Journal of South American Earth Sciences 42 (2013) 216e241

Contents lists available at SciVerse ScienceDirect

Journal of South American Earth Sciences

journal homepage: www.elsevier.com/locate/jsames

Ostracods (Crustacea) and their palaeoenvironmental implication for the Solimões Formation (Late Miocene; Western Amazonia/Brazil)

Martin Gross a,*, Maria Ines Ramos b, Marco Caporaletti c, Werner E. Piller c a Department for Geology and Palaeontology, Universalmuseum Joanneum, Weinzöttlstrasse 16, 8045 Graz, Austria b Coordenação de Ciências da Terra e Ecologia, Museu Paraense Emílio Goeldi, Avenida Perimetral, s/n Terra Firme, Belém-PA 66077-830, Brazil c Institute for Earth Sciences, Karl-Franzens-University, Heinrichstrasse 26, 8010 Graz, Austria article info abstract

Article history: Western Amazonia’s landscape and biota were shaped by an enormous wetland during the Miocene Received 5 March 2012 epoch. Among the most discussed topics of this ecosystem range the question on the transitory influx of Accepted 5 October 2012 marine waters. Inter alia the occurrence of typically brackish water associated ostracods is repeatedly consulted to infer elevated salinities or even marine ingressions. The taxonomical investigation of Keywords: ostracod faunas derived from the upper part of the Solimões Formation (Eirunepé; W-Brazil) documents Western Amazonia a moderately diverse assemblage (19 species). A wealth of freshwater ostracods (mainly Cytheridella, Late Miocene Penthesilenula) was found co-occurring with taxa (chiefly Cyprideis) usually related to marginal marine Solimões Formation d18 d13 Palaeoenvironments settings today. The observed faunal compositions as well as constantly very light O- and C-values Ostracoda obtained by measuring both, the freshwater and brackish water ostracod group, refer to entirely freshwater conditions. These results corroborate with previous sedimentological and palaeontological observations, which proposed a fluvial depositional system for this part of western Amazonia during the Late Miocene. We demonstrate that some endemic, “brackish” water ostracods (i.e., Cyprideis)havebeen effectively adapted to freshwater conditions. Thus, their occurrence is no univocal evidence for the influence of brackish or marine waters in western Amazonia during the Miocene. Ó 2012 Elsevier Ltd. All rights reserved.

1. Introduction and bivalves have been studied intensively during the last years and revealed (aside essential palaeoenvironmental and bio- Today, lowland Amazonia is famous for its biodiversity, which stratigraphical results) prime examples for speciation events obviously root in pre-Quaternary times (e.g., Haffer, 2008; Hoorn related to a variety of ecological factors. Thus, fundamental ques- et al., 2010a; Jaramillo et al., 2010; Wesselingh et al., 2010). The tions of evolutionary biology are touched based on the fossil record evolution of the modern Amazon system and its precursors is (e.g., Wesselingh, 2007; Anderson et al., 2010). Comparably, ostra- synoptically summarised recently (Hoorn and Wesselingh, 2010; cods experienced an extensive radiation during the “Pebas phase” Hoorn et al., 2010a). However, it remains heavily disputed by e.g., (especially the genus Cyprideis; e.g., Whatley et al., 1998), which is, Latrubesse et al. (2007, 2010). Nevertheless, from a palaeobiological however, less well documented and understood. point of view the proposed Middle to (early) Late Miocene “Pebas Studies on Neogene ostracods from the “Pebas system”, started system” (Hoorn et al., 2010a) is of outstanding interest. Albeit of its with the fundamental work of Purper (1977, 1979), followed by still debated nature (mega-lake, e.g., Wesselingh et al., 2002; mega- contributions of Sheppard and Bate (1980), Purper and Pinto (1983, wetland, e.g., Hoorn et al., 2010b; mega-fan (partly), e.g., Latrubesse 1985), Purper and Ornellas (1991) and Swain (1998). Later, the et al., 2010) and chronology, this vast wetland (w1 million km2; comprehensive research of Muñoz-Torres et al. (1998), Whatley Fig. 1a) shaped western Amazonia’s landscapes and life for several et al. (1998, 2000) significantly improved ostracod taxonomy and millions of years. Among aquatic biota it holds a spectacularly stimulated an initial ostracod based biozonation as well as phylo- diverse, largely endemic mollusc and ostracod fauna. Gastropods genetic hypotheses. Additional publications come from Ramos (2006), Celestino and Ramos (2007), Ramos et al. (2009). Lately, the state of the art in ostracodological research is reviewed by * Corresponding author. Tel.: þ43 316 8017 9733; fax: þ43 316 8017 9671. Wesselingh and Ramos (2010). E-mail addresses: [email protected] (M. Gross), mramos@ museu-goeldi.br (M.I. Ramos), [email protected] (M. Caporaletti), The current paper investigates ostracod faunas originating from [email protected] (W.E. Piller). a region, which is suggested to be situated at the edge of the “Pebas

0895-9811/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jsames.2012.10.002 M. Gross et al. / Journal of South American Earth Sciences 42 (2013) 216e241 217

Fig. 1. Location of the study area around Eirunepé (western Amazonia). (a) Schematic illustration of the maximal extent of the “Pebas system” (¼ hatched area) during the Middle Miocene after Wesselingh (2008). (b) Location of investigated outcrops along the cut banks of the Juruá and Tarauacá River.

system” e both in terms of palaeogeography and time (Eirunepé (1977), Purper (1979), Hoorn (1994), Latrubesse et al. (1997, area, western Brazil; Wesselingh et al., 2006a; Wesselingh and 2010), Hoorn et al. (2010b) and Silva-Caminha et al. (2010). Ramos, 2010; Figueiredo, 2012; Fig. 1a, b). We aim to provide a profound taxonomical base, which is crucial for forthcoming palaeoecological, biostratigraphical and phylogenetic research. Our 3. Materials and methods systematic evaluation is supplemented by compilations of the autecology as well as by detailed illustrations, inclusively of For micropalaeontological investigations bulk samples (w1e2kg) different sexes and ontogenetic stages. Ostracodological results of were taken from all outcrops under investigation (Gross et al., 2011). the present study are complementary to an earlier sedimentolog- 500 g of dried sediment (40 C, 24 h) were washed by using diluted fl ical analysis, which documented an aggrading uvial system as hydrogen superoxide for disintegration through standard sieves depositional environment for the examined outcrops (Gross et al., (H2O2:H2O ¼ 1:5; 63/125/250/500 mm). Wet sieve residuals were 2011). washed with ethanol (70%) before drying (40 C, 24 h). Residuals 250 mm were picked out completely and subject of 2. Geological setting detailed taxonomic investigations. From the 125 mm sieve-residual 0.2 g/sample were picked, which contained mainly, hardly distin- The sampled sections are situated along the river banks of the guishable juvenile and/or fragmented ostracod valves. Therefore Juruá and Tarauacá River, NE respectively SE of Eirunepé (state of the following taxa are not further differentiated within this fraction Amazonia; Solimões Basin (Jandiatuba Sub-Basin); e.g., Caputo, (for details see ESM 1): Cypria sp. and Physocypria sp. are counted as 1991; Wanderley-Filho et al., 2010; Fig. 1a, b). Aside Quaternary Cypria/Physocypria. Cyprideis graciosa, Cyprideis longispina, Cypri- deposits (alluvium, terraces) the studied outcrops (Remanso, deis pebasae, C. aff. pebasae, Cyprideis sp. 1 and 2 are summarised Aquidabã, Morada Nova, Pau D’Alho, Torre da Lua, Barro Branco) under Cyprideis “ornate” as well as Cyprideis aff. machadoi and expose sediments of the upper part of the Solimões Formation Cyprideis ?olivencai are subsumed under Cyprideis “smooth”. (Del’ Arco et al., 1977; Maia et al., 1977; Paz et al., in press). Prior to stable isotope analyses (d18O, d13C; 48 measurements) Detailed sedimentological descriptions of these locations and ostracod valves were additionally washed with distilled water in an interpretations are already presented in Gross et al. (2011) to ultrasonic bath and rinsed in ethanol finally. Adults and juveniles which we refer here. (A-1 instars) of 5 species were measured: C. pebasae, C. graciosa, The more than 1000 m thick Solimões Fm. covers most of Cytheridella danielopoli, Penthesilenula olivencae (only adults), western Amazonia and comprises peliticesandy alternations, Rhadinocytherura amazonensis (only adults). The number of valves lignitic intercalation as well as paleosols. Diverging views on in its required for analyses (w50 mg) varied between 2 and 7. For analyses definition, its stratigraphical and geographical range as well of its a Thermo-Finnigan Kiel II automated reaction system and depositional environments clearly mirror the controversially a Thermo-Finnigan Delta Plus isotope-ratio mass spectrometer debated history of Amazonia through Neogene times. In-depth were used (University of Graz; standard deviation ¼ 0.1& relative reviews of the Solimões Fm. are provided by e.g., Del’ Arco et al. to NBS-19; results in per mille relative to VPDB). 218 M. Gross et al. / Journal of South American Earth Sciences 42 (2013) 216e241

4. Results Juveniles of Vestalenula Rossetti & Martens, 1998 have antero- and posteroventral teeth in left valves (which are absent in adults) 4.1. Systematic palaeontology and no external posteroventral keel in right valves (which is diag- nostic for adults; e.g., Artheau, 2007; Minati et al., 2008; Smith and All figured specimens are housed in the collection of the Museu Kamiya, 2008). Due to the presence of a brooding cavity and the Paraense Emílio Goeldi, Belém (Inv. No. MPEG-90-M to MPEG-193- anterocentral position of the central muscle scars, here mentioned M), additional material is stored at the Universalmuseum Joan- specimens represent adult (female) individuals, which are also neum, Department for Geology & Palaeontology, Graz (Inv. No. larger than species of Vestalenula (Rossetti and Martens, 1998). UMJG&P 210.903). Moreover, the posterior internal tooth of left juvenile valves of The suprageneric classification follows Meisch (2000) with Vestalenula is situated posteroventrally and not caudal (Artheau, some adaptations after Martens and Savatenalinton (2011). General 2007; Smith and Kamiya, 2008; Minati et al., 2008). descriptions of species are provided by Purper (1979), Sheppard The genus Isabenula Rossetti, Pinto & Martens, 2011 can be and Bate (1980) and Muñoz-Torres et al. (1998; Whatley et al., excluded due to the presence of a short, external posteroventral 1998; Ramos, 2006). Further characteristics or deviations are dis- keel in right valves as well as it develops only an anteroventral, cussed in the remarks. internal tooth (Rossetti et al., 2011). Abbreviations: R ¼ right valve; L ¼ left valve; \ ¼ female; The genus Alicenula Rossetti & Martens, 1998 with its elongated _ ¼ male; A-1, ... ¼ juvenile stages; l ¼ length, h ¼ height (both in valves and internal teeth in left valves (Martens et al., 2003)is millimetres; length of spines not included in measurements), similar. Because it was initially described without possessing n ¼ number of measured specimens. internal teeth (Ballent and Díaz, 2011), we follow the original diagnosis of Rossetti and Martens (1998), Pinto et al. (2004) and Class Ostracoda Latreille, 1806 attribute our material to the genus Penthesilenula (diagnostic Order Podocopida Sars, 1866 characters of valves: elongated to sub-squarish in lateral view, Suborder Podocopina Sars, 1866 absence of an external keel in the right valve, internal teeth in the Superfamily Darwinuloidea Brady & Norman, 1889 left valve). Family Darwinulidae Brady & Norman, 1889 Probably, within Penthesilenula the current specimens belong to Genus Penthesilenula Rossetti & Martens, 1998 the “incae”-group (more elongated valves than the “africana”- Penthesilenula olivencae (Purper, 1984) comb. nov. group; pointed caudal- and anteroventral internal tooth); the latter Pl. 1, figs. 1e10, 22e23 feature comparable with the teeth in the recent Penthesilenula 1977 Darwinula sp. e Purper: 365, pl. 4, figs. 5e8. aotearoa Rossetti, Eagar & Martens, 1998; see Rossetti and Martens * 1979 Darwinula fragilis Purper, sp. nov. e Purper: 225, pl. 1, figs. (1998) and Pinto et al. (2004). 4e10. In the Eirunepé material, adult P. olivencae specimens are ? 1980 Darwinula sp. e Sheppard and Bate: 117e118, pl. 13, fig. 7. associated with smaller darwinulid valves with comparable outline 1984 Darwinula olivencae e Purper: 1371. and central muscle scar pattern. Although internal teeth are 1998 D. fragilis Purper, 1979 e Muñoz-Torres et al.: 90, pl. 1, figs. missing (a feature which can change during ontogeny; compare 1e3. Vestalenula), the more postero(dorsal) position of central muscle 2006 D. fragilis Purper, 1979 e Ramos: 89e90, figs. 6aec. scars clearly points to juvenile darwinulids, which we consider to 2009 D. fragilis Purper, 1979 e Ramos et al.: 116, fig. 305-I. represent juveniles of P. olivencae. 2010 Alicenula (Darwinula) fragilis Purper,1979 e Wesselingh and Ecology: Darwinulid ostracods live in various freshwater habi- Ramos: 309, figs. 18.7eef. tats such as lakes, ponds, springs, rivers, as well as in (semi-) terrestrial and interstitial habitats but also in mixohaline environ- Material: 271 adults, 7 juveniles (250 mm sieve fraction), 742 ments (e.g., Rossetti and Martens, 1998; Higuti et al., 2009a). mainly juveniles and some adult fragments (125 mm sieve fraction). Extant species of Penthesilenula are found in lakes, rivers, Dimensions:R\l ¼ 0.67e0.73 (0.69), h ¼ 0.26e0.28 (0.26; n ¼ 7); swamps, in interstitial and terrestrial habitats (wet leaf litter or L\l ¼ 0.68e0.74 (0.70), h ¼ 0.27e0.30 (0.28; n ¼ 10); R?A-1 l ¼ 0.54, moss; Van Doninck et al., 2003; Pinto et al., 2004; Ballent and Díaz, h ¼ 0.20 (n ¼ 1); L?A-1 l ¼ 0.55e0.58, h ¼ 0.23e0.24 (n ¼ 2); R?A-2 2011). It is demonstrated experimentally to tolerate a wide salinity l ¼ 0.39, h ¼ 0.15 (n ¼ 1); L?A-2 l ¼ 0.40, h ¼ 0.17 (n ¼ 1). range (distilled water to 25 PSU). However, in natural habitats Remarks: Based on their similar valve morphology (outline, size) Penthesilenula is usually restricted to salinities <3 PSU and is the present specimens are assigned to the species D. fragilis of recorded only rarely in brackish, estuarine/lagoonal environments Purper (1979) which she renamed 1984 into D. olivencae according (e.g., Würdig et al., 1990; Ferrero, 1996; Van Doninck et al., 2003). to article 57.2 of ICZN. However, neither descriptions nor figures of Species of the “incae”-group (i.e., Penthesilenula incae (Dela- the listed publications provide adequate details (especially of chaux, 1928), P. aotearoa) are almost exclusively known from internal valve features) for a robust synonymisation. Left valves “warm” (>10 C) freshwater habitats and are especially abundant in from Eirunepé exhibit prominent caudal- and anteroventral lotic habitats with sandy substrate (Van Doninck et al., 2003; Higuti internal teeth (pl. 1, fig. 3). Such teeth can be anticipated in fig. 6 et al., 2009a; Ballent and Díaz, 2011). (plate 1) of Purper (1979) and support the assignment. In our Occurrence: Western Amazonia, early Middle to early Late material a very small anteroventral internal tooth is developed in Miocene (Peru: Cyprideis aulakoseCyprideis caraione zone; right valves too (pl. 1, fig. 4, 10). There is no posteroventral external Colombia and Brazil: Cyprideis obliquosulcataeCyprideis cyrtoma keel developed in right valves; a posterior brood pouch is visible. zone; Muñoz-Torres et al., 2006; chronostratigraphic correlation Currently, six Darwinulidae genera are described (Rossetti and after Wesselingh and Ramos, 2010). This work: Aquidabã, Morada Martens, 1998; Rossetti et al., 2011). Nova, Pau D’Alho, Torre da Lua, Barro Branco. The genus Darwinula Brady & Robertson, 1885 lacks internal teeth in left valves. Microdarwinula Danielopol, 1968 is much Superfamily Cypridoidea Baird, 1845 smaller and its valves are rounded in lateral view and lack a brood Family Candonidae Kaufmann, 1900 pouch. Thus, both genera are not further considered for the present Subfamily Candoninae Kaufmann, 1900 material. Genus Pseudocandona Kaufmann, 1900 M. Gross et al. / Journal of South American Earth Sciences 42 (2013) 216e241 219

