized by Berkeley Cotter (1892: 285) , but presumably all specimens from Praia are from close to the Miradouro de Macela and are actually J. krejcii sp. nov. It is also possible that all specimens from Pinheiros are J. krejcii . The neotype of J. typica is from Ponta do Norte; also seen from Ponta do Castelo. The writer has seen no material from Pinheiros or from Feiteirinhas (currently known as Ponta das Salinas). Janthina typica also has been recorded recently from La Esfinge, on La Isleta, north-eastern Gran Canaria, Canary Islands, with photographs confirming the identification ( Meco et al ., 2015: 61 , fig. Appendix 5A; six examined specimens listed above). Janthina typica is also recorded from Selvagem Grande Island, north of the Canary Islands ( Gagel, 1911 ; Joksimowitsch, 1911 ) and from São Vicente on the north coast of Madeira (Mayer, 1864a, b; Krejci-Graf et al ., 1958: 336 ). However, the stratigraphy and age at these sites has not been described in modern times and these records possibly refer to J. chavani . The type material of Acrybia chouberti , from a well a short distance east of Casablanca, Morocco , is the sole record from another eastern Atlantic locality. The holotype of Eunaticina abyssalis ( Simone, 2014: 586 ) is a specimen of J. typica found in situ in rock dredged in 1500–1575 m off Espiritu Santo State, Brazil , SW Atlantic, in a case remarkably parallel to that of the earlier-described species Kaneconcha knorri (see below under J. chavani ). In Japan , Janthina typica is recorded from relatively few late Miocene–Pliocene (Messinian–early Piacenzian) localities along the Pacific coast of Honshu, Shikoku and Kyushu Islands. In Australia , Janthina typica is uncommon and occurs at one locality in Jemmys Point Formation near Lakes Entrance in eastern Victoria , in Grange Burn Formation at Muddy Creek, western Victoria , possibly in the unnamed early Pliocene limestone overlying Kingscote Limestone on Kangaroo Island, and a single specimen is recorded from the lower part of Hallett Cove Sandstone on the coast southwest of Adelaide. In New Zealand , it occurs rarely in Messinian (Kapitean) rocks along the coast between East Cape and Te Araroa and in the Gisborne district , the northernmost locations where latest Miocene rocks are recorded in New Zealand . It also is moderately common at Kaawa Creek, SW Auckland and at Cape Kidnappers, Hawke’s Bay (both Opoitian, Zanclean), and occurs widely but uncommonly in Zanclean rocks in N Hawke’s Bay (22 localities) and Westland (10 localities). Schofield (1958: 252) , based on identifications by C. A. Fleming, also recorded a specimen of “ Hartungia postulata Bart. ” from GS3611, R12/f7541, bank near Glasson’s Creek, north of the Waiuku–Runciman main highway, Auckland , in a re-collection of GS3611, originally listed by Marwick (1948: 8) . Localities in northern Hawke’s Bay were listed (in part) by Marwick (1965 : table 4) and mapped (by GNS collection numbers) on the geological map in Marwick (1965) , although the sole discussion in the text ( Marwick, 1965: 10 ) incorrectly referred to “ Hartungia postulata ” as one of the species first appearing in the Opoitian Stage. Whitten (1973) also recorded Waipipian specimens from South Taranaki between Inaha Stream and Hawera, not seen, in AUGD. Whitten (1973) described 23 new members of Tangahoe Formation, but recorded Janthina typica (as Hartungia postulata ) from three members only: Whareroa Shellbed (one locality, Q21/ f6639), Ohawe Sandstone (three localities, Q21/f6553, f6627, f6631, plus an earlier record from Hawera: Laws 1940b ), and Waingongoro Member (five localities, Q21/ f6613, f6615–f6618). These localities are listed by Whitten (1973 , Appendix I) as: Waingongoro Member: Q21/f6613, grid reference Q21/ 108792, 150 m E of Inaha Stream mouth; Q21/f6615, grid ref. Q21/ 109792, 175 m E of Inaha Stream mouth; Q21/f5616, grid ref. Q21/109792, between the two previous localities; Q21/f6617, grid ref. Q21/111792, 325– 375 m E of Inaha Stream mouth; Q21/f6618, grid ref. Q21/ 112792, 450 m E of Inaha Stream mouth; Ohawe Sandstone: Q21/f6553, grid ref. Q21/134789, SE