Plate1. (1e10) Penthesilenula olivencae. (1) Le (0.74/0.30; AQ5-2_69), (2) Re (0.71/0.27; AQ6-2_24R), (3) Li (¼ fig. 1), (4) Ri (¼ fig. 2), (5) Li?A-1 (0.58/0.24; AQ19_46), (6) Ri?A-1 (0.54/ 0.20; AQ19_49), (7) Li?A-2 (0.40/0.17; AQ19_48), (8) Ri?A-2) (0.39/0.15; slightly deformed; AQ19_51), (9) central muscle scars of Li (detail of fig. 3), (10) anteroventral internal tooth in Ri (detail of fig. 4). (11e13) Pseudocandona sp. (11) Re (0.65/0.40; AQ16_02), (12) Ri (¼ fig. 11), (13) central muscle scars of Re fragment (transmitted light, redrawn; AQ16_03R). (14e15) Cypria sp. (14) Le (0.61/0.43; AQ6-2_25), (15) Li (¼ fig. 14; 15b: anteroventral “contact knob”.(16e21) Physocypria sp. (16) Le (0.53/0.36; AQ6-2_26), (17) Re (0.50/0.34; AQ8- 2_06), (18) Li (¼ fig. 16), (19) Ri (¼ fig. 17; 19b: posteroventral marginal tubercles), (20) Ld (¼ fig. 16), (21) Rd (0.53/0.35; AQ8-2_05). (22e23) Penthesilenula olivencae (22) Ld (0.70/ 0.28; TO12_02), (23) Rd (0.71/0.27; TO12_04). (24e26) Cypretta sp. (24) Re (0.69/0.50; BA7a_01), (25) Ri (¼ fig. 24; 25b: nodes at the anteroventral margin), 26) Rd (¼ fig. 24). (Abbreviations: R ¼ right valve; L ¼ left valve; e ¼ external view; d ¼ dorsal view; i ¼ internal view; \ ¼ female; _ ¼ male; A-1, . ¼ juvenile stages. In brackets length (l) and height (h) are indicated in millimetres (e.g., 0.74/0.30 ¼ length 0.74 mm, height 0.30 mm; the length of spines is not included) followed by a specimen code (sample number_specimen number; e.g., AQ5-2_69). Sample number ¼ abbreviated location followed by the layer number; e.g., AQ5-2 ¼ Aquidabã, layer number 5-2; AQ ¼ Aquidabã, BA ¼ Barro Branco, MN ¼ Morada Nova, PD ¼D Pau ’Alho, TO ¼ Torre da Lua (Gross et al., 2011)). 220 M. Gross et al. / Journal of South American Earth Sciences 42 (2013) 216e241

Pseudocandona sp. antero- and posteroventrally (pl. 1, figs. 19aeb; Van Morkhoven, Pl. 1, figs. 11e13 1963; Meisch, 2000). 2010 ?Heterocypris sp. e Wesselingh and Ramos: 309, figs. Ecology: Cypria and Physocypria are active swimmers occurring 18.7geh. in freshwater habitats like ponds and lakes but can be found also in oligo- to mesohaline waters (Van Morkhoven, 1963; Hartmann, Material: 1 right valve, 1 badly preserved carapace, 3 fragments 1989; Meisch, 2000). (250 mm sieve fraction). Occurrence (of C. aqualica): Western Amazonia, early Middle to Dimensions:Rl¼ 0.65, h ¼ 0.40 (n ¼ 1). early Late Miocene (Peru: C. aulakoseCyprideis minipunctata zone; Remarks: Only a few specimens are on hand. Internal features Colombia and Brazil: C. obliquosulcataeC. cyrtoma zone; Muñoz- are largely obscured by adhering sediment. However, one fragment Torres et al., 2006; chronostratigraphic correlation after Wesselingh displays central muscle scars (pl. 1, fig. 13) typical for candonid and Ramos, 2010). This work (Cypria sp.): Aquidabã, Morada Nova, ostracods, which exclude their allocation to Heterocypris Claus, Pau D’Alho, Torre da Lua, Barro Branco. 1892 (Van Morkhoven, 1963; Meisch, 2000). Small pustules at the anterior and posterior margins of right valves are missing, which Genus Physocypria Vávra, 1897 are usually developed in Heterocypris (e.g., Purper and Würdig- Physocypria sp. Maciel, 1974; Meisch, 2000). Pl. 1, figs. 16e21 Although an assignment to a specific Candoninae genus based solely on similarities of the valve morphology remains tentative, we Material: 11 mainly fragmented valves (250 mm sieve fraction). attribute the current specimens to the candonid genus Pseudo- Dimensions:Rl¼ 0.50e0.54 (0.52), h ¼ 0.32e0.35 (0.34; n ¼ 5); candona. Perhaps, they represent a species of the “compressa”- Ll¼ 0.53e0.54, h ¼ 0.34e0.36 (n ¼ 2). group (short and stout in lateral view; dorsal margin straight, Remarks: This is the first record of the genus Physocypria from slightly sloping anteriorly; ventral margin almost straight; Meisch, the Solimões Formation. More material is necessary for further 2000; Namiotko and Danielopol, 2004). taxonomic analyses. Perhaps, the present material comprises Ecology: Candonids are mainly found in freshwater habitats (e.g., juvenile valves only. pools, lakes, swampy habitats) but can tolerate slightly saline Ecology: see Cypria. (brackish) waters (Van Morkhoven, 1963). Species of the Pseudo- Occurrence (this work): Aquidabã, Torre da Lua. candona “compressa”-group prefer shallow freshwater environ- ments but are recorded from oligo- to mesohaline settings too Family Ilyocyprididae Kaufmann, 1900 (Meisch, 2000). Subfamily Pelocypridinae Triebel, 1962 Occurrence: Thus far, Candoninae are only recorded by Genus Pelocypris Klie, 1939 Wesselingh and Ramos (2010) and Linhares et al. (2011) from the Pelocypris zilchi Triebel, 1953 Neogene of western Amazonia. This work: Aquidabã. Pl. 2, figs. 1e13 * 1953 P. zilchi n. sp. e Triebel: 2e4, pl. 1, figs. 1e8. Subfamily Cyclocypridinae Kaufmann, 1900 1980 P. zilchi Triebel, 1953 e Sheppard and Bate: 104e106, text- Genus Cypria Zenker, 1854 fig. 3; pl. 10, figs. 8e13. Cypria sp. 1980 Pelocypris sp. e Sheppard and Bate: 108, pl. 10, fig. 14. Pl. 1, figs. 14, 15aeb 2010 Ilyocypris (Pelocypris) zilchi (Triebel, 1953) e Wesselingh ? 1998 Cypria aqualica Sheppard & Bate, 1980 e Muñoz-Torres and Ramos: 309, fig. 18.7c. et al.: 91, pl. 1, figs. 4e6. Material: 8 adults, 15 juveniles (both mainly fragmented; Material: 33 mainly fragmented valves (250 mm sieve fraction). 250 mm sieve fraction). Dimensions:Ll¼ 0.61e0.67 (0.65), h ¼ 0.43e0.46 (0.44; n ¼ 3). Dimensions:R?_ l ¼ 1.62, h ¼ 0.90 (n ¼ 1); L?\ l ¼ 1.45e1.55 Remarks: These specimens are placed in Cypria due their later- (n ¼ 2), h ¼ 0.88e1.03 (0.97; n ¼ 3); R?A-1 l ¼ 1.17e1.25, h ¼ 0.72 ally compressed, sub-circular valves and the presence of one (n ¼ 2); R?A-2 l ¼ 1.11, h ¼ 0.67 (n ¼ 1); R?A-3 l ¼ 0.85 (broken), anteroventral “contact-knob” in left valves (pl. 1, figs. 15aeb; Van h ¼ 0.56 (n ¼ 1); L?A-2 l ¼ 0.97e1.07 (all broken), h ¼ 0.57e0.64 Morkhoven, 1963; Malz, 1977; Meisch, 2000). (0.61; n ¼ 3); L?A-3 l ¼ 0.80 (broken), h ¼ 0.49 (n ¼ 1); L?A-4 The only formally described species of Cypria in the Neogene of l ¼ 0.76, h ¼ 0.42 (n ¼ 1); L?A-5 l ¼ 0.58e0.60 (n ¼ 2), h ¼ 0.38 (n ¼ 1). western Amazonia is C. aqualica, which differs due its almost Remarks: These ilyocypridid ostracods clearly belong to Pelocypris straight ventral margin for the current form (Sheppard and Bate, (generic characters of the valve: the bifurcate sulcus does not reach 1980). Usually, the attribution of Cypria valves to that species is the central muscle scar area); it lacks central depressions corre- based on very few specimens (Purper, 1979; Sheppard and Bate, sponding to central muscle scars; at the anterior margin an inner list 1980; Ramos, 2006; Ramos et al., 2009; Wesselingh and Ramos, is missing (Tressler,1949; Triebel,1953; Lister,1975; Hartmann,1989). 2010). Thus, little is known about intraspecific variability or onto- Within the current material noded as well as un-noded instars genetic stages (note that Muñoz-Torres et al., 1998 suggested the co-occur, while this feature is not observed on adult valves. These type-material of Sheppard and Bate (1980) to represent juveniles). tubercles correspond to normal pore canal openings. The position Equally, the present investigation is founded on scarce, largely of the main tubuli respectively pores can be traced through fragmented and possibly juvenile valves, which hampers species different ontogenetic stages up to adult specimens with their only identification. Although the specimens of Muñoz-Torres et al. weakly developed pore tubuli. By considering the large ecophe- (1998) are significantly larger, they are similar to the material notypic variability of tubercles in other ilyocypridids (Ilyocypris; from Eirunepé. With reservation we synonymise the herein Meisch, 2000; Yang et al., 2002), the noded specimen of Sheppard mentioned specimens with the material from Muñoz-Torres et al. and Bate (1980: Pelocypris sp.) is included in the synonymy here. (1998). Different ecological parameters are discussed to be related to the Physocypria sp. (see below) differs in being smaller and having presence/absence of nodes in Ilyocypris (e.g., temperature, salinity, a more truncate posterior and almost straight dorsal margin. locomotion; Van Harten, 1979; Yang et al., 2002) and might play Moreover, right valves bear generotypical, marginal tubercles a comparable role in Pelocypris. M. Gross et al. / Journal of South American Earth Sciences 42 (2013) 216e241 221

Plate 2. (1e13) Pelocypris zilchi (1) Le?\ (1.55/1.03; AQ8-2_12), (2) Le?A-2 (1.07/0.64; AQ19_34), (3) Le?A-2 (0.97/0.57; AQ19_35), (4) Le?A-3 (0.80/0.49; AQ8-2_09), (5) Le?A-4 (0.76/0.42; AQ8-2_10), (6) Le?A-5 (0.60/0.38; AQ8-2_11), (7) Re?_ (1.62/0.90; AQ19_42), (8) Re?A-1 (1.25/0.72; AQ19_43), (9) Re?A-2 (1.11/0.67; AQ19_45), (10) Li?\ (1.45/0.88; AQ8-2_13), (11) Ri?A-1 (¼ fig. 8), (12) Ld?\ (¼ fig. 1), (13) Le?A-4 (¼ fig. 5; anterior marginal denticles). (14e19) Cytheridella danielopoli (14) central muscle scars of Li_ (0.88/0.47; TO12_05), (15) Ld\ (1.01/0.59; AQ19_01), (16) Rd\ (1.03/0.61; AQ19_37), (17) Rd\ (1.00/0.58; AQ5-2_64), (18) Ld_ (0.94/0.51; AQ19_09), (19) Rd_ (0.95/0.51; AQ19_23). (For abbreviations see plate 1).

So-called marginal denticles are well-developed along the free too (pl. 2, fig. 5, 13). Due to limited material a discussion of the valve margin of right adult valves of P. zilchi but are lacking in left validity of the species-diagnostic criteria for P. zilchi and its delin- ones (Triebel, 1953; Sheppard and Bate, 1980). However, tiny eation from the evidently related recent Pelocypris lenzi Klie, 1939 is denticles are present in left juvenile valves of the Eirunepé material left open for subsequent investigations (differential diagnosis to 222 M. Gross et al. / Journal of South American Earth Sciences 42 (2013) 216e241

P. lenzi: P. zilchi is smaller, has a better developed ornamentation, shorter pore conuli and marginal denticles only in right valves; Triebel, 1953). The differentiation of different sexes and ontogenetic stages is provisional because of the lack of sufficient specimens (some are additionally fragmented). Ecology: Pelocypris is considered to inhabit freshwater settings like streams and not saline playa lakes (Sheppard and Bate, 1980; Hartmann, 1989; Horne, 1996; Reheis et al., 2005). Occurrence: Western Amazonia, Early Miocene (La Tagua; Sheppard and Bate, 1980; mollusc zone 2 according to Wesselingh et al. (2006b); chronostratigraphic correlation after Wesselingh and Ramos, 2010); El Salvador, Pleistocene (Barranca El Sisimico; Triebel, 1953; see Cisneros, 2005 for age). This work: Aquidabã.

Family Cyprididae Baird, 1845 Subfamily Cyprettinae Hartmann, 1971 Genus Cypretta Vávra, 1895 Cypretta sp. Pl. 1, figs. 24e26 Fig. 2. Length/height diagram for Cytheridella danielopoli.

Material: 1 right valve (250 mm sieve fraction). Material: 131 adults, 1863 juveniles (250 mm sieve fraction), 176 Dimensions:Rl¼ 0.69, h ¼ 0.50 (n ¼ 1). juveniles and fragments (125 mm sieve fraction). Remarks: One single, subovate, finely reticulated right valve with Dimensions:R\l ¼ 1.00e1.08 (1,00), h ¼ 0.58e0.63 (0,57; n ¼ 7); a remarkable central tubercle was found. It largely coincides with L\l ¼ 0.96e1.07 (1.01), h ¼ 0.53e0.64 (0.58; n ¼ 7); R_l ¼ 0.91e0.95 the diagnostic features of the genus Cypretta (e.g., Furtos, 1934; (0.92), h ¼ 0.48e0.51 (0.50; n ¼ 7); L_l ¼ 0.93e0.95 (0.94), h ¼ 0.51e Sohn and Kornicker, 1973; Victor and Fernando, 1981). Unfortu- 0.53 (0.51; n ¼ 4); RA-1 l ¼ 0.72e0.85 (0.79), h ¼ 0.39e0.48 (0.44; nately, the valve is not translucent enough to proof the presence of n ¼ 10); LA-1 l ¼ 0.72e0.83 (0.78), h ¼ 0.41e0.47 (0.44; n ¼ 11);RA-2 diagnostic septa along the anterior margin. Because only one valve l ¼ 0.62e0.64 (0.63), h ¼ 0.35e0.36 (0.35; n ¼ 7); LA-2 l ¼ 0.62e0.66 is available, we avoided to fracture the valve for studying the (0.64), h ¼ 0.35e0.38 (0.36; n ¼ 9); RA-3 l ¼ 0.52e0.59 (0.56), marginal valve structure. However, small nodes are developed h ¼ 0.30e0.33 (0.32; n ¼ 7); LA-3 l ¼ 0.52e0.61 (0.56), h ¼ 0.31e0.33 between the flange and the selvage along the anterior and poster- (0.32; n ¼ 11); LA-4 l ¼ 0.44, h ¼ 0.26 (n ¼ 1). oventral margin. Remarks: Females of C. danielopoli differ from males due to their This specimen resembles extant Cypretta brevisaepta Furtos, larger size and their well expressed brood pouch (compare Purper, 1934, a species with reverse valve overlap (left over right overlap), 1974; Pérez et al., 2010a; pl. 2, figs. 15e19). The valve surface is which is visible here likewise (e.g., Keyser, 1976; Smith and usually pitted, which is less pronounced along the anterior margin Delorme, 2010; for discussions on the taxonomical value of this and at the dorsomedian sulcus. Nevertheless, some adult speci- feature see: Sohn and Kornicker, 1973; Keyser, 1976; Victor and mens are almost smooth and display a faint reticulation only (pl. 3, Fernando, 1981; Holmes, 1998). fig. 17). Beside ornamentation there are no additional differences, Ecology: Cypretta is an actively swimming, freshwater ostracod, neither in shape nor in internal features; the position of pore tubuli which is rarely found in oligohaline environments. It is recorded is strikingly equivalent. Also within instars the degree of orna- mainly from tropical and subtropical climates (Sohn and Kornicker, mentation and the expression of sulci vary (pl. 3, figs. 5, 12, 13). 1973; Keyser, 1976; Holmes, 1998; Pérez et al., 2010a,b). The current material almost match with the type-specimens of Occurrence (this work): Barro Branco. Purper (1979), which are somewhat smaller, a little bit stronger punctated and their dorsal margin is slightly less inclined towards Superfamily Cytheroidea Baird, 1850 anterior. Based on the observed variability in ornamentation and Family Limnocytheridae Klie, 1938 outline the Eirunepé material is conspecific with C. danielopoli. C. Subfamily Timiriaseviinae Mandelstam, 1960 postornata of Sheppard and Bate (1980), which has a smooth valve Genus Cytheridella Daday, 1905 surface anterior and a pitted ornamentation posterior of the sulcus, Cytheridella danielopoli Purper, 1979 is here considered to be a junior synonym of that species (compare Fig. 2; pl. 2, figs. 14e19; pl. 3, figs. 1e28 van den Bold, 1986). C. danielopoli of Muñoz-Torres et al. (1998) 1977 Cytheridella sp.nov. A e Purper: 365, pl. 4, figs. 1e4. belongs to A-1 instars as already suggested by these authors. *1979 C. danielopoli Purper, sp. nov. e Purper: 243e244, pl. 7, figs. Further, C. purperae of Ramos (2006; compare Ramos et al., 2009; 21e27. Wesselingh and Ramos, 2010) is based on juvenile valves (probably 1980 Cytheridella postornata sp. nov. e Sheppard and Bate: 108e A-2) and range within the intraspecific variability of C. danielopoli 110, pl. 10, figs. 1e7. (pl. 3, figs. 5, 12, 13). Likewise, Cytheridella sp. from Ramos (2006), 1998 C. danielopoli Purper, 1979 e Muñoz-Torres et al.: 104, pl. 6, Ramos et al. (2009) represents the smooth morphotype of figs. 13e14. C. danielopoli. 2006 Cytheridella purperae sp. nov. e Ramos: 92e93, figs. 7oev. The type-species of Cytheridella, Cytheridella ilosvayi Daday, 2006 Cytheridella sp. e Ramos: 93e94, figs. 7xey. 1905, is very similar to C. danielopoli and belongs to the same 2009 Cytheridella sp. e Ramos et al.: 116, figs. 299e300-I. lineage (for C. ilosvayi see: Löffler, 1961; Purper, 1974; Pérez et al., 2009 C. purperae Ramos, 2006 e Ramos et al.: 116, figs. 301e 2010a; Smith and Delorme, 2010). Differences to other species are 304-I. discussed in Purper (1979) and Sheppard and Bate (1980). 2010 C. purperai [sic!] Ramos, 2006 e Wesselingh and Ramos: Ecology: Cytheridella is almost exclusively restricted to fresh- 309, figs. 18.7aeb. water environments (Purper, 1974; Colin et al., 1997). Extant M. Gross et al. / Journal of South American Earth Sciences 42 (2013) 216e241 223

Plate 3. (1e28) Cytheridella danielopoli (1) Le\ (1.01/0.59; AQ19_01 ¼ pl. 2, fig. 15), (2) Le_ (0.93/0.51AQ19_07), (3) LeA-1 (0.81/0.46; AQ5-2_51), (4) LeA-1 (0.80/0.43; AQ5-2_50), (5) LeA-2 (0.66/0.37; TO12_11; forma purperae), (6) LeA-2 (0.66/0.38; AQ5-2_53L), (7) LeA-3 (0.55/0.33; AQ19_14), (8) LeA-4 (0.44/0.26; AQ5-2_58), (9) Re\ (1.03/0.61; AQ19_37 ¼ pl. 2, fig. 16), (10) Re_ (0.95/0.51; AQ19_23 ¼ pl. 2, fig. 19), (11) ReA-1 (0.82/0.47; AQ5-2_59), (12) ReA-1 (0.79/0.44; AQ5-2_60; forma purperae), (13) ReA-2 (0.64/0.36; AQ19_25; forma purperae), (14) ReA-2 (0.62/0.35; TO12_21), (15) ReA-3 (0.54/0.32; AQ19_29), (16) LiA-4 (¼ fig. 8), (17) Re\ (1.00/0.58; AQ5-2_64 ¼ pl. 2, fig. 17), (18) Ri\ (¼ fig. 17), (19) RiA-3 (¼ fig. 15), (20) RiA-2 (¼ fig. 14), (21) RiA-1 (0.76/0.42; AQ5-2_62), (22) Ri_ (¼ fig. 10), (23) Ri\ (¼ fig. 9), (24) LiA-3 (¼ fig. 7), (25) LiA-2 (¼ fig. 5), (26) LiA-1 (0.82/0.47; AQ5-2_49), (27) Li_ (0.94/0.51; AQ19_09 ¼ pl. 2, fig. 18), (28) Li\ (¼ fig. 1). (For abbreviations see plate 1). 224 M. Gross et al. / Journal of South American Earth Sciences 42 (2013) 216e241

Plate 4. (1e17) Cyprideis graciosa (1) Le\ (0.82/0.46; MN14_08), (2) Le_ (0.88/0.45; MN14_06), (3) LeA-1 (0.61/0.35; PD25_04), (4) LiA-1 (¼ fig. 3), (5) Re\ (0.79/0.44; MN14_04), (6) Re_ (0.84/0.41; MN14_02), (7) ReA-1 (0.61/0.34; PD25_07), (8) RiA-1 (¼ fig. 7), (9) Li_ (¼ fig. 2), (10) Ri_ (¼ fig. 6), (11) Li\ (¼ fig. 1), (12) central muscle scars of Ri (detail of fig. 13), (13) Ri\ (¼ fig. 5), (14) Ld_ (0.86/0.44; MN14_07), (15) Rd_ (¼ fig. 6), (16) Ld\ (¼ fig. 1), (17) Rd\ (¼ fig. 5). (18e34) Cyprideis longispina (18) Le\ (0.90/0.48; AQ5-2_01), (19) Le_ (0.89/0.43; AQ5-2_06), (20) LeA-1 (0.66/0.36; TO12_37), (21) LiA-1 (¼ fig. 20), (22) Re\ (0.87/0.44; AQ5-2_12), (23) Re_ (0.89/0.41; AQ5-2_17), (24) ReA-1 (0.64/0.35; AQ5-2_21), (25) RiA-1 (¼ fig. 24), (26) Li\ (¼ fig. 1), (27) Ri\ (¼ fig. 22), (28) Li_ (¼ fig. 19), (29) central muscle scars of Ri_ (0.87/0.40; AQ5-2_16), (30) Ri_ (¼ fig. 23), (31) Ld_ (0.89/0.42; AQ6- 2_02), (32) Rd_ (¼ fig. 23), (33) Ld\ (¼ fig. 18), (34) Rd\ (¼ fig. 22). (For abbreviations see plate 1). M. Gross et al. / Journal of South American Earth Sciences 42 (2013) 216e241 225

C. ilosvayi is found in shallow (<40 m), permanent, freshwater to spines are observed in right valves). Similar to adult specimens the slightly saline (<3.2 PSU) waters in subtropical/tropical (>20 C) anterior hinge element is more elongated in C. longispina-instars climates (Alvarez Zarikian et al., 2005; Pérez et al., 2010a,b). than in juveniles of C. graciosa. Occurrence: Western Amazonia, late Middle to early Late Ecology: The genus Cyprideis is a holoeuryhaline (freshwatere Miocene (Peru: C. caraione zone; Colombia and Brazil: hypersaline), pandemic ostracod genus; however, most C. obliquosulcataeC. cyrtoma zone; Muñoz-Torres et al., 2006; commonly found in shallow, oligo- and mesohaline (brackish) chronostratigraphic correlation after Wesselingh and Ramos, 2010). water environments (e.g., lakes, lagoons, estuaries, salt marshes). It This work: Aquidabã, Pau D’Alho, Torre da Lua, Barro Branco. is able to cope with highly fluctuating salinities, aberrant water chemistries, variable temperatures and oxygenation. Cyprideis is Family Cytherideidae Sars, 1925 a bisexually reproducing taxon with internal brood-care, which Subfamily Cytherideinae Sars, 1925 makes it susceptible to withstand unstable conditions and to Genus Cyprideis Jones, 1857 passive dispersal (e.g., Sandberg, 1964; Jahn et al., 1996; De Deckker Cyprideis graciosa (Purper, 1979) et al., 1999; Meisch, 2000; Frenzel and Boomer, 2005; Gross et al., Pl. 4, figs. 1e17 2008; and references therein). 1977 Cytheridea sp. nov. D e Purper: 363, pl. 3, figs. 5e6. Occurrence: Western Amazonia, latest Middle to early Late * 1979 C. graciosa Purper, sp. nov. e Purper: 229e230, pl. 3, figs.1e9. Miocene (Colombia and Brazil: C. obliquosulcataeC. cyrtoma zone; 1991 C. graciosa Purper, 1979 e Purper and Ornellas: 26e28, pl. 1, Muñoz-Torres et al., 2006; chronostratigraphic correlation after figs. 10e15. Wesselingh and Ramos, 2010). Linhares et al. (2011) recorded ? 1998 C. graciosa (Purper, 1979) e Muñoz-Torres et al.: 96, pl. 3, C. graciosa also in the caraione zone. This work: Remanso, Morada figs. 1e3. Nova, Pau D’Alho, Torre da Lua, Barro Branco. ? 1998 C. graciosa (Purper, 1979) e Whatley et al.: 234, text-fig. 2; pl. 1, figs. 11e15. Cyprideis longispina (Purper, 1979) 2006 C. graciosa Purper, 1979 e Ramos: 92, figs. 7deh. Pl. 4, figs. 18e34 2009 C. graciosa Purper, 1979 e Ramos et al.: 114, fig. 289-I. * 1979 C. longispina Purper, sp. nov. e Purper: 230e231, pl. 3, figs. 2010 C. graciosa Purper, 1979 e Wesselingh and Ramos: 308, figs. 10e21. 18.5eef. 1998 C. longispina (Purper, 1979) e Muñoz-Torres et al.: 96e98, ?2011 C. graciosa e Linhares et al: 96, fig. 3/9e10. pl. 3, figs. 12e14. 1998 C. longispina (Purper, 1979) e Whatley et al.: 235, text-fig. 2; Material: 149 adults, 84 juveniles (250 mm sieve fraction). pl. 1, figs. 21e25. Dimensions:R\l ¼ 0.77e0.79 (0.78), h ¼ 0.42e0.44 (0.43; n ¼ 5); 2006 C. longispina (Purper, 1979) Whatley et al., 1998 e Ramos: L\l ¼ 0.81e0.83 (0.82), h ¼ 0.46e0.47 (0.46; n ¼ 6); R_l ¼ 0.82e 92, figs. 7ien. 0.84 (0.83), h ¼ 0.41e0.42 (0.41; n ¼ 5); L_l ¼ 0.86e0.88 (0.87), 2009 C. longispina (Purper, 1979) e Ramos et al.: 114, figs. 291e h ¼ 0.44e0.45 (0.45; n ¼ 5); RA-1 l ¼ 0.60e0.61, h ¼ 0.34e0.35 294-I. (n ¼ 2); LA-1 l ¼ 0.61e0.62 (0.61), h ¼ 0.34e0.35 (0.35; n ¼ 3). 2010 C. longispina (Purper, 1979) e Wesselingh and Ramos: 308, Remarks: The present material matches with the original figs. 18.5a, b. description of Purper (1979; Purper, 1977) as well as with the specimens documented by Purper and Ornellas (1991) and Material: 133 adults, 117 juveniles (250 mm sieve fraction). Wesselingh and Ramos (2010). C. graciosa of Muñoz-Torres et al. Dimensions:R\l ¼ 0.84e0.88 (0.86), h ¼ 0.43e0.45 (0.44; n ¼ 7); (1998; Whatley et al., 1998) is considerably larger. The posterior L\l ¼ 0.87e0.90 (0.88), h ¼ 0.46e0.48 (0.47; n ¼ 5); R_ l ¼ 0.85e0.89 margin of males in these publications is more rounded and females (0.87), h ¼ 0.40e0.41 (0.41; n ¼ 4); L_l ¼ 0.89e0.91 (0.89), h ¼ 0.42e are posterior higher. Posteroventral spines are not visible in left 0.44 (0.43; n ¼ 5); RA-1 l ¼ 0.61e0.65 (0.63), h ¼ 0.34e0.36 (0.35; valves of these specimens, which are a characteristic feature n ¼ 10); LA-1 l ¼ 0.63e0.68 (0.66), h ¼ 0.35e0.39 (0.36; n ¼ 5). according to Purper (1979). However, such spines were mentioned Remarks: The present material largely coincides with the given in the diagnosis of Whatley et al. (1998). To date, it is not clear if synonyms. Purper (1979) mentioned that males differ from females these differences range within the intraspecific variability of this by the presence of a set of four posteroventral spines in right valves. species. C. graciosa, discussed by Ramos (2006; Ramos et al., 2009), In our adult specimens, regularly we found one spine occurring is noticeably smaller. The figured carapace of a female (Ramos, equally in both sexes. Nevertheless, right juvenile valves with up to 2006, figs. 7feh) has a triangular shape and possibly represents three posteroventral spines were observed. C. longispina of Muñoz- a juvenile individual. Torres et al. (1998; compare Whatley et al., 1998) is somewhat C. graciosa is similar to C. longispina (pl. 4, figs. 18e34). None- larger and females are less elongated in outline in lateral view than theless, adults of C. longispina bear posteroventral spines (gener- the material from the Eirunepé region. The specimens of Ramos ally one spine) in the right valves only. The most ventrally located (2006) are slightly more elongated. Differences to C. graciosa are and strongest one of these spines originates higher up on the discussed above. flange and is more horizontally directed than in C. graciosa. Post- Ecology: see C. graciosa. eroventrally, C. longispina develops a significantly extended flange Occurrence: Western Amazonia, early to latest Middle Miocene in right valves, which is missing in C. graciosa. Anterior spines are (Peru: C. aulakoseC. caraione zone; Colombia and Brazil: restricted to the lower half of the anterior margin and the anterior C. obliquosulcata zone; Muñoz-Torres et al., 2006; chronostrati- hinge element is longer and narrower than in C. graciosa. Whereas graphic correlation after Wesselingh and Ramos, 2010). This work: in C. graciosa the anteroventral and central anterior surface of the Aquidabã, Morada Nova, Torre da Lua, Barro Branco. valve is usually smooth, in C. longispina the ornament is reduced in the anteroventral area only. In dorsal view the anterior portion is Cyprideis pebasae (Purper, 1979) slightly more tapered in C. graciosa. Pl. 5, figs. 1e17 Juveniles of C. graciosa and C. longispina can be distinguished by 1977 Cytheridea sp. nov. C e Purper: 363, pl. 3, figs. 1e4. the development of marginal spines, which is equivalent to adult * 1979 C. pebasae Purper, sp. nov. e Purper: 228e229, pl. 2, figs. valves (note: in juveniles of C. longispina up to three posteroventral 11e23. 226 M. Gross et al. / Journal of South American Earth Sciences 42 (2013) 216e241

Plate 5. (1e17) Cyprideis pebasae (1) Le\ (0.75/0.43; AQ5-2_26), (2) Le_ (0.78/0.40; AQ5-2_30), (3) LeA-1 (0.59/0.34; AQ5-2_32), (4) LiA-1 (¼ fig. 3), (5) Re\ (0.73/0.41; AQ5-2_37), (6) Re_ (0.78/0.37; AQ5-2_41), (7) ReA-1 (0.60/0.34; AQ5-2_44), (8) RiA-1 (¼ fig. 7), (9) Li\ (¼ fig. 1), (10) Ri\ (¼ fig. 5), (11) Li_ (¼ fig. 2), (12) central muscle scars of Li_ (0.82/0.42; MN14_10), (13) Ri_ (¼ fig. 6), (14) Ld\ (¼ fig. 1), (15) Rd\ (¼ fig. 5), (16) Ld_ (¼ fig. 2), (17) Rd_ (¼ fig. 6). (18e34) Cyprideis aff. pebasae (18) Le\ (0.80/0.44; TO12_29), (19) Le_ (0.81/ 0.42; TO10_09), (20) LeA-1 (0.63/0.35; TO12_30), (21) LiA-1 (¼ fig. 20), (22) Re\ (0.76/0.42; TO12_32), (23) Re_ (0.86/0.43; TO12_33), (24) ReA-1 (0.64/0.36; TO12_35), (25) RiA-1 (¼ fig. 24), (26) Li\ (¼ fig. 18), (27) Ri\ (¼ fig. 22), (28) Li_ (¼ fig. 19), (29) central muscle scars of Ri_ (0.77/0.39; MN9_11), (30) Ri_ (¼ fig. 23), (31) Ld\ (¼ fig. 18), (32) Rd\ (¼ fig. 22), (33) Ld_ (¼ fig. 19), (34) Rd_ (¼ fig. 29). (For abbreviations see plate 1). M. Gross et al. / Journal of South American Earth Sciences 42 (2013) 216e241 227