end of Ohawe Beach; Q21/f6627, grid ref. Q21/156776, mouth of small stream 1.1 km NW of Waihi Beach; Q21/f6631, grid ref. Q21/167768, mouth of Waihi Stream; Q21/f6639, grid ref. Q21/214746, Whareroa Shellbed, 250–450 m NW of small stream near Hawera rifle range. So in Waipipian (early Piacenzian) rocks it occurs in S Auckland (three localities), N Hawke’s Bay (nine localities), the Whanganui-S Taranaki coast (11 localities), the Kaniere-Arahura district, Westland (three localities) and the Chatham Islands (one locality, on NW Pitt Island). These widely separated Atlantic, Japanese, Australian and New Zealand localities indicate that J. typica had a cosmopolitan distribution in tropical and temperate seas, just as living Janthina species do. Dimensions . See Table 3 . Table 3 . Dimensions of Janthina typica . Stage abbreviations: Wo , Opoitian New Zealand Stage (Zanclean, 5.3–3.7 Ma); Wp, Waipipian New Zealand Stage (early Piacenzian, 3.7–3.0 Ma).

locality height		diam.	NSF	H/D
Janthina typica neotype, Santa Maria Is.† 15.4		19.3	13	0.79
Acrybia chouberti holotype*	30+	32	—	—
Turbo postulatus neotype, G5721	31.75	32.1	12	0.99
Hartungia elegans holotype (distorted)*	19.1+	29.8	—	—
Eunaticina abyssalis holotype*	20.6	26.6	(12)	0.77
Australian specimens
SAMA T1494A, H. dennanti lectotype	30.4	28.9	10	1.05
SAMA T1494B, paralectotype	17.5	19.0	10	0.92
NMV P28906, Muddy Creek	+	+	12	—
NMV P26906, Muddy Creek	+	24.5	9	—
NMV P26908, Muddy Creek	27.8	29.3	9	0.95
NMV P26909, Muddy Creek	24.2	25.0	10	0.97
NMV P40662, Muddy Creek	19.5	21.0	8	0.93
NMV P40663, Muddy Creek	25.3	25.2	9	1.00
NMV P40666, Muddy Creek	30+	29.8	9	1.01
NMV P40667, Muddy Creek	29.5	29.8	9	0.99
NMV P316445, Muddy Creek	26.5	30.8	9	0.86
NMV 406640, Muddy Creek	23.2	25.2	10	0.92
GNS WM7656, Muddy Creek	24.7	24.8	9	1.00
NMV P40668, Bunga Ck, Gippsland	19.8	24.8	11	0.79
NMV P27905, Bunga Ck, Gippsland	17.9	22.2	9	0.81
New Zealand specimens
GS5513, Kaawa Ck, SW Auckland, Wo	25.0+	23.1	11	1.08
GS5513, Kaawa Creek	26.0	25.3	10	1.03
GS5513, Kaawa Creek	31.4+	31.7	11	0.99
GS5513, Kaawa Creek	22.8+	24.7	11	0.92
AUGD, Kaawa Creek	26.0	23.1	9	1.13
AUGD, Kaawa Creek	21.9	23.3	9	0.94
AUGD, Kaawa Creek	23.1	22.7	11	1.02
AUGD, Kaawa Creek	22.9	22.0	9	1.04
AUGD, Kaawa Creek	17.3	18.5	9	0.94
AUGD, Kaawa Creek	23.6	28.2	9	0.84
AUGD, Kaawa Creek	23.6	23.4	10	1.01
AUGD, Kaawa Creek	40.7	38.4	11	1.06
AUGD, Kaawa Creek	31.6	31.8	10	0.99
AUGD, Kaawa Creek	35.6	37.2	10	0.96
AUGD, Kaawa Creek	31.8	31.3	10	1.02
GS875, Manaia Beach, S Taranaki, Wp	26.3	29.5	9	0.89
GS1172, Waingongoro R mouth, Wp	29.3	29.0	11	1.01

+ incomplete * dimensions from original publications † dimensions supplied by C. Marques da Silva, Lisbon diam . diameter NSF number of spiral folds Diagnosis . Teleoconch moderately large (to c. 40 mm wide), whorls evenly convex; most specimens heliciform throughout growth, spire consistently low; completely covered with fine, closely spaced axial ridges; 8–12 evenly convex spiral folds per whorl (9 or 10 on most specimens) over entire teleoconch surface, as prominent on sutural ramp as elsewhere. Outer lip sinus relatively small, narrow, semicircular, located at base of lip, generating lowermost, wide spiral fold parallel to columella. Protoconch not seen. Original description . Bronn’s (1861: 119–120) original description of Hartungia typica reads: “ Hartungia typica n. g. sp. [new genus and species]. A very delicate thin and also Ianthina -like shell, filled with rock, 18 mm high and 22 mm wide, with three whorls, which (as in I. communis ) [i.e., Janthina janthina ] form a flatly arched upper side and of which the first two [whorls] are only 5 mm in height. In contrast, the wide ovate aperture measures 17 mm in height and 15mm in width, while its complete lower edge (as in I. nitens Menke ) [i.e., Janthina globosa ] wraps around vertically towards the base. In the same way the umbilicus is not open, but is just in the form of a narrow chink behind the inner lip, which lies conspicuously on the penultimate whorl, as in the said species. The dense fine and elegant vertical striation also recalls it [presumably referring to