? 1980 Cyprideis purperi colombiaensis subsp. nov. e Sheppard and pars 2006 C. pebasae (Purper, 1979) Whatley et al., 1998 emend. e Bate: 101, pl. 8, figs. 3e9. Ramos: 90e91, figs. 6eeh. [non figs. 6iey] non 1998 C. pebasae (Purper, 1979) e Muñoz-Torres et al.: 100, pl. 2006 C. lacrimata Muñoz-Torres et al.,1998 e Ramos: 92, figs. 7aec. 4, figs. 8e10. 2009 C. lacrimata Muñoz-Torres et al., 1998 e Ramos et al.: 114, non 1998 C. pebasae (Purper, 1979) e Whatley et al.: 236, text-fig. fig. 290-I. 2; pl. 2, figs. 16e20. 2010 C. lacrimata Muñoz-Torres et al., 1998 e Wesselingh and ? 1998 Cyprideis lacrimata sp. nov. e Muñoz-Torres et al.: 96, text- Ramos: 308, fi gs. 18.5g, h. fig. 2; pl. 3, figs. 7e11. ?2011 Cyprideis sp. 4 e Linhares et al.: 97, fig. 4/12. pars 2006 C. pebasae (Purper, 1979) Whatley et al., 1998 emend. e Ramos: 90e91, figs. 6iey. [non figs. 6eeh] Material: 54 adults, 52 juveniles (250 mm sieve fraction). 2010 C. pebasae (Purper, 1979) e Wesselingh and Ramos: 308, Dimensions:R\l ¼ 0.72e0.76, h ¼ 0.38e0.42 (n ¼ 2); L\l ¼ 0.75e figs. 18.5c, d. 0.80 (0.78), h ¼ 0.40e0.44 (0.43; n ¼ 3); R_l ¼ 0.77e0.86 (0.81), h ¼ 0.39e0.44 (0.41; n ¼ 5); L_l ¼ 0.81e0.85, h ¼ 0.42e0.46 (n ¼ 2); Material: 521 adults, 260 juveniles (250 mm sieve fraction). RA-1 l ¼ 0.60e0.65 (0.62), h ¼ 0.34e0.37 (0.36; n ¼ 5); LA-1 Dimensions:R\l ¼ 0.66e0.74 (0.71), h ¼ 0.36e0.41 (0.38; l ¼ 0.61e0.65 (0.63), h ¼ 0.34e0.37 (0.35; n ¼ 3). n ¼ 11); L\l ¼ 0.71e0.80 (0.75), h ¼ 0.39e0.43 (0.41; n ¼ 7); Remarks: These valves are identical with C. lacrimata of Ramos R_l ¼ 0.75e0.79 (0.77), h ¼ 0.36e0.38 (0.37; n ¼ 8); L_l ¼ 0.74e (2006), Ramos et al. (2009), Wesselingh and Ramos (2010). Also 0.84 (0.79), h ¼ 0.38e0.44 (0.40; n ¼ 15); RA-1 l ¼ 0.57e0.60 the specimens of C. pebasae in Ramos (2006: figs. eeh) with a less (0.58), h ¼ 0.33e0.34 (0.34; n ¼ 5); LA-1 l ¼ 0.55e0.61 (0.59), “pronounced anterior and ventroposterior marginal border” and h ¼ 0.33e0.36 (0.34; n ¼ 12). numerous, small denticles at the anterior margin match with the Remarks: C. pebasae of Muñoz-Torres et al. (1998), compare present specimens. The one valve described by Purper (1979) as Whatley et al. (1998) is obviously closely related to the original Hulingsina? sp. is also very similar. C. pebasae in Muñoz-Torres et al. material of Purper (1977, 1979). However, those specimens exhibit (1998), Whatley et al. (1998), which differs particularly due its numerous, small denticles along the anterior margin, which is in infracurvate anterior margin and Cyprideis sp. 4 of Linhares et al. contrast to the species description of Purper (1979: “about seven (2011), which is more elongated, are possibly synonymous. spines”). Further, the flange along the anterior margin and at the Nonetheless, we consider C. lacrimata of Ramos (2006), posteroventral corner is thinner than in C. pebasae (Purper, 1979). Wesselingh and Ramos (2010) to be not equivalent to C. lacrimata of For these reasons we believe that C. pebasae of Muñoz-Torres et al. Muñoz-Torres et al. (1998) (see also Whatley et al., 1998 and (1998; see Whatley et al., 1998) is not identical with C. pebasae remarks to C. pebasae above). Although the current specimens (Purper, 1979). Consequently, only the specimens with a “well belong to the species group around C. pebasae, they cannot be pronounced anterior and ventroposterior marginal border” as well ascribed to C. pebasae of Purper (1979). as with a “coarsely spinose” flange of Ramos (2006: figs. 6iey) are Whether C. lacrimata Muñoz-Torres et al., 1998 is a junior supposed to belong to C. pebasae. synonym of C. pebasae (Purper, 1979) or not, a new species has to be C. lacrimata of Muñoz-Torres et al. (1998) is very similar to the established for the Cyprideis species under discussion here. Since present species but differs by its more rectangular outline (“trun- the aim of this paper is beyond describing a new species, it is left in cated posterior margin”) and its slightly coarser ornamentation open nomenclature and recorded under C. aff. pebasae at this time. along the free valve margin. Apparently, C. lacrimata of Muñoz- Ecology: see C. graciosa. Torres et al. (1998) and C. pebasae (Purper, 1979) are very closely Occurrence (of C. pebasae sensu Muñoz-Torres et al., 1998): related or are even synonyms. Western Amazonia, latest Middle to early Late Miocene (Colombia C. purperi colombiaensis Sheppard and Bate, 1980 resembles and Brazil: C. obliquosulcataeC. cyrtoma zone; Muñoz-Torres et al., C. pebasae from the Eirunepé region. It is slightly larger, the median 2006; chronostratigraphic correlation after Wesselingh and Ramos, sulcus is more pronounced, its ornamentation is faintly finer and 2010). This work (C. aff. pebasae): Aquidabã, Morada Nova, Pau the anterior rim is less well expressed. However, based on the given D’Alho, Torre da Lua, Barro Branco. documentation an assured synonymisation is not possible to date (compare also Swain, 1998). Cyprideis aff. machadoi (Purper, 1979) Ecology: see C. graciosa. Pl. 6, figs. 1e20, 22 Occurrence (of C. lacrimata Muñoz-Torres et al., 1998): Western non 1979 Chlamydocytheridea machadoi Purper, gen. et sp. nov. e Amazonia, (early) late Middle to early Late Miocene (Peru: C. caraione Purper: 237e238, pl. 6, figs. 1e6. zone; Colombia and Brazil: C. obliquosulcataeC. cyrtoma zone; ?1979 Cyprideis truncata Purper, sp. nov. e Purper: 232e233, pl. 4, Muñoz-Torres et al., 2006; chronostratigraphic correlation after figs. 12e22. Wesselingh and Ramos, 2010). This work (C. pebasae): Remanso, 2010 Cyprideis machadoi (Purper, 1979) e Wesselingh and Aquidabã, Morada Nova, Pau D’Alho, Torre da Lua, Barro Branco. Ramos: 308, figs. 18.5m, n. 2011 Cyprideis machadoi e Linhares et al.: 96, fig. 3/13e14. Cyprideis aff. pebasae (Purper, 1979) Pl. 5, figs. 18e34 Material: 51 adults, 4 juveniles (250 mm sieve fraction). non 1979 Cytheridea pebasae Purper, sp. nov. e Purper: 228e229, Dimensions: R\l ¼ 0.85e0.94, h ¼ 0.44e0.49 (n ¼ 2); pl. 2, figs. 11e23. L\l ¼ 0.85e0.93 (0.89), h ¼ 0.46e0.51 (0.49; n ¼ 4); R_l ¼ 0.89e 1979 Hulingsina sp. e Purper: 239e240, pl. 7, figs. 1e5. 1.01 (0.95), h ¼ 0.44e0.48 (0.46; n ¼ 4); L_l ¼ 0.91e1.03, non 1998 C. lacrimata sp. nov. e Muñoz-Torres et al.: 96, text-fig. 2; h ¼ 0.47e0.52 (n ¼ 2); RA-1 l ¼ 0.69, h ¼ 0.37 (n ¼ 1); LA-1 pl. 3, figs. 7e11. l ¼ 0.65, h ¼ 0.37 (n ¼ 1). ? 1998 C. pebasae (Purper, 1979) e Muñoz-Torres et al.: 100, pl. 4, Remarks: Due to the well developed hinge and inner lamella as figs. 8e10. well as a pronounced sexual dimorphism, these specimens are ?1998 C. pebasae (Purper,1979) e Whatley et al.: 236, text-fig. 2; pl. regarded to represent adults, which show a considerable variation 2, figs. 16e20. in size (e.g., pl. 6, figs. 1 and 3, figs. 6 and 7). 228 M. Gross et al. / Journal of South American Earth Sciences 42 (2013) 216e241

Plate 6. (1e20) Cyprideis aff. machadoi (1) Le\ (0.92/0.51; MN14_12), (2) Le_ (1.03/0.52; MN14_11), (3) Le\ (0.86/0.46; MN9_28), (4) Li\ (¼ fig. 3), (5) Re\ (0.94/0.49; MN14_14), (6) Re_ (1.01/0.48; MN14_13), (7) Re_ (0.89/0.44; MN9_33), (8) Ri_ (0.96/0.46; MN9_31), (9) Li\ (¼ fig. 1), (10) Ri\ (¼ fig. 5), (11) Li_ (¼ fig. 2), (12) Ri_ (¼ fig. 6), (13) LeA-1 (0.65/0.37; PD20_01), (14) ReA-1 (0.69/0.37; MN14_17), (15) LiA-1 (¼ fig. 13), (16) RiA-1 (¼ fig. 14), (17) Ld_ (¼ fig. 2), (18) Rd_ (¼ fig. 6), (19) Ld\ (¼ fig. 1), (20) Rd\ (¼ fig. 5). (21) Cyprideis ? olivencai,Ld\ (0.84/0.43; PD20_03). (22) Cyprideis aff. machadoi, central muscle scars of Ri\ (¼ fig. 10). (23e27) Cyprideis ?olivencai (23) Le\ (¼ fig. 21), (24) Li\ (¼ fig. 23), (25) ReA-1 (0.63/0.33; PD25_01), (26) RiA-1 (¼ fig. 25), (27) ReA-2 (0.51/0.28; PD25_03). (For abbreviations see plate 1).

The present species clearly belongs to the “smooth lineage” of outline, especially due to the development of a very wide area Amazonian Cyprideis (Whatley et al., 1998, p. 237). But it seems not between the flange and the selvage anteriorly (Purper, 1979; to be identical with the similar C. machadoi, which is much larger, Muñoz-Torres et al., 1998; Whatley et al., 1998: pl. 2, figs. 6e10 has stronger terminal hinge elements and differs noticeably in [sic!]). M. Gross et al. / Journal of South American Earth Sciences 42 (2013) 216e241 229

However, Muñoz-Torres et al. (1998), Whatley et al. (1998) Brazil: C. obliquosulcataeC. cyrtoma zone; Muñoz-Torres et al., 2006; consider C. machadoi as a variable species in relation to the shape chronostratigraphic correlation after Wesselingh and Ramos, 2010). and development of the anterior margin and its flange. Those This work (C.?olivencai): Morada Nova, Pau D’Alho, Torre da Lua. authors placed Paulacoutoia kroemmelbeini Purper, 1979, Otar- ocyprideis elegans Sheppard & Bate, 1980 and Chlamydocytheridea Cyprideis sp. 1 kotzianae Purper & Ornellas, 1991 in the synonymy of C. machadoi. Pl. 7, figs. 1e5 Further, they suggest the “scattered punctate” Cyprideis truncata Purper, 1979 to be a juvenile of C. machadoi. Material: 3 adults (250 mm sieve fraction). The present material of adult valves largely matches with Dimensions:R\l ¼ 0.74, h ¼ 0.40; R?_l ¼ 0.66, h ¼ 0.33; L?_ C. truncata aside the punctate surface. Slight punctae are observed l ¼ 0.69, h ¼ 0.35. on all valves, which correspond to normal pore canals. On somewhat Remarks: These specimens are related to C. longispina but seem corroded specimens these depressions become more prominent and not to range within the variability of that species. Cyprideis sp. 1 is look like punctate (pl. 6, fig. 7). Nevertheless, further investigations noticeably smaller, has a coarse reticulate ornamentation and the of the type material are needed to clarify the status of C. truncata. flange is less pronounced posteroventrally. Because only three Here, we assign the present specimens to C. aff. machadoi. adult valves are available to date, no further species identification is Ecology: see C. graciosa. attempted. Occurrence (of C. machadoi): Western Amazonia, early Middle to Ecology: see C. graciosa. Occurrence (this work): Barro Branco. early Late Miocene (Peru: C. aulakoseC. minipunctata zone; Colombia and Brazil: C. obliquosulcataeC. cyrtoma zone; Muñoz-Torres et al., Cyprideis sp. 2 2006; chronostratigraphic correlation after Wesselingh and Ramos, Pl. 7, figs. 6e15, 38 2010). This work (C. aff. machadoi): Morada Nova, Pau D’Alho.

Cyprideis ?olivencai (Purper, 1979) Material: 17 adults, 3 juveniles (250 mm sieve fraction). Pl. 6, figs. 21, 23e27 Dimensions:R\l ¼ 0.64e0.68, h ¼ 0.33e0.35 (n ¼ 2); ?1979Paulacoutoia olivençai Purper, gen. et sp. nov. e Purper: R_l ¼ 0.64e0.69 (0.67), h ¼ 0.32e0.33 (0.33; n ¼ 4); L_l ¼ 0.62e 235e236, pl. 5, figs. 10e17. 0.67 (0.65), h ¼ 0.30e0.32 (0.31; n ¼ 3); RA-1 l ¼ 0.55e0.56 ? 1998 C. olivencai (Purper, 1979) e Muñoz-Torres et al.: 100, pl. 4, (0.55), h ¼ 0.30e0.31 (0.30; n ¼ 3). figs. 5e7. Remarks: Cyprideis sp. 2 is close to Pseudoparakrithella paralela ? 1998 C. olivencai (Purper, 1979) e Whatley et al.: 236, text-fig. 2; Purper, 1979 (compare also Purper and Pinto, 1983). Size, outline in pl. 2; figs. 1e5 [sic!]. lateral and dorsal view, development of the inner lamella as well as 2010 C. olivencai (Purper, 1979) e Wesselingh and Ramos: 308, the central muscle scars are equivalent. Both, Cyprideis sp. 2 and fig. 18.5o, p. P. paralela, exhibit an inverse hinge (and inverse valve size) with ?2011 C. olivencai e Linhares et al.: 97, fig. 4/1e2. crenulated anterior and posterior bars and a short median groove in left valves. Material: 6 adults, 3 juveniles (250 mm sieve fraction). However, the specimens from the Eirunepé area are stronger Dimensions:L\l ¼ 0.84e0.87, h ¼ 0.43e0.47 (n ¼ 2); R?_l ¼ 0.85, ornamented (coarsely punctate to reticulate, whereas P. paralela is h ¼ 0,39 (n ¼ 1); RA-1 l ¼ 0.63e0.64, h ¼ 0.32e0.35 (n ¼ 2); RA-2 punctate), marginal pore canals are either simple or branched (in l ¼ 0.51, h ¼ 0.28 (n ¼ 1). P. paralela they are simple) and two to three, very inconspicuous Remarks: Only few specimens were found which resemble in posteroventral spines on some left valves are observed, which are lateral view C. olivencai of Muñoz-Torres et al. (1998), Whatley et al. not mentioned by Purper (1979). (1998), Linhares et al. (2011) and, especially, of Wesselingh and Pseudoparakrithella Purper, 1979 is considered by Whatley et al. Ramos (2010). For the later no inner view is figured, which (1998); see also Muñoz-Torres et al. (1998) to be a synonym of the hampers further comparisons. Because the valves of Wesselingh genus Cyprideis (here we follow that assumption). Moreover, these and Ramos (2010) originate from Morada Nova, probably they are authors place P. paralela into the synonymy of the “very variable synonymous with our material. species” C. olivencai (Purper, 1979)(Whatley et al., 1998, p. 236). The specimens of Muñoz-Torres et al. (1998) and Whatley et al. Neither the reversed hinge nor differences in ornamentation (oli- (1998) diverge due to the posteroventrally much wider inner lamella vencai was originally described to be punctate; in the diagnosis and in having only a narrow anterior vestibulum (the material of Eir- given by Whatley et al. (1998) it is characterised to be smooth), the unepé displays a wide anterior and a narrow posterior vestibulum). development of the inner lamella (fused zone narrower in paralela The most striking difference is, however, the dentate anterior and of Purper (1979) than in olivencai of Whatley et al. (1998)) and posterior hinge elements, which are smooth in here studied valves. differences in marginal pore canals (simple in paralela; simple, Although C. olivencai seems to be a rather variable taxon (e.g., bifurcating or polyfurcating in olivencai sensu Whatley et al., 1998) valve size, width of the inner lamella, development of vestibuli, are supposed to justify a species delineation. ornamentation) by considering the given synonyms in Muñoz- Based on the scarce material, we cannot attest if Cyprideis sp. 2 Torres et al. (1998), Whatley et al. (1998), compare Purper (1979), belongs to C. paralela (Purper, 1979) or represents a new Cyprideis- Sheppard and Bate (1980), Purper and Pinto (1983), an assured species. Nevertheless, we consider Cyprideis sp. 2 not to be identification is not possible based on the available material. synonymous with C. olivencai sensu Whatley et al. (1998), Muñoz- A few, smooth-shelled, juvenile valves resemble in outline Torres et al. (1998). adults of C. olivencai sensu Muñoz-Torres et al. (1998) and Whatley Ecology: see C. graciosa. Occurrence (this work): Aquidabã, et al. (1998). They can be distinguished due to their less inclined Morada Nova, Torre da Lua, Barro Branco. dorsal margin and the narrower anterior inner lamella from juve- niles of C. aff. machadoi and are allocated to C.?olivencai here. Family Cytheridae Baird, 1850 Ecology: see C. graciosa. Genus Perissocytheridea Stephenson, 1938 Occurrence (of C. olivencai): Western Amazonia, early Middle to Perissocytheridea acuminata (Purper, 1979) early Late Miocene (Peru: C. aulakoseC. caraione zone; Colombia and Pl. 7, figs. 16e24 230 M. Gross et al. / Journal of South American Earth Sciences 42 (2013) 216e241

Plate 7. (1e5) Cyprideis sp. 1 (1) Le?_ (0.69/0.35; BA7b_03), (2) Re?_ (0.66/0.33; BA7b_18), (3) Li?_ (¼ fig. 1), (4) Ri?_ (¼ fig. 2), (5) central muscle scars of Li?_ (¼ fig. 3). (6e15) Cyprideis sp. 2 (6) Ld_ (0.66/0.30; TO6_01), (7) Rd\ (0.64/0.33; BA7a_02), (8) Le_ (¼ fig. 6), (9) Re\ (¼ fig. 7), (10) Li_ (¼ fig. 6), (11) Ri\ (¼ fig. 7), (12) ReA-1 (0.56/0.31; BA7b_26), (13) RiA-1 (¼ fig. 12), (14) Re_ (0.68/0.33; MN9_51), (15) Ri_ (¼ fig. 14). (16e24) Perissocytheridea acuminata (16) Le (0.48/0.24; PD20_04), (17) Re (0.47/0.22; PD20_05), (18) ReA-1 (0.39/0.20; BA7b_34), (19) RiA-1 (¼ fig. 18), (20) Li (¼ fig. 16), (21) Ri (¼ fig. 17), (22) Re?A-2 (0.31/0.18; MN8b_01), (23) Ri?A-2 (¼ fig. 22), (24) central muscle scars of Ri (¼ fig. 21). (25e27) Perissocytheridea sp. (25) Le (0.31/0.15; PD20_06), (26) Li (¼ fig. 25), (27) Ld (¼ fig. 25). (28e34) Rhadinocytherura amazonensis (28) Le (0.33/0.18; BA7a_06), (29) Re (0.33/ 0.18; BA7a_07), (30) Li (¼ fi g. 28), (31) Ri (0.32/0.18; BA7a_08), (32) Ld (0.32/0.18; BA7a_03), (33) Rd (0.33/0.17; BA7a_10), (34) detail of Ri (anterodorsal area with sieve pores; 0.31/ 0.17; BA7a_04). (35e37) Rhadinocytherura sp. (35) central muscle scars of Ri (0.32/0.17; AQ6-2_27), (36) Re (¼ fig. 35), (37) Ri (¼ fig. 35). (38) Cyprideis sp. 2, central muscle scars of Ri_ (¼ fig. 14). (For abbreviations see plate 1). M. Gross et al. / Journal of South American Earth Sciences 42 (2013) 216e241 231

1977 Ostracoda C n.g.,n.sp. e Purper: 365, pl. 4, figs. 15e16. complementary; Sheppard and Bate, 1980) largely coincide with * 1979 P. acuminata Purper gen et sp.nov. e Purper: 241e242, pl. 7, the given synonyms. figs. 15e20. Muñoz-Torres et al. (1998) noticed variations in ornamentation 1998 P. acuminata (Purper, 1979) e Muñoz-Torres et al.: 102, pl. 5, of the valves. Most specimens from the Eirunepé area bear a post- figs. 5e8. eroventral spine in both valves, which is not reported by Sheppard and Bate (1980) and Muñoz-Torres et al. (1998). Here, we consider Material: 2 adults, 2 juveniles (125 mm sieve fraction). that feature to range within the variability of the species. Dimensions:R\l ¼ 0.47, h ¼ 0.22 (n ¼ 1); L\l ¼ 0.48, h ¼ 0.24 Originally, normal pores are mentioned to be simple by (n ¼ 1); RA-1 l ¼ 0.39, h ¼ 0.20 (n ¼ 1); ?RA-2: l ¼ 0.31, h ¼ 0.18 (n ¼ 1). Sheppard and Bate (1980). In contrast, large sieve-plates are Remarks: In lateral view, the rare specimens from Eirunepé are evident in the current material. Nevertheless, we believe our slightly less acuminate and display a small convexity at the centre specimens to be conspecific with R. amazonensis. of the dorsal margin. Nonetheless, they match well with the Ecology: Sheppard and Bate (1980) consider R. amazonensis as documentation given by Purper (1977, 1979) and Muñoz-Torres a marine to brackish water taxon. et al. (1998). Occurrence: Western Amazonia, early Middle to early Late Ecology: Perissocytheridea is considered a euryhaline (mainly Miocene (Peru: C. caraione zone; Colombia and Brazil: mesohaline) form, typical for lagoonal environments (e.g., Keyser, C. obliquosulcataeC. cyrtoma zone; Muñoz-Torres et al., 2006; 1977; Colin et al., 1996; Muñoz-Torres et al., 2006; Mebrouk et al., chronostratigraphic correlation after Wesselingh and Ramos, 2010). 2011). This work: Aquidabã, Morada Nova, Pau D’Alho, Torre da Lua, Barro Occurrence: Western Amazonia, early Middle to early Late Branco. Miocene (Peru: C. aulakoseC. caraione zone; Colombia and Brazil: C. obliquosulcataeC. cyrtoma zone; Muñoz-Torres et al., 2006; Rhadinocytherura sp. chronostratigraphic correlation after Wesselingh and Ramos, 2010). Pl. 7, figs. 35e37 This work: Morada Nova, Pau D’Alho, Barro Branco. Material: 4 adult valves (125 mm sieve fraction). Perissocytheridea sp. Dimensions:Rl¼ 0.32, h ¼ 0.17 (n ¼ 2); L l ¼ 0.30, h ¼ 0.17 Pl. 7, figs. 25e27 (n ¼ 1). ? 1998 Perissocytheridea sp. 1 e Muñoz-Torres et al.: 104, pl. 6, figs. Remarks: Four valves were found, which resemble 1e6. R. amazonensis (Sheppard and Bate, 1980; Muñoz-Torres et al., 1998; and above). However, these specimens differ significantly Material: 1 adult valve (125 mm sieve fraction). due to their entirely straight dorsal margin, the caudal process is Dimensions:Ll¼ 0.31, h ¼ 0.15 (n ¼ 1). situated at the half of the valves’ height, the posterior tooth in the Remarks: One left, due to its well developed hinge and inner left valve is more robust and the ornament is regularly pitted. lamella clearly adult valve is available, which we place in Peri- The hinge is developed as characteristic for the genus. Again, ssocytheridea (compare generic diagnosis emended by Pinto and normal pores of sieve type are observed (compare remarks to R. Ornellas (1970)). amazonensis above). This specimen comes close to Perissocytheridea sp. 1 of Muñoz- Ecology: see R. amazonensis. Occurrence (this work): Aquidabã, Torres et al. (1998), which is: somewhat larger; its posterior margin Torre da Lua. is more rounded; it has a better expressed bifurcate sulcus that borders a node (below an additional small node is developed); and, 4.2. Ostracod occurrence and geochemical results its inner lamella is much narrower. Because our material consists of only one specimen, it is left in open nomenclature. Thirty-three bulk samples delivered a moderately diverse fauna Ecology: see P. acuminata. with 19 species, belonging to 10 genera and 8 families. Four Occurrence (of Perissocytheridea sp. 1 of Muñoz-Torres et al. samples were barren of ostracods (AQ6, AQ15/2, MN1, MN2). (1998)): Western Amazonia, latest Middle Miocene (Colombia Overall, the fauna is dominated by the limnocytherid and Brazil: C. obliquosulcata zone; Muñoz-Torres et al., 2006; Cytheridella danielopoli (41.6%), the cytherideid Cyprideis (36.6%; 8 chronostratigraphic correlation after Wesselingh and Ramos, 2010). species) and the darwinulid Penthesilenula olivencae (19.6%; Fig. 3; This work: Pau D’Alho. ESM 1). The other 9 taxa just account with 2.2% to the total ostracod assemblage. Among Cyprideis the “ornate” group (Whatley et al., Family Cytheruridae Müller, 1894 1998) predominates with w35% (C. pebasae w15%; C. longispina Genus Rhadinocytherura Sheppard & Bate, 1980 w4.8%; C. graciosa w4.5%; C. aff. pebasae w2%), whereas the Rhadinocytherura amazonensis Sheppard & Bate, 1980 “smooth” group (C. aff. machadoi, C.?olivencai) yields w1.6% only Pl. 7, figs. 28e34 (ESM 1). All isotopic analyses provided very light values with * 1980 R. amazonensis sp. nov. e Sheppard and Bate: 112e113, pl. a range for d18O from 6.62 to 9.89& and for d13C with 8.36 to 11, figs. 10e16. 13.63& (Fig. 4; Table 1). 1998 R. amazonensis Sheppard & Bate, 1980 e Muñoz-Torres et al.: 104, pl. 6, figs. 7e12. 4.2.1. Remanso Location. 27.4 km NE Eirunepé (S 0631022.000/W 06935042.800; Material: 18 adults (including one fragmented carapace), 1 altitude: w105 m; section thickness: w15 m), left cutbank of the juvenile, 3 fragments (125 mm sieve fraction). Juruá River (Figs. 1b, 5). Dimensions:Rl¼ 0.31e0.35 (0.33), h ¼ 0.17e0.20 (0.18; n ¼ 8); L l ¼ 0.31e0.33 (0.32), h ¼ 0.17e0.19 (0.18; n ¼ 6); L?A-1 l ¼ 0.27, Sedimentological interpretation (Gross et al., 2011). Overbank h ¼ 0.15 (n ¼ 1). deposits at the base of the section are cut by a channel, which Remarks: Size, shape as well as the development of the inner became filled by pelitic, afterwards pedogenically overprinted lamella and the characteristic hinge (left valve: dentate anterior sediments (sample RE4). It is overlain by crevasse splay deposits, tooth, smooth median bar, loculate posterior socket; right valve sandy bedforms as well as possible bar-top assemblages. 232 .Gose l ora fSuhAeia at cecs4 21)216 (2013) 42 Sciences Earth American South of Journal / al. et Gross M. e 241

Fig. 3. Distribution of ostracods in the samples (warranged in stratigraphical order; palaeoenvironmental interpretation after Gross et al., 2011;note:Cyprideis juv. “ornate” and Cyprideis juv. “smooth” refer to specimens from the 125 mm sieve-residual; AQ ¼ Aquidabã, BA ¼ Barro Branco, MN ¼ Morada Nova, PD ¼ Pau D’Alho, RE ¼ Remanso, TO ¼ Torre da Lua). M. Gross et al. / Journal of South American Earth Sciences 42 (2013) 216e241 233

Fig. 4. d18O- and d13C-isotopic ratios of ostracod valves from the Eirunepé area (freshwater taxa: Cytheridella, Penthesilenula; “brackish/marine” taxa: Cyprideis, Rhadinocytherura; note: Cytheridella ilosvayi ¼ recent material from two oxbow lakes: Lago Barro Branco: S 0650018.300/W 06945037.000, Lago Comprido: S 0643052.700/W 06944033.900; Fig. 1b). The indicated range for modern rivers and floodplain lakes is based on aragonitic mollusc shells (after Wesselingh et al., 2006c), which give somewhat heavier values for the same environmental parameters compared to ostracod calcite (Grossman and Ku, 1986).

Ostracods. Sample RE4 delivered a very poor fauna with a few Sedimentological interpretation (Gross et al., 2011). The basal layers juveniles of C. graciosa, C. pebasae and some juveniles of the are referred to crevasse splay and/or crevasse channel sediments, “ornate” Cyprideis group. introduced into a floodplain lake or, alternatively, are channel deposits of an abandoned avulsive river arm (samples AQ5, 6, 3/ 4.2.2. Aquidabã 2). Subordinately, a small channel fill (floodplain pond) is inter- Location. 22.0 km NE Eirunepé (S 0631040.800/W 06939052.000; calated (AQ5/2, 6/2). Up-section, the influx of crevassing altitude: w106 m; section thickness: w8 m), left cutbank of the respectively the active channel decreased and led to the forma- Juruá River (Figs. 1b, 5). tion of a successively drying up floodplain lake within a densely

Table 1 Isotope data of Eirunepé ostracod valves (juv. ¼ juvenile, ad. ¼ adult; no.v. ¼ number of valves used for analysis; B. Branco, Comprido ¼ Lago Barro Branco, Lago Comprido, both recent oxbow lakes; for species abbreviations see Fig. 4).

Sample no. Species no. v. d13C d18O Sample no. Species no. v. d13C d18O BA7b C. pebasae juv. 3 10.59 7.11 AQ19 C. danielopoli juv. 4 13.04 9.15 BA7b C. pebasae juv. 3 9.87 7.49 AQ19 C. danielopoli ad. 2 8.36 6.68 BA7b C. pebasae juv. 3 10.87 7.94 AQ6/2 C. pebasae juv. 4 11.59 9.01 BA7b C. pebasae ad. 2 10.08 7.17 AQ6/2 C. pebasae juv. 4 11.56 8.24 BA7b C. pebasae ad. 2 10.97 7.62 AQ6/2 C. pebasae ad. 2 12.49 8.38 BA7b R. amazonensis ad. 7 12.50 9.38 AQ6/2 C. pebasae ad. 2 13.09 9.66 BA7a C. pebasae juv. 4 9.98 7.07 AQ6/2 C. pebasae ad. 2 12.10 8.72 BA7a C. pebasae ad. 2 9.69 6.62 AQ6/2 C. pebasae ad. 2 13.27 9.82 BA7a C. pebasae ad. 2 10.33 6.94 AQ6/2 C. pebasae ad. 2 11.95 8.14 BA7a C. pebasae ad. 2 9.96 6.86 AQ6/2 C. danielopoli juv. 3 11.81 9.29 TO12 C. pebasae juv. 3 11.04 7.03 AQ6/2 C. danielopoli juv. 3 12.01 9.66 TO12 C. pebasae ad. 2 11.69 7.12 AQ6/2 C. danielopoli juv. 3 11.54 9.09 TO12 C. pebasae ad. 2 11.88 8.09 AQ6/2 C. danielopoli ad. 2 9.43 7.69 TO12 C. pebasae ad. 2 13.00 8.73 AQ6/2 P. olivencae ad. 6 12.95 9.55 TO12 C. pebasae ad. 2 11.38 8.38 AQ6/2 P. olivencae ad. 6 12.81 9.89 TO12 C. graciosa ad. 4 12.26 7.49 AQ5/2 C. pebasae ad. 2 12.36 8.70 PD20 C. graciosa ad. 3 13.63 9.31 AQ5/2 C. pebasae ad. 2 13.03 8.66 MN14 C. pebasae ad. 2 10.16 6.72 AQ5/2 C. pebasae ad. 2 11.83 8.41 MN14 C. pebasae ad. 3 10.95 8.78 AQ5/2 C. pebasae ad. 2 12.83 9.70 MN14 C. pebasae ad. 3 9.82 7.34 AQ5/2 C. danielopoli juv. 3 11.60 8.79 MN9 C. pebasae ad. 2 9.67 8.52 AQ5/2 C. danielopoli juv. 3 12.13 8.92 MN9 C. pebasae ad. 3 9.68 7.45 AQ5/2 C. danielopoli juv. 3 12.64 8.95 MN9 C. pebasae ad. 3 9.34 7.34 AQ5/2 C. danielopoli ad. 2 10.66 9.06 MN9 C. graciosa ad. 3 9.74 7.06 B. Branco C. ilosvayi 1 13.69 10.73 AQ19 C. danielopoli juv. 4 12.29 8.84 Comprido C. ilosvayi 1 11.34 7.84 234 M. Gross et al. / Journal of South American Earth Sciences 42 (2013) 216e241

Fig. 5. Lithological sections of (a) Remanso, (b) Aquidabã and (c) Morada Nova as well as interpretation of depositional environment (after Gross et al., 2011, where more detailed information is presented; CS ¼ crevasse splay, FF (CH) ¼ abandoned channel, FPL ¼ floodplain lake, OF ¼ overbank fines; legend also used in Fig. 6). vegetated backswamp (AQ8/2, 13/2, 16, 19, 15/2, 22a, 22b). P. olivencae; other taxa are lacking). In contrast, rich and diverse Sporadically, it was influenced by flash floods. Finally, a swampy faunas were found in the samples in between (AQ5/2eAQ19). In the environment established (AQ16/2). three most productive samples C. danielopoli is the dominant species (AQ5/2, 6/2, 19). Remarkably, Cyprideis spp. is completely absent in Ostracods. Samples from Aquidabã yielded a rich ostracod fauna AQ19. (2403 determined specimens; 12 taxa). In total, C. danielopoli (64.3%) dominates the assemblage. Cyprideis accounts with 19.6% (4 Geochemistry. Juveniles and adults of C. pebasae (AQ5/2, 6/2) range species; only of the “ornate” group) and Pelocypris olivencae with within 8.14 to 9.82& in respect to d18O(d13C: 11.56 to 13.6% to the fauna. Pseudocandona sp. and Pelocypris zilchi are 13.27&). Similar are the results for adults of P. olivencae (AQ6/2; exclusive for Aquidabã. d18O: 9.55 to 9.89&; d13C: 12.81 to 12.95&) and for juveniles Samples from the base of the outcrop (AQ5, 6, 3/2) and from of C. danielopoli (AQ5/2, 6/2,19; d18O: 8.79 to 9.66&; d13C: 11.54 the upper part (AQ15/2, 22a, 22b AQ16/2) yielded rather poor to 13.04&). Only adults of C. danielopoli (AQ19) yielded somewhat faunas (predominantly Cyprideis spp., only rarely C. danielopoli and heavier values (d18O: 6.68 to 9.06&; d13C: 8.36 to 10.66&). M. Gross et al. / Journal of South American Earth Sciences 42 (2013) 216e241 235

4.2.3. Morada Nova Samples MN1 and MN2 were barren of ostracods; MN8a and MN8b Location. 17.9 km NE Eirunepé (S 0632051.100/W 06942039.400; contained only a few ostracod remains. In samples MN9 and MN14 altitude: w107 m; section thickness: w8 m), left cutbank of the a moderately rich fauna was found, which is highly dominated by Juruá River (Figs. 1b, 5). Cyprideis spp.