Janthina exigua ], but does not form a re-entrant sinus in the middle of the outer lip as in Ianthina , but is straight there; in contrast, further down opposite the end of the umbilical chink, [the striation] bends in on a spiral rib [to form] a somewhat insignificant small arch [sinus]. However, what distinguishes this gastropod immediately from all known Ianthina species and would better accord with Narica are 8 flat, broadly rounded spiral cords, which extend down along the outermost convexity of the last whorl, remaining somewhat distant from the suture and even more from the umbilicus, and [of] which the fourth, without forming a keel, is situated furthest towards the outside and scarcely exceeds its upper and lower neighbours in strength. The height of two spiral cords conforms to the width of 6–7 vertical striations. Thus this species is distinguished from Ianthina by the form of the sinuosity and the sculpture of the shell, with which it stands closely in a family and seems to form its own genus, which we name after the indefatigable explorer of the west European islands. This genus probably has some similarity in shape and lip sinus with Neritoma Morris from the Portland beds, but Neritoma is bi-sinuate, somewhat thick-shelled and not umbilicate, behind the inner lip somewhat canal-forming, the growth lines simple. The shell seems too thin for the early whorls of a Magilus and the aperture too regular” (slightly modified translation from German by T. A. Darragh, NMV, pers. comm. 12 Nov 2015 ). Remarks . Bronn (1861: 119–120) compared Hartungia typica with “ Ianthina ” species in several places in his description, stating that it “stands closely in a family” with Janthina , and clearly regarded Hartungia as a new genus of Janthinidae . He appreciated that H. typica is similar to living Janthina species, differing in its small basal sinus and the presence of spiral folds. Mayer (1864: 62) later redescribed the same species in Janthina , apparently changing the species name to maintain the association with the collector, Georg Hartung. Mayer’s (1864b : i–ii) foreword explained that Bronn had died, and so Hartung asked Mayer “in the autumn of 1861” to describe the fossils collected by Reiss at Madeira and Porto Santo. On examining the fossils he found they had a lot in common with those described by Bronn in the works by Hartung and Reiss on the Azores. Therefore, he thought the whole assemblage should be written up together, and asked Hartung and Reiss to help arrange a loan of Bronn’s material from the University of Heidelberg. “With the greatest willingness Professors Blum and Pagenstecher immediately fulfilled my wish, sending me the material able to be found, and so in the first half of the last winter I could attend to my expanded task” (translation from German by T. A. Darragh, NMV, pers. comm. 12 Nov 2015 ). This statement suggests that some of Bronn’s material already could not be traced in 1861. Mayer (1864: 62) listed the description of Hartungia typica by Bronn (1861: 119 , pl. 19, fig. 3) in a chresonymy below the species heading for Janthina hartungi , along with the listing of H. typica by Bronn in Reiss (1862: 32) . So it is clear that he renamed Bronn’s species, although he did not state a reason for doing so. He also compared Janthina hartungi with J. communis (i.e., J. janthina ) and with J. capreolata Montrouzier (i.e., J. exigua ). He stated the dimensions as “Alt. 19, lat. 22 mill.”, only subtly different from the dimensions of 18 mm high and 22 mm wide provided by Bronn (1861: 119) . The list of material at the end of the description makes it clear that Mayer had actual shells before him, as he said the six examples he had included three typical ones from Feiteirinhas (i.e., Ponta das Salinas), another from “Ponta dos Mattos” (presumably the modelling clay impression in Fig. 24M ), one without spiral folds from Pinheiros and one with weak folds from São Vicente (Madeira). The lack of an express statement of a type specimen suggests that Mayer regarded all these as type material of J. hartungi . Mayer’s and Bronn’s illustrations are particularly