Sedimentological interpretation (Gross et al., 2011). The lower part of Geochemistry. Values obtained from adults of C. pebasae (MN9, 14) the outcrop is formed by a succession of paleosols (samples MN1, range between 6.72 and 8.78& for d18O and 9.34 to 10.95& 2). The paleosols are overlain by point bar deposits (only present in for d13C; the result for C. graciosa adult (MN9) is comparable (d18O: Sections 2e4 as shown in Gross et al., 2011) in which a subordinate 7.06&; d13C ¼9.74&). channel (?chute channel fill) is intersected in the NW (MN8a, 8b). Up-section, the influx of the active channel decreased and a short- 4.2.4. Pau D’Alho lived floodplain lake developed within a richly vegetated back- Location. 12.6 km NE Eirunepé (S 0633055.500W06946011.7 00; swamp (MN9, 14). altitude: w108 m; section thickness: w10 m), left cutbank of the Juruá River (Figs. 1b, 6). Ostracods. The small ostracod fauna of Morada Nova (306 deter- mined specimens; 11 taxa) comprises 94.8% Cyprideis spp. (7 Sedimentological interpretation (Gross et al., 2011). The section species). P. olivencae is present with 3.9%, C. danielopoli is lacking. starts with fluvial overbank deposits, probably related to levee and/

Fig. 6. Lithological sections of (a) Pau D’Alho, (b) Torre da Lua and (c) Barro Branco as well as interpretation of depositional environment (after Gross et al., 2011; where more detailed information is presented; CR ¼ crevasse channel). 236 M. Gross et al. / Journal of South American Earth Sciences 42 (2013) 216e241 or crevasse splay sedimentation with possibly intercalated short- 5. Discussion lived ponds (sample PD2). The layers up-section (samples PD18, 20, 23, 25) are assigned to the formation of a shallow floodplain 5.1. Palaeoecological indication of the Eirunepé ostracod fauna pond, which was episodically influenced by crevasse splays. Above, floodplain pond conditions ended due to enhanced clastic input 5.1.1. Freshwater versus brackish or marine waters from an approaching channel, which finally intersected the One of the most controversially discussed issues in western succession and became filled up by point bar deposits. Amazonia’s history is the influenceofmarineincursionsduring Neogene times. Their existence, chronology, origin as well as Ostracods. The limited ostracod fauna of Pau D’Alho (309 deter- their spatial extent is still disputed (e.g., Hoorn et al., 2010a,b; mined specimens; 11 taxa) is largely dominated by Cyprideis spp. Hovikoski et al., 2010; Latrubesse et al., 2010; Ruskin et al., 2011). (80.3%; 5 species), followed by the darwinulid P. olivencae (12.6%). Aside sedimentological and ichnological indications (e.g., C. danielopoli is virtually absent (Fig. 3). Gingras et al., 2002; Hovikoski et al., 2007, 2010), palae- ontological evidences (i.e., mangrove pollen, foraminifers, Geochemistry. The isotopic analysis of four adults of C. graciosa specific molluscs, barnacles) were used to infer transitorily (PD20) resulted in light d18O(9.31&) and d13C(13.63&) values. marine influences (e.g., Hoorn, 1993, 2006; Wesselingh et al., 2002; Vonhof et al., 2003; Linhares et al., 2011). Additionally, 4.2.5. Torre da Lua the occurrence of brackish water ostracods (particularly Cypri- Location. 17.5 km SE Eirunepé (S 064902300/W 0694700400; alti- deis) and some rarely recorded marine species, motivated several tude: w117 m; section thickness: w12 m), left cutbank of the Tar- authors to propose brackish (mixohaline) waters (Purper, 1979; auacá River (Figs. 1b, 6). Purper and Pinto, 1983, 1985; Purper and Ornellas, 1991; Whatley et al., 1998) or marine transgressions (Sheppard and Bate, 1980; Sedimentological interpretation (Gross et al., 2011). The basal layers Swain, 1998). are interpreted to be deposited in a floodplain lake, which was In total the ostracod fauna of the Eirunepé region comprise influenced by crevasse splays (with increasing tendency up- about 62.8% freshwater ostracods (Penthesilenula, Pseudo- section; samples TO1, 6). In layer 10 (sample TO10) a crevasse candona, Cypria, Physocypria, Pelocypris, Cypretta, Cytheridella). channel reached the floodplain lake via a crevasse delta. Up- Cyprideis (36.6%) is associated with them, accompanied by Per- section, the crevasse channel became abandoned and floodplain issocytheridea and Rhadinocytherura (0.6%). Cyprideis and Peri- lake conditions were reinstalled (TO12), which were, however, ssocytheridea are euryhaline taxa, today typically occurring in frequently influenced by crevasse splays during floods (TO20, 23). marginal marine settings (for palaeoecology of the taxa see 4.1. .) Rhadinocytherura is considered as a marine to brackish water Ostracods. The rich ostracod fauna of Torre da Lua (1635 deter- form (Sheppard and Bate, 1980); however, it is endemic for mined specimens; 12 taxa) is dominated by C. danielopoli (37.8%), western Amazonia. followed by P. olivencae (36.0%) and Cyprideis spp. (25.6%; 6 In our samples fresh- and brackish water elements co-occur species). Whereas the sample from the base of the section (TO1) (Fig. 3). Usually various ontogenetic stages of the taxa were found contained a moderately abundant and diverse fauna, TO6 yielded and valves are well preserved (translucent, no signs of corrosion), only Cyprideis spp. Up-section, samples were rich in ostracods with which excludes a significant relocation (Sheppard and Bate, 1980; C. danielopoli as the dominant taxon in TO10 and TO12. Above Whatley, 1988; Purper and Ornellas, 1991; Whatley et al., 1998; (TO20, 23), C. danielopoli almost vanished and the spectrum is Ramos, 2006). largely dominated by P. olivencae (mainly juveniles), accompanied Stable isotope analyses (d18O, d13C; 137 valves in total) per- by Cyprideis spp. formed on both groups (freshwater: Penthesilenula olivencae, Cytheridella danielopoli; “brackish/marine”: Cyprideis graciosa, C. Geochemistry. C. pebasae (adults and juveniles) and C. graciosa pebasae, Rhadinocytherura amazonensis) yielded constantly very (adult) range within 7.03 to 8.73& in d18O and between 11.04 light values (Fig. 4, Table 1). No clear trends between different taxa, and 13.00& in d13C. adult and juvenile valves, localities or layers could be observed. However, such depleted d18O and d13C are compatible with 4.2.6. Barro Branco a freshwater system (Leng and Marshall, 2004) and does not permit Location. 22.1 km SSE Eirunepé (S 0652018.300/W 06947005.100; to conjecture about brackish waters or marine influences here. Our altitude: w120 m; section thickness: w10 m), left cutbank of the results are consistent with previous isotopic investigations (mainly Tarauacá River (Figs. 1b, 6). molluscs; Vonhof et al., 2003; Kaandorp et al., 2006; Ramos, 2006; Wesselingh et al., 2006c; Wesselingh, 2008; Wesselingh and Sedimentological interpretation (Gross et al., 2011). The lower part of Ramos, 2010). Moreover, pilot measurements on recent material the section comprises a succession of paleosols, which is overlain (Cytheridella ilosvayi) from two oxbow lakes along the Tarauacá by floodplain lake sediments (BA7a, 7b). Afterwards, a series of River show comparable values (d18O: 7.84 to 10.73&; d13C: crevasse splay deposits follows, which is topped by pedogenically 11.34 to 13.69&). overprinted floodplain fines. Consequently, based on the faunal composition as well as on stable isotope results, there is no hint that Cyprideis and Rhadino- Ostracods. At Barro Branco Cyprideis spp. dominates the fauna cytherura (most likely also Perissocytheridea) indicate brackish (84.8%); P. olivencae is the second most abundant taxon (10.1%) and waters or a marine influx in the Eirunepé outcrops. Apparently C. danielopoli occurs only in a low amount (1.3%). Cyprideis sp. 1 and these taxa have been successfully adapted to freshwater settings, Cypretta sp. are exclusive for Barro Branco. which is also well documented for Cyprideis today (Lake Tanga- nyika; Wouters and Martens, 2001, 2007) and proposed by Ramos Geochemistry. Values for adults and juveniles of C. pebasae (BA7a, (2006) and Latrubesse et al. (2010) earlier. Hence, palaeosalinity 7b) range within: d18O: 6.62 to 7.94&; d13C: 9.69 reconstructions of western Amazonia’s Neogene environments to 10.97&. The analysis of adults of R. amazonensis (BA7b) resul- based on the occurrence of highly endemic ostracod taxa should be ted in light values (d18O: 9.38&; d13C: 12.50&). used with some concern. M. Gross et al. / Journal of South American Earth Sciences 42 (2013) 216e241 237

5.1.2. Ostracod distribution a cluster analysis (Q-mode; software package PAST 1.92; Hammer Earlier sedimentological analyses (Gross et al., 2011) referred and Harper, 2008, Fig. 7) the most ostracod rich samples are the deposits of the sections around Eirunepé to a suspended load related to floodplain lakes or abandoned channels, which seem to river system (possibly to an anastomosing system) and demon- be less affected by the hydrodynamics of the river. These faunas are strated the inexistence of a long-lived lake s.str. there (“Lake dominated by C. danielopoli (cluster A). More fluvially characterised Pebas”; Wesselingh, 2007). Aside sandy channel deposits, the bulk layers display lower ostracod abundances and are dominated by of the sediments belongs to a variety of overbank environments Cyprideis spp. and/or P. olivencae (cluster B). Extant C. ilosvayi, (4.2.). Lacustrine sediments are related to abandoned channels and which is closely related to the C. danielopoli, prefers permanent, short-lived floodplain ponds/lakes, which were frequently affected lentic environments (Pérez et al., 2010a,b), while species of the by flooding events. The herein presented freshwater ostracod “incae” group of Penthesilenula are especially abundant in lotic assemblages corroborate well with this interpretation. Probably, habitats (inclusively floating macrophytes; Higuti et al., 2009a). this wetland is comparable to the Quaternary mega-fans of the Hence, a coarse, probably hydrodynamically controlled differenti- Chaco plain and the modern Pantanal as suggested by Latrubesse ation is suggested. Cyprideis is a euryoecious taxon, usually et al. (2007, 2010) for the uppermost Solimões Fm. (Late common in stressful environments. Based on the observed ostracod Miocene) in the south-western Brazil (Acre). distribution, Cyprideis seems to be better predisposed for unstable Thus, and in accordance with this continental sedimentation settings like short-lived ponds and more fluvially affected envi- model, we can assume a highly structured fluvial landscape with ronments. A further differentiation e as indicated by the cluster diverse habitats and substrates. An array of regional (e.g., flood analysis e is tentative because the sub environments cannot be pulses, seasonality) and local (e.g., physicochemical properties of characterised in more detail. the water, nutrient supply) parameters may have affected One sample from Torre da Lua (TO23; possibly also TO20) is ostracod assemblages in such a setting (Kaandorp et al., 2005; highly dominated by juveniles of P. olivencae, associated by few, Higuti et al., 2010). Since we suppose to deal with an avulsive mainly juveniles of Cyprideis spp. This association could be an river system, flooding events might have played an important role example for an allochthonous fauna (?related to floating meadows; in ostracod distribution (e.g., mixing of riverine and floodplain Higuti et al., 2010). Remarkably, among almost thousand specimens lake faunas, inclusively due to floating macrophytes; sudden (largely C. danielopoli) not even one fragment of Cyprideis was anoxia due to water level rise in floodplain lakes; Higuti et al., recovered from sample AQ19. To date this observation remains 2007, 2010). unexplained (ecological parameters?, simply patchiness of occur- Despite all these virtually unpredictable variables, a rough rence?). The absence of ostracods in some samples is clearly linked differentiation is signalised by the ostracod distribution. Based on to (palaeo-)pedogenic alteration (MN1, MN2 and AQ15/2).

Fig. 7. Dendrogram resulting from cluster analysis of ostracod data and generalised facies interpretation (FPL ¼ floodplain lake, FPP ¼ floodplain pond, CH ¼ channel). 238 M. Gross et al. / Journal of South American Earth Sciences 42 (2013) 216e241

5.2. Relation of the Eirunepé ostracod fauna to Pebasian faunas and colonisation of this herein treated, patchy structured environment remarks to the western Amazonian Cyprideis flock (Sandberg, 1964; Van Harten, 1990; Higuti et al., 2010). Within the Eirunepé ostracod fauna Cyprideis is the most Except a few works (e.g., Ramos, 2006), investigations of fossil specious taxon and clearly related to the Cyprideis species flock of ostracods from western Amazonia deal with material originating Neogene western Amazonia (Whatley et al., 1998). Comparable from the BrazilianeColombianePeruvian borderland, roughly radiation events are documented from modern freshwater Lake centred around the small town Pebas (w165 km ENE of Iquitos; e.g., Tanganyika (Wouters and Martens, 2001, 2007) as well as from the Purper, 1979; Muñoz-Torres et al., 1998, 2006; Whatley et al., 1998). brackish, Late Miocene Lake Pannon (Central Europe; Krstic, The ostracod composition of these “classical” Pebasian fauna, 1968a,b; Van Harten, 1990), both famous examples for long-lived where Cyprideis totally dominates the assemblages with up to 93%, lakes. Muñoz-Torres et al. (2006), Whatley et al. (1998) regarded differs notably from the Eirunepé fauna (only 36.6% Cyprideis). the palaeoenvironmental stability of a large, brackish, lacustrine Nonetheless, out of the 19 species recorded in Eirunepé, 9 species habitat to be the trigger for this extensive radiation. At least on are shared with the Pebasian fauna. Seven additional species e left a local scale, stable palaeoenvironments are missing in the Eir- in open nomenclature e show some similarity and only 3 taxa are unepé region (Gross et al., 2011). The combination of Cyprideis’ new records for the Neogene of western Amazonia (Fig. 8). Thus, euryoecious biology, its passive dispersal capacity and its sexual from taxonomical point of view, large affinities to the Pebasian reproduction mode, possibly contributed significantly to this fauna are evident. Depending on the dating of the Eirunepé fauna speciation event within a locally unstable but on a regional scale (late Middle Miocene or Late Miocene; see below, 5.3.), these long-lived wetland (Cohen and Johnston, 1987; Wouters and relations can be referred to a marginal position to the “Pebas Martens, 1994; Wilkinson et al., 2006). wetland” (Ramos, 2006; Celestino and Ramos, 2007; Wesselingh and Ramos, 2010) or, more likely, to the fade out of the Pebasian 5.3. Biostratigraphical remarks fauna before the onset of the Amazon system (Wesselingh et al., 2006a; Hoorn et al., 2010a; Gross et al., 2011). “One of the frustrating problems for a palaeontologist is the Interestingly, 97.8% of the ostracod fauna of the Eirunepé sections dating of Cenozoic fresh and brackish water deposits. They often is composed of only 3 genera, the darwinuloid genus Penthesilenula contain only endemic, and specialised, environmentally controlled and the cytheroid genera Cytheridella and Cyprideis. Taxa of the faunas.” (van den Bold, 1986, p. 141). Large progress is achieved Cypridoidea (e.g., candonids, ilyocyprinids, cypridids) were found during the last two decades but e at least in detail e this statement only subordinately, which are otherwise common in modern still characterises the “assumption-laden” (Wesselingh, 2008,p.15) freshwater settings (Martens et al., 2008; Higuti et al., 2009b). situation of western Amazonia’s stratigraphy. Whatley et al. (1998) argued for an elevated (brackish) salinity to be Until now, palynological biozonations are still used for correla- responsible for their rare occurrence. In Eirunepé that could not be tions across Amazonia and provide a rough temporal framework the case due to exclusively freshwater conditions. Extant Penthesi- (Hoorn, 1993, 1994). Nevertheless, definitions of pollen zones and lenula, Cytheridella and Cyprideis exhibit brood care, whereas cyp- their chronostratigraphic allocation are far beyond from settled ridoid ostracods to not brood (Smith and Delorme, 2010). down (Jaramillo et al., 2010; Latrubesse et al., 2010; Silva-Caminha Hypothetically, this mode of reproduction inherits a reproductive et al., 2010; Dino et al., 2012; see also Jaramillo et al., 2011 for advantage and/or facilitates dispersal (?rivers, ?birds) and detailed palynological zonations of the Colombian Llanos Basin). An in-depth appraisal of the mollusc contents resulted in a detailed mollusc biozonation, which is, however, tightly linked to the palynological concept (Wesselingh et al., 2006b). Similarly, Muñoz- Torres et al. (2006) proposed an ostracod biozonation for the Pebas/ Solimões Formation, based on a comprehensive analysis of Peru- vian, Colombian and Brazilian faunas. Again, the chronostrati- graphic calibration of the zones follows palynostratigraphy. Recently, Wesselingh and Ramos (2010; Wesselingh, 2008) adjusted the chronostratigraphic framework of mollusc and ostracod zones by placing the end of the “Pebas system” to w11 Ma in accordance to the onset of the modern Amazon system as esti- mated by Figueiredo et al. (2009; note that other authors argue for a latter onset; e.g., Wesselingh et al., 2006b; Hoorn et al., 2010a (Late Miocene); Latrubesse et al., 2010 (Latest MioceneeEarly Pliocene); Campbell, 2010 (Late Pliocene); Fig. 8). The herein reported ostracod taxa exhibit long ranges based on the available literature data (Muñoz-Torres et al., 2006; mainly early Middle to early Late Miocene according to the proposed chronology of Wesselingh and Ramos, 2010) and do not permit a more precise biostratigraphical dating today. Key taxa sensu Muñoz-Torres et al. (2006) are missing in the Eirunepé material. Only the co-occurrence of C. graciosa and C. longispina may hint to the C. obliquosulcata zone (latest Middle Miocene after Wesselingh and Ramos, 2010; but: early Late Miocene after Wesselingh et al., Fig. 8. Stratigraphic range of ostracod taxa based on Muñoz-Torres et al. (2006; 2006a,b; Fig. 8). chronostratigraphic correlation of mollusc and ostracod zones after Wesselingh and The mollusc faunas of Aquidabã and Torre da Lua are considered Ramos, 2010). For discussion on the estimated age of Eirunepé outcrops see text by Wesselingh et al. (2006a,b) to slightly post-date the Pebas fauna (thick lines ¼ characteristic faunal element according to Muñoz-Torres et al. (2006); w stippled lines ¼ not recorded; grey lines ¼ uncertain taxonomic equivalence; asterisks and to be of Late Miocene origin (possibly 9 Ma; but Wesselingh mark taxa shared with Pebasian faunas). and Ramos, 2010: late Middle Miocene). However, based on the M. Gross et al. / Journal of South American Earth Sciences 42 (2013) 216e241 239 occurrence of some palynological index taxa a Late Miocene age References (probably Asteraceae zone sensu Lorente, 1986; Jaramillo et al., 2010) is proposed (Gross et al., 2011; Paz et al., in press). This Alvarez Zarikian, C.A., Swart, P.K., Gifford, J.A., Blackwelder, P.L., 2005. Holocene paleohydrology of Little Salt Spring, Florida, based on ostracod assemblages allocation is strongly supported by the vertebrate fauna of Torre da and stable isotopes. Palaeogeography, Palaeoclimatology, Palaeoecology 225, Lua (Ramos, 2006), which belongs to the well documented “Acre 134e156. fauna” and is correlated to the Late Miocene (Huayquerian South Anderson, L.C., Wesselingh, F.P., Hartman, J.H., 2010. A phylogenetic and morpho- w e logic context for the radiation of an endemic fauna in a long-lived lake: Cor- American Land Mammal Age; 9.0 6.5 Ma; Cozzuol, 2006; bulidae (Bivalvia; Myoidea) in the Miocene Pebas Formation of western Latrubesse et al., 2007, 2010; Negri et al., 2010). In view of these Amazonia. Paleobiology 36, 534e554. palaeontological indications as well as the palaeogeographic Artheau, M., 2007. Geographical review of the ostracod genus Vestalenula (Dar- background of this region (Latrubesse et al., 1997, 2007, 2010; winulidae) and a new subterranean species from southern France. Invertebrate Systematics 21, 471e486. Cozzuol, 2006) a Late Miocene age (<9 Ma?) seems most plausible Ballent, S.C., Díaz, A.R., 2011. Contribution to the taxonomy, distribution and to date (Fig. 8). paleoecology of the early representatives of Penthesilenula Rossetti & Martens, 1998 (Crustacea, Ostracoda, Darwinulidae) from Argentina, with the description of a new species. Hydrobiologia. http://dx.doi.org/10.1007/s10750-011-0658-8. 6. Conclusions Campbell, K.E., 2010. Late Miocene onset of the Amazon River and the Amazon deep-sea fan: evidence from the Foz do Amazonas Basin: Comment. Geology 38, 212. The taxonomical analysis of ostracod faunas from the upper part Caputo, M.V., 1991. Solimões megashear: intraplate tectonics in northwestern of the Solimões Formation (Eirunepé area) documents a moder- Brazil. Geology 19, 246e249. ately diverse assemblage (19 species). Freshwater genera (particu- Celestino, E.A., Ramos, M.I.F., 2007. Sistemática de ostracodes e reconstruçã pale- w oambiental no Neógeno da bacia do Solimões, Formação Solimões, sudoeste do larly Cytheridella and Penthesilenula; 3/2 of the total fauna) were estada do Amazonas. In: XII Congresso Latino-Americano de Ciências do Mar, found co-occurring with usually brackish water indicating taxa Florianópolis, Associação Brasileira de Oceanografia, pp. 1e3. (especially Cyprideis; w1/3). Stable isotope analyses (d18O, d13C), Cisneros, J.C., 2005. New Pleistocene vertebrate fauna from El Salvador. Revista e performed on both groups, furnished for all taxa very negative Brasileira de Paleontologia 8, 239 255. Cohen, A.S., Johnston, M.R., 1987. Speciation in brooding and poorly dispersing values throughout and refer to exclusively freshwater conditions. lacustrine organisms. Palaios 2, 426e435. This is in concert with the fluvial sedimentation model presented Colin, J.-P., Tambareau, Y., Krasheninikov, V.A., 1996. Ostracodes limniques et ’ earlier (Gross et al., 2011). Consequently, the occurrence of Cypri- lagunaires dans le Crétacé supérieur du Mali (Afrique de l Ouest): systématique, paléoécologie et affinités paléobiogéographiques. Revue de Micropaléontologie deis and probably of some other “brackish/marine” taxa (Peri- 39, 211e222. ssocytheridea, Rhadinocytherura) is per se no general indicator for Colin, J.-P., Tambareau, Y., Krasheninikov, V.A., 1997. An early record of the genus the presence brackish waters or the influx of marine waters in this Cytheridella Daday, 1905 (Ostracoda, Limnocytheridae, Timiriaseviinae) from the Upper Cretaceous of Mali, West Africa: palaeobiogeographical and palae- region as mentioned previously (e.g., Purper, 1979; Whatley et al., oecological considerations. Journal of Micropalaeontology 16, 91e95. 1998) but already challenged by Ramos (2006) and Latrubesse Cozzuol, M.A., 2006. The Acre vertebrate fauna: age, diversity, and geography. et al. (2010). In combination with sedimentological interpreta- Journal of South American Earth Sciences 21, 185e203. De Deckker, P., Chivas, A.R., Shelley, J.M.G., 1999. Uptake of Mg and Sr in the tions a rough, hydrodynamically controlled differentiation of the euryhaline ostracod Cyprideis determined from in vitro experiments. Palae- faunas is suggested (more fluvially influenced: Cyprideise ogeography, Palaeoclimatology, Palaeoecology 148, 105e116. Penthesilenula-dominated; less fluvially characterised: Cyther- Del’ Arco, J.O., Santos, R.O.B., Rivetti, M., Olivera Alves, E.D., Fernandes, C.A.C., Silva, L.L., 1977. Folha SB. 19 Juruá. IeGeologia. Projeto Radambrasil. Levanta- idella-dominated). mento de Recursos Naturais 15, 19e88. Based on the taxonomical composition, a relation to the Dino, R., Soares, E.A.A., Antonioli, L., Riccomini, C., Nogueira, A.C.R., 2012. Paly- Middlee(early) Late Miocene faunas of the “Pebas system” is nostratigraphy and sedimentary facies of Middle Miocene fluvial deposits of obvious. Eight species of Cyprideis were found, which are clearly the Amazonas Basin, Brazil. Journal of South American Earth Sciences 34, 61e80. related to the Pebasian Cyprideis species flock. However, due to the Ferrero, L., 1996. Paleoecologia de ostrácodos holocenos del estuario del Rio Que- lack of index taxa sensu Muñoz-Torres et al. (2006) among ostra- quén Grande (Provincia de Buenos Aires). Ameghiniana 33, 209e222. ‘ cods, we consider the fauna of Eirunepé to post-date the “Pebas Figueiredo, J.J.P., 2012. Comment by J.P. Figueiredo, & Hoorn, C. on Late Miocene sedimentary environments in south-western Amazonia (Solimões Formation; system” as signified by other palaeontolgical evidences. We Brazil)’ by Martin Gross, Werner E. Piller, Maria Ines Ramos, Jackson Douglas da propose a Late Miocene age (<9 Ma?), somewhat prior to the Silva Paz. Journal of South American Earth Sciences 35, 74e75. effective installation of the modern Amazon system. Figueiredo, J., Hoorn, C., Ven van der, P., Soares, E., 2009. Late Miocene onset of the Amazon River and the Amazon deep-sea fan: evidence from the Foz do Ama- zonas Basin. Geology 37, 619e622. Acknowledgements Frenzel, P., Boomer, I., 2005. The use of ostracods from marginal marine, brackish waters as bioindicators of modern and Quaternary environmental change. Palaeogeography, Palaeoclimatology, Palaeoecology 225, 68e92. This study was financially supported by the Austrian Science Furtos, N.C., 1934. Two new species of Cypretta (Ostracoda) from the Marquesas Fund project P21748-N21. We acknowledge the Conselho Nacional Island and Florida with notes on the distribution of the genus. Bernice P. Bishop e fi de Desenvolvimento Científico e Tecnológico/Ministério da Ciência Museum Bulletin 114, 279 286 [Preprint 1934, Paci c Entomological Survey, article 21]. e Tecnologia (CNPq/MCT; process number EXC 010389/2009-1) as Gingras, M.J., Räsänen, M.E., Pemberton, S.G., Romero, L.P., 2002. Ichnology and well as the Departamento Nacional de Produção Mineral (DNPM/ sedimentology reveal depositional characteristics of bay-margin parasequences Manaus, especially Gert Woeltje) for sampling and shipping in the Miocene Amazonian foreland basin. Journal of Sedimentary Research 72, 871e883. licences. The authors are grateful to Jackson Douglas da Silva Paz Gross, M., Minati, K., Danielopol, D.L., Piller, W.E., 2008. Environmental changes and (Universidade Federal de Mato Grosso), Miguel Sombra (Eirunepé) diversification of Cyprideis in the Late Miocene of the Styrian Basin (Lake e and Norbert Winkler (Joanneum, Graz) for their assistance during Pannon, Austria). Senckenbergiana Lethaea 88, 161 181. fi Gross, M., Piller, W.E., Ramos, M.I., Paz, J.D.S., 2011. Late Miocene sedimentary eldwork. Finally, we would like to thank the two reviewers of this environments in south-western Amazonia (Solimões Formation; Brazil). Journal paper e Elsa Gliozzi and Edgardo M. Latrubesse e for their of South American Earth Sciences 32, 169e181. constructive comments. Grossman, E.L., Ku, T.-L., 1986. Oxygen and carbon isotope fractionation in biogenic aragonite: temperature effects. Chemical Geology 59, 59e74. Haffer, J., 2008. Hypotheses to explain the origin of species in Amazonia. Brazilian Appendix A. Supplementary data Journal of Biology 68, 917e947. Supplement. Hammer, Ø., Harper, D., 2008. Paleontological Data Analysis. Blackwell, Oxford, 351 pp. Hartmann, G., 1989. Ostracoda. Dr. H.G. Bronns Klassen und Ordnungen des Tier- Supplementary data related to this article can be found at http:// reichs, 5, 1. Abteilung, 2. Buch, 4. Teil, 5. Lieferung, VEB Gustav Fischer Verlag, dx.doi.org/10.1016/j.jsames.2012.10.002. Jena, pp. 787e1067. 240 M. Gross et al. / Journal of South American Earth Sciences 42 (2013) 216e241