similar, including the unusual lateral view of the angular columellar base. As Mayer had Bronn’s material before him, the very similar illustrations (Mayer’s: copied as Figs 25D–F ; Bronn’s: copied as Figs 25O–P, S ) likely are different artist’s drawings of the same specimen. In view of the brittleness of Janthina fossils, this material possibly has disintegrated over the intervening years. Janthina typica is characterized by its moderately large size, reaching about 40 mm in diameter and 38 mm in height, although most specimens are 25–35 mm in diameter; its consistent, more-or-less equidimensional, heliciform shape, with a moderately low spire; the few, rapidly expanding teleoconch whorls; the low, thin, closely spaced axial ridgelets covering the entire teleoconch surface, about 1 mm apart over the periphery of large specimens; the quite prominent, evenly convex spiral folds with equally wide, evenly concave interspaces that also cover the entire teleoconch surface, 8–12 on the last whorl (most specimens have 9 or 10 folds), visible on spire whorls as well as on the last whorl; and its relatively small, semicircular sinus in the outer lip, situated at the base of the lip against the columella base. In most specimens, the sinus generates a particularly prominent, wide spiral ridge parallel to the other spiral folds and to the columella and the inner lip of the aperture, higher and wider than the normal spiral folds, although this basal fold is not obviously differentiated in a few specimens. The whorl outline is regularly and strongly convex on most specimens, although a few large shells develop a slightly to quite strongly concave sutural ramp over the last half-whorl. Some of these develop a thickened, smoothly rounded lip edge over the concave area, whereas the lip edge is simple and thin in all other specimens. Traces of the axial ridgelets are visible on almost all specimens, but they are much better preserved on some than on others, and on well-preserved specimens from Kaawa Creek, southwest Auckland , New Zealand (Opoitian, Zanclean) and, in particular, on specimens from Grange Burn Formation (also Zanclean) at Muddy Creek, near Hamilton in western Victoria , Australia , quite high, thin lamellae are preserved. The protoconch has not been observed, despite careful searching of all available material. A supposed protoconch on the holotype of Eunaticina abyssalis ( Simone, 2014: 586 , figs 10I–J) is actually the broken and corroded apex of the Janthina teleoconch ( Fig. 25G ). The teleoconch of J. typica is thin, brittle and calcitic, apart from an interior white aragonitic layer, most obvious on the columella, and chalky in many specimens. Most freshly exposed specimens also have a distinctive pale brownish coloration, reflecting their calcitic composition. Because of the calcitic composition, even small, pale brown fragments of shell bearing axial ridgelets are diagnostic of Janthina in many Pliocene outcrops. Finlay (1931: 5) claimed that a few New Zealand specimens still showed “traces … of the characteristic dark bluish-violet colour”, but the writer has not observed colour on any specimens, and other New Zealand paleontologists consulted also have not observed it. The diagnostic spiral folds are widest and most prominent at the periphery in most specimens, and grade into lower, narrower and more closely spaced ones on the sutural ramp and on the base, except for the wide one generated by the sinus, which on most specimens is almost twice as wide as one peripheral fold. Considerable variation is observed in most characters. The spire is significantly lower in small than in large specimens, as in Janthina chavani and J. janthina , although the weak allometrical growth in spire height observed in J. chavani is not obvious in J. typica . The two specimens observed from Santa Maria Island and the drawings published by Bronn (1861 : pl. 19, fig. 3) and Mayer (1864a, b: pl. 6, fig. 41) show specimens that would not be surprising to find among samples collected in New Zealand or southern Australia . The six specimens observed from the material of Meco et al . (2015 , 2016 ) from La Esfinge, Gran Canaria Island ( Figs 25N, Q–R, T ) are rather fragile, but most are well-preserved, with low to moderately high spires and relatively weak sculpture, but are not as weakly sculptured as the holotype of Eunaticina abyssalis ( Simone, 2014 ) and provide a good basis for understanding the variation that might be expected in the Santa Maria Island population. The two smallest specimens from Gran Canaria are internal moulds, lacking shell. The total range of variation is no greater than is observed between widely separated populations of living Janthina species. Some consistent differences are observed between Australian and New Zealand collections, attributed solely to differences in preservation. Specimens from Grange Burn Formation (Kalimnan Australian Stage, Zanclean) at Muddy Creek and its tributary Grange Burn, near Hamilton, western Victoria , Australia ( Figs 25A–C, L ) are easily removed from weakly consolidated sediment and are much the best-preserved in the world. However, the material from La Esfinge, Gran Canaria Island, and the neotype of Hartungia typica are reasonably well-preserved. Almost all others observed have the axial ridgelets abraded to some extent, and most others are at least a little distorted— some dramatically so. Fine details are more consistently preserved in southern Australian than in New Zealand material, and most New Zealand specimens from localities other than Kaawa Creek are slightly to severely crushed. The few New Zealand Messinian specimens are all either very poorly preserved internal moulds of crushed shells or are very incomplete. Most Japanese specimens also are crushed and many are internal moulds only, although a few are excellently preserved (e.g., Tomida & Itoigawa, 1982: pl. 19, figs 1–3; Tomida & Kitao, 2002: fig. 2). Specimens from the Azores, Canary Islands, Japan , Australia and New Zealand are so similar that there is little doubt they are conspecific, displaying less variation than is observed in the living population of Janthina janthina . Chavan’s (1951: 135 , fig. 1) pen drawing of the holotype of Acrybia ( Hartungia ) chouberti illustrated the sculpture diagrammatically. The identity of this specimen was resolved when Ludbrook (1978 : pl. 12, figs 17–19) published photographs of Chavan’s holotype ( Figs 24J–K, O ) rather than of the poor plaster casts available previously, although the specimen has not been whitened before photography and focus is not ideal. The casts in MNHN and GNS are of a specimen with almost no sculpture, likely made from the paratype internal mould mentioned by Chavan rather than from the holotype . Ludbrook’s (1978 : pl. 12, figs 17–19) illustrations reveal that the holotype has prominent spiral folds all over, including unusually prominent ones on the sutural ramp, and falls within the range of variation of Janthina typica . Ludbrook (1978: 121) cautiously concluded that “while there may be good reason to suspect that the Australian and New Zealand material is identifiable at the species level with the Azores Hartungia typica and its synonym Janthina hartungi ”, she thought that could not be confirmed because of the lack of authentic Azores material for comparison. Now that this has been resolved, the writer sees no reason to doubt that fossil Janthina species had the same cosmopolitan geographical ranges as living Janthina species, and that Heligmope dennanti , “ Turbo ” postulatus , Acrybia chouberti , Hartungia elegans and Eunaticina abyssalis are all synonyms of Janthina typica . It is unfortunate not to retain Georg Hartung’s name. He collected the fossils described by Bronn (1861) and some of those described by Mayer (1864a, b). Pinto & Bouheiry (2007) provided an account of Hartung’s research in the Azores , Madeira and Canary Islands and his extensive collaboration with Charles Lyell. Time range . Messinian–early Piacenzian (latest Miocene– early late Pliocene); c. 7–3.0 Ma, to judge from its range in New Zealand (Kapitean–Waipipian Stages; Messinian–early Piacenzian; Cooper, 2004 : fig. 13.1). This time range is confirmed less precisely in Japan and southern Australia .