Higuti, J., Velho, L.F.M., Lansac-Tôha, F.A., Martens, K., 2007. Pleuston communities Linhares, A.P., Ramos, M.I.F., Gross, M., Piller, W.E., 2011. Evidence for marine influx are buffered from regional flood pulses: the example of ostracods in the Paraná during the Miocene in southwestern Amazonia, Brazil. Geología Colombiana 36, River floodplain, Brazil. Freshwater Biology 52, 1930e1943. 91e103. Higuti, J., Lansac-Tôha, F.A., Velho, L.F.M., Pinto, R.L., Vieira, L.C.G., Martens, K., Lister, K.H., 1975. Quaternary Freshwater Ostracoda from the Great Salt Lake Basin. 2009a. Composition and distribution of Darwinulidae (Crustacea, Ostracoda) in The University of Kansas Paleontological contributions, Utah. Paper 78, 1e34. the alluvial valley of the upper Paraná River, Brazil. Brazilian Journal of Biology Löffler, H., 1961. Zur Systematik und Ökologie der chilenischen 69, 253e262. Süßwasserentomostraken. Beiträge zur neotropischen Fauna 2, 197e206. Higuti, J., Lansac-Tôha, F.A., Velho, L.F.M., Martens, K., 2009b. Biodiversity of non- Lorente, M.A., 1986. Palynology and Palynofacies of the Upper Tertiary in Venezuela. marine ostracods (Crustacea, Ostracoda) in the alluvial valley of the upper Dissertationes Botanicae, vol. 99, pp. 1e225. Paraná River, Brazil. Brazilian Journal of Biology 69, 661e668. Maia, R.G.N., Godoy, H.K., Yamaguti, H.S., Moura, P.A., Costa, F.S.F., Holanda, M.A., Higuti, J., Declerck, S.A.J., Lansac-Tôha, F.A., Velho, L.F.M., Martens, K., 2010. Vari- Costa, J.A., 1977. Projeto Carvão no Alto Solimões. Relatório Final. Companhia de ation in ostracod (Crustacea, Ostracoda) communities in the alluvial valley of Pesquisa de Recursos Minerais-Departamento Nacional da Produçáo Mineral, the upper Paraná River (Brazil) in relation to substrate. Hydrobiologia 644, Manaus, 142 pp. 261e278. Malz, H., 1977. Cypridopsine Ostracoden aus dem Tertiär des Mainzer Beckens. Holmes, J.A., 1998. A late Quaternary ostracod record from Wallywash Great Pond, Senckenbergiana Lethaea 58, 219e261. a Jamaican marl lake. Journal of Paleolimnology 19, 115e128. Martens, K., Savatenalinton, S., 2011. A subjective checklist of the Recent, free-living Hoorn, C., 1993. Marine incursions and the influence of Andean tectonics on the non-marine Ostracoda (Crustacea). Zootaxa 2855, 1e79. Miocene depositional history of northwestern Amazonia: results of a palynos- Martens, K., Rossetti, G., Horne, D.J., 2003. How ancient are ancient asexuals? tratigraphic study. Palaeogeography, Palaeoclimatology, Palaeoecology 105, Proceedings of the Royal Society of London, B 270, 723e729. 267e309. Martens, K., Schön, I., Meisch, C., Horne, D.J., 2008. Global diversity of ostracods Hoorn, C., 1994. An environmental reconstruction of the palaeo-Amazon River (Ostracoda, Crustacea) in freshwater. Hydrobiologia 595, 185e193. system (MiddleeLate Miocene, NW Amazonia). Palaeogeography, Palae- Mebrouk, F., Colin, J.-P., Hennache, F., 2011. Un gisement d’ostracodes non-marins oclimatology, Palaeoecology 112, 187e238. dans l’Éocène inférieur du Djebel Amour, Atlas saharien central, Algérie: tax- Hoorn, C., 2006. Mangrove forests and marine incursions in Neogene Amazonia onomie, paléoécologie et paléobiogéographie. Carnets de Géologie (Notebooks (Lower Apaporis River, Colombia). Palaios 21, 197e209. on Geology) 2011/04, 83e97. Hoorn, C., Wesselingh, F.P. (Eds.), 2010. Amazonia, Landscape and Species Evolution: Meisch, C., 2000. Freshwater ostracoda of western and central Europe. In: A Look into the Past. Wiley-Blackwell, Oxford, p. 464. Schwoerbel, J., Zwick, P. (Eds.), Süßwasserfauna von Mitteleuropa, vol. 8/3. Hoorn, C., Wesselingh, F.P., ter Steege, H., Bermudez, M.A., Mora, A., Sevink, J., Spektrum Akademischer Verlag, Heidelberg/Berlin, 522 pp. Sanmartín, I., Sanchez-Meseguer, A., Anderson, C.L., Figueiredo, J.P., Jaramillo, C., Minati, K., Cabral, M.C., Pipík, R., Danielopol, D.L., Linhart, J., Neubauer, W., 2008. Riff, D., Negri, F.R., Hooghiemstra, H., Lundberg, J., Stadler, T., Särkinen, T., Morphological variability among European populations of Vestalenula cylindrica Antonelli, A., 2010a. Amazonia through time: Andean uplift, climate change, (Straub) (Crustacea, Ostracoda). Palaeogeography, Palaeoclimatology, Palae- landscape evolution, and biodiversity. Science 330, 927e931. oecology 264, 296e305. Hoorn, C., Wesselingh, F.P., Hovikoski, J., Guerrero, J., 2010b. The development of the Muñoz-Torres, F., Whatley, R.C., Van Harten, D., 1998. The endemic non-marine Amazonian mega-wetland (Miocene; Brazil, Colombia, Peru, Bolivia). In: Miocene ostracod fauna of the Upper Amazon Basin. Revista Española de Hoorn, C., Wesselingh, F.P. (Eds.), Amazonia, Landscape and Species Evolution: A Micropaleontología 30, 89e105. Look into the Past. Wiley-Blackwell, Oxford, pp. 123e142. Muñoz-Torres, F.A., Whatley, R.C., Van Harten, D., 2006. Miocene ostracod (Crus- Horne, F.R., 1996. Ostracods of Texas playas. The Southwestern Naturalist 41, tacea) biostratigraphy of the upper Amazon Basin and evolution of the genus 450e455. Cyprideis. Journal of South American Earth Sciences 21, 75e86. Hovikoski, J., Gingras, M., Räsänen, M., Rebata, L.A., Guerrero, J., Ranzi, A., Namiotko, T., Danielopol, D.L., 2004. Review of the eremita species-group of the Melo, J., Romero, L., Nuñez del Prado, H., Jaimes, F., Lopez, S., 2007. The Pseudocandona Kaufmann (Ostracoda, Crustacea), with the description of a new nature of Miocene Amazonian epicontinental embayment: high-frequency species. Revista Española de Micropaleontología 36, 109e125. shifts of the low-gradient coastline. Geological Society of America Bulletin Negri, F.R., Bocquentin-Villanueva, J., Ferigolo, J., Antoine, P.-O., 2010. A review of 119, 1506e1520. Tertiary mammal faunas and birds from western Amazonia. In: Hoorn, C., Hovikoski, J., Wesselingh, F.P., Räsänen, M., Gingras, M., Vonhof, H.B., 2010. Marine Wesselingh, F.P. (Eds.), Amazonia, Landscape and Species Evolution: A Look into influence in Amazonia: evidence from the geological record. In: Hoorn, C., the Past. Wiley-Blackwell, Oxford, pp. 245e258. Wesselingh, F.P. (Eds.), Amazonia, Landscape and Species Evolution: A Look into Paz, J.D.S., Ramos, M.I.F., Moraes-Santos, H.M., Silva, S. Depósitos fluviais da for- the Past. Wiley-Blackwell, Oxford, pp. 143e161. mação Solimões, Bacia do Solimões, Eirunepé (AM), Brasil. Latin American Jahn, A., Gamenick, I., Theede, H., 1996. Physiological adaptations of Cyprideis torosa Journal of Sedimentology and Basin Analysis, in press. (Crustacea, Ostracoda) to hydrogen sulphide. Marine Ecology Progress Series Pérez, L., Lorenschat, J., Brenner, M., Scharf, B., Schwalb, A., 2010a. Extant freshwater 142, 215e223. ostracodes (Crustacea: Ostracoda) from Lago Petén Itzá, Guatemala. Revista de Jaramillo, C., Hoorn, C., Silva, S.A.F., Leite, F., Herrera, F., Quiroz, L., Dino, R., Biologia Tropical 58, 871e895. Antonioli, L., 2010. The origin of the modern Amazon rainforest: implications of Pérez, L., Lorenschat, J., Bugja, R., Brenner, M., Scharf, B., Schwalb, A., 2010b. the palynological and palaeobotanical record. In: Hoorn, C., Wesselingh, F.P. Distribution, diversity and ecology of modern freshwater ostracodes (Crus- (Eds.), Amazonia, Landscape and Species Evolution: A Look into the Past. Wiley- tacea), and hydrochemical characteristics of Lago Petén Itzá, Guatemala. Journal Blackwell, Oxford, pp. 317e334. of Limnology 69, 146e159. Jaramillo, C.A., Rueda, M., Torres, V., 2011. A palynological zonation for the Cenozoic Pinto, I., Ornellas, L.P., 1970. A new brackishwater ostracode Perissocytheridea of the Llanos and Llanos Foothills of Colombia. Palynology 35, 46e84. krömmelbeini Pinto & Ornellas, sp. nov., from southern Brazil. Escola de Geo- Kaandorp, R.J.G., Vonhof, H.B., Wesselingh, F.P., Romero Pittman, L., Kroon, D., Van logia de Porto Alegre, Publicação Especial 20, 1e19. Hinte, J.E., 2005. Seasonal Amazonian rainfall variation in the Miocene climate Pinto, R.L., Rocha, C.E.F., Martens, K., 2004. On the genus Penthesilenula Rossetti and optimum. Palaeogeography, Palaeoclimatology, Palaeoecology 221, 1e6. Martens, 1998 (Crustacea, Ostracoda, Darwinulidae) from (semi-) terrestrial Kaandorp, R.J.G., Wesselingh, F.P., Vonhof, H.B., 2006. Ecological implications from habitats in São Paulo State (Brazil), with the description of a new species. geochemical records of Miocene Western Amazonian bivalves. Journal of South Journal of Natural History 38, 2567e2589. American Earth Sciences 21, 54e74. Purper, I., 1974. Cytheridella boldii Purper, sp. nov. (Ostracoda) from Venezuela and Keyser, D., 1976. Zur Kenntnis der brackigen mangrovenbewachsenen Weichböden a Revision of the Genus Cytheridella Daday, 1905. Anais da Academia Brasileira Südwest-Floridas unter besonderer Berücksichtigung ihrer Ostracodenfauna de Ciências 46, 635e662. thesis., University of Hamburg, Hamburg, 142 pp. Purper, I., 1977. Some ostracodes from the Upper Amazon Basin. Brazil. Environ- Keyser, D., 1977. Brackwasser-Cytheracea aus Süd-Florida. Abhandlungen und Ver- ment and age. In: Löffler, H., Danielopol, D.L. (Eds.), Aspects of Ecology and handlungen des naturwissenschaftlichen Vereins in Hamburg, Neue Folge, 20, Zoogeography of Recent and Fossil Ostracoda. W. Junk Publishers, The Hague, pp. 43e85. pp. 353e367. Krstic, N., 1968a. Ostracodes des couches congeriennes: 1. Cyprideis I. Bulletin du Purper, I., 1979. Cenozoic Ostracodes of the upper Amazon Basin, Brazil. Pesquisas Museum d’Histoire Naturelle, A 23, 107e151. 12, 209e281. Krstic, N., 1968b. Ostracodes des couches congeriennes: 3. Cyprideis II. Bulletin du Purper, I., Ornellas, L.P., 1991. New Ostracodes of the endemic fauna of the Pebas Museum d’Histoire Naturelle, A 23, 153e183. Formation, Upper Amazon Basin, Brazil. Pesquisas 18, 25e30. Latrubesse, E.M., Bocquentin, J., Santos, J.C.R., Ramonell, C.G., 1997. Paleoenvir- Purper, I., Pinto, I.D., 1983. New genera and species of ostracodes of the Upper onmental model for the late Cenozoic of southwestern Amazonia: paleontology Amazon Basin, Brasil. Pesquisas 15, 113e126. and geology. Acta Amazonica 27, 103e118. Purper, I., Pinto, I.D., 1985. New data and new ostracods from Pebas Formation e Latrubesse, E., Silva, S.A.F., Cozzuol, M., Absy, M.L., 2007. Late Miocene continental Upper Amazon Basin. In: 8 Congresso Brasileiro de Paleontologia, Rio de sedimentation in the southwestern Amazonia and its regional significance: Janeiro, 1985, Seria Geologia, Paleontologia/Estratigrafia, DNPM vol.27, pp. biotic and geological evidence. Journal of South American Earth Sciences 23, 427e441. 61e80. Purper, I., Würdig-Maciel, N.L., 1974. Occurrence of Heterocypris incongruens Latrubesse, E.M., Cozzuol, M., Silva-Caminha, S.A.F., Rigsby, C.A., Absy, M.L., (Ramdohr), 1808 e Ostracoda e in Rio Grande do Sul, Brazil. Discussion on the Jaramillo, C., 2010. The Late Miocene paleogeography of the Amazon Basin and allied genera: Cyprinotus, Hemicypris, Homocypris and Eucypris. Pesquisas 3, the evolution of the Amazon River system. Earth-Science Reviews 99, 99e124. 69e91. Leng, M.J., Marshall, J.D., 2004. Palaeoclimate interpretation of stable isotope data Ramos, M.I.F., 2006. Ostracods from the Neogene Solimões Formation (Amazonas, from lake sediment archives. Quaternary Science Reviews 23, 811e831. Brazil). Journal of South American Earth Sciences 21, 87e95. M. Gross et al. / Journal of South American Earth Sciences 42 (2013) 216e241 241

Ramos, M.I.F., Santos, H.M.M., Costa, S.A.R.F., Toledo, P.M., 2009. CatáIogo de Fósseis. Wanderley-Filho, J.R., Eiras, J.F., Cunha, P.R.C., Ven van der, P.H., 2010. The Paleozoic In: CoIeção Paleontológica do Museu Paraense Emilio Goeldi. Museu Paranese Solimões and Amazonas basins and the Acre foreland basin of Brazil. In: Emilio Goeldi, Belém, 172 pp. Hoorn, C., Wesselingh, F.P. (Eds.), Amazonia, Landscape and Species Evolution: A Reheis, M.C., Laabs, B.J.C., Forester, R.M., McGeehin, J.P., Kaufman, D.S., Bright, J., Look into the Past. Wiley-Blackwell, Oxford, pp. 29e37. 2005. Surficial Deposits in the Bear Lake Basin. U.S. Geological Survey, Open-File Wesselingh, F.P., 2007. Long-lived lake molluscs as island faunas: a bivalve Report 2005-1088, pp. 1e30. perspective. In: Renema, W. (Ed.), Biogeography, Time, and Place: Distributions, Rossetti, G., Martens, K., 1998. Taxonomic revision of the Recent and Holocene Barriers, and Islands. Springer, Dordrecht, pp. 275e314. representatives of the family Darwinulidae (Crustacea, Ostracoda), with Wesselingh, F.P., 2008. Molluscan radiations and landscape evolution in Miocene a description of three new genera. Bulletin van het Koninklijk Belgisch Instituut Amazonia. Annales Universitates Turkuensis, Biologica-Geographica-Geologica voor Natuurwetenschappen, Biologie 68, 55e110. 232, 1e41. Rossetti, G., Pinto, R.L., Martens, K., 2011. Description of a new genus and two new Wesselingh, F.P., Ramos, M.I.F., 2010. Amazonian aquatic invertebrate faunas species of Darwinulidae (Crustacea, Ostracoda), from Christmas Island (Indian (Mollusca, Ostracoda) and their development over the past 30 million years. In: Ocean) with some considerations on the morphological evolution of ancient Hoorn, C., Wesselingh, F.P. (Eds.), Amazonia, Landscape and Species Evolution: A asexuals. Belgian Journal of Zoology 141, 55e74. Look into the Past. Wiley-Blackwell, Oxford, pp. 302e316. Ruskin, B.G., Dávila, F.M., Hoke, G.D., Jordan, T.E., Astini, R.A., Alonso, R., 2011. Stable Wesselingh, F.P., Räsänen, M.E., Irion, G., Vonhof, H.B., Kaandorp, R., Renema, W., isotope composition of middle Miocene carbonates of the Frontal Cordillera and Romero Pittman, L., Gingras, M., 2002. Lake Pebas: a palaeoecological recon- Sierras Pampeanas: did the Paranaense seaway flood western and central struction of a Miocene, long-lived lake complex in western Amazonia. Cainozoic Argentina? Palaeogeography, Palaeoclimatology, Palaeoecology 308, 293e303. Research 1, 35e81. Sandberg, P.A., 1964. The ostracod genus Cyprideis in the Americas. Acta Uni- Wesselingh, F.P., Ranzi, A., Räsänen, M.E., 2006a. Miocene freshwater Mollusca from versitatis Stockholmiensis 12, 1e178. western Brazilian Amazonia. Scripta Geologica 133, 419e437. Sheppard, L.M., Bate, R.H., 1980. Plio-Pleistocene ostracods from the Upper Amazon Wesselingh, F.P., Hoorn, M.C., Guerrero, J., Räsänen, M.E., Romero Pittmann, L., of Colombia and Peru. Palaeontology 23, 97e124. Salo, J., 2006b. The stratigraphy and regional structure of Miocene deposits in Silva-Caminha, S.A.F., Jaramillo, C.A., Absy, M.L., 2010. Neogene palynology of the western Amazonia (Peru, Colombia and Brazil), with implications for late Solimões Basin, Brazilian Amazonia. Palaeontographica, B 284, 13e79. Neogene landscape evolution. Scripta Geologica 133, 291e322. Smith, A.J., Delorme, L.D., 2010. Chapter 19. Ostracoda. In: Thorp, J.H., Covich, A.L. Wesselingh, F.P., Kaandorp, R.J.G., Vonhof, H.B., Räsänen, M.E., Renema, W., (Eds.), Ecology and Classification of North American Freshwater Invertebrates. Gingras, M., 2006c. The nature of aquatic landscapes in the Miocene of western Academic Press, London, pp. 725e772. Amazonia: an integrated palaeontological and geochemical approach. Scripta Smith, R.J., Kamiya, T., 2008. The ontogeny of two species of Darwinuloidea Geologica 133, 363e393. (Ostracoda, Crustacea). Zoologischer Anzeiger 247, 275e302. Wesselingh, F.P., Hoorn, C., Kroonenberg, S.B., Antonelli, A., Lundberg, J.G., Sohn, I.G., Kornicker, L.S., 1973. Morphology of Cypretta kawatai Sohn and Kornicker, Vonhof, H.B., Hooghiemstra, H., 2010. On the origin of Amazonian landscapes 1972 (Crustacea, Ostracoda), with a discussion of the genus. Smithsonian and biodiversity: a synthesis. In: Hoorn, C., Wesselingh, F.P. (Eds.), Amazonia, Contributions to Zoology 141, 1e28. Landscape and Species Evolution: A Look into the Past. Wiley-Blackwell, Oxford, Swain, F.M., 1998. Ostracoda from the Pliocene? Pebas Formation at Iquitos, Peru. pp. 421e431. http://www.geo.umn.edu/people/profs/swain/iquitos_Peru.pdf. Whatley, R.C., 1988. Population structure of ostracods: some general principles for Tressler, W.L., 1949. Fresh-water Ostracoda from Brazil. Proceedings of the United the recognition of palaeoenvironments. In: De Deckker, P., Colin, J.P., States National Museum 100, 61e83. Peypouquet, J.P. (Eds.), Ostracoda in the Earth Sciences. Elsevier, Amsterdam, Triebel, E., 1953. Eine fossile Pelocypris (Crust., Ostr.) aus El Salvador. Senck- pp. 245e256. enbergiana Lethaea 34, 1e4. Whatley, R.C., Muñoz-Torres, F., Van Harten, D., 1998. The Ostracoda of an isolated Van Doninck, K., Schön, I., Maes, F., Bruyn, L., Martens, K., 2003. Ecological strategies Neogene saline lake in the western Amazon Basin, Peru. Bulletin du Centres de in the ancient asexual animal group Darwinulidae (Crustacea, Ostracoda). Recherches Elf Exploration-Production, Memoires 20, 231e245. Freshwater Biology 48, 1285e1294. Whatley, R.C., Muñoz-Torres, F., Van Harten, D., 2000. Skopaeocythere: a minute new van den Bold, W.A., 1986. Fresh and brackish water ostracoda from the Neogene of limnocytherid (Crustacea, Ostracoda) from the Neogene of the Amazon Basin. northern Venezuela. Tulane Studies in Geology and Paleontology 19, 141e157. Ameghiniana 37, 163e167. Van Harten, D., 1979. Some new shell characters to diagnose the species of the Wilkinson, M.J., Marshall, L.G., Lundberg, J.G., 2006. River behavior on megafans and Ilyocypris gibba-biplicata-bradyi group and their ecological significance. In: potential influences on diversification and distribution of aquatic organisms. Krstic, N. (Ed.), Taxonomy, Biostratigraphy and Distribution of Ostracods. Journal of South American Earth Sciences 21, 151e172. Serbian Geological Society, Beograd, pp. 71e76. Wouters, K., Martens, K., 1994. Contribution to the knowledge of the Cyprideis Van Harten, D., 1990. The Neogene evolutionary radiation in Cyprideis Jones species flock (Crustacea: Ostracoda) of Lake Tanganyika, with the description of (Ostracoda: Cytheracea) in the Mediterranean area and the Paratethys. Courier three new species. Bulletin de l’Institut Royal des Sciences Naturelles de Bel- Forschungsinstitut Senckenberg 123, 191e198. gique, Biologie 64, 111e128. Van Morkhoven, F.P.C.M., 1963. Post-palaeozoic Ostracoda, Their Morphology, Wouters, K., Martens, K., 2001. On the Cyprideis species flock (Crustacea, Ostracoda) Taxonomy, and Economic Use. In: Generic Descriptions, vol 2. Elsevier in Lake Tanganyika, with the description of four new species. Hydrobiologia Publishing Company, Amsterdam/London/New York, 478 pp. 450, 111e127. Victor, R., Fernando, C.H., 1981. Freshwater Ostracods (Crustacea: Ostracoda) of Wouters, K., Martens, K., 2007. Three new species of the Cyprideis species flock the genus Cypretta Vavra, 1895 from Malaysia, Indonesia and the Philippines. (Crustacea, Ostracoda) of Lake Tanganyika (East Africa). Bulletin de l’Institut Internationale Revue der gesamten Hydrobiologie und Hydrographie 66, Royal des Sciences Naturelles de Belgique, Biologie 77, 147e160. 415e433. Würdig, N.L., Freitas, S.F., Fausto, I.V., 1990. Comunidade de ostracods associada ao Vonhof, H.B., Wesselingh, F.P., Kaandorp, R.J.G., Davies, G.R., Van Hinte, J.E., bentos e nacrofitas aquatica da lagoa do Gentil, Tramandai, Rio Grande do Sul. Guerrero, J., Räsänen, M., Romero-Pittman, L., Ranzi, A., 2003. Paleogeography Acta Limnologica Brasiliensia 3, 807e828. of Miocene Western Amazonia: Isotopic composition of molluscan shells Yang, F., Sun, Z., Zhang, Y., Qiao, Z., 2002. Taxonomic significance of nodal orna- constrains the influence of marine incursions. Geological Society of America mentation of Quaternary genus Ilyocypris (Ostracoda) from Qaidam Basin, Bulletin 115, 983e993. Qinghai province. Acta Micropalaeontologica Sinica 19, 15e32.