Lower and Middle Triassic foraminifera from the Eros Limestone, Hydra Island, Greece

The systematics and stratigraphic ranges (constrained by conodont dating) of abundant and well preserved foraminiferal faunas from six sections in the Lower and Middle Triassic Eros Limestone of central and western Hydra (Argolis Peninsula, Greece) are described. A joint analysis of the conodonts, foraminifera and bivalves has enabled the Scythian and Anisian stages to be recognized with some certainty within the Eros Limestone carbonate platform. The foraminifera have affinities with those of many other Tethyan localities, in particular the Dinarides, Balkans, Carpathians and the Southern Alps.


INTRODUCTION
The small Island of Hydra is located in the Aegean just to the south of the Argolis Peninsula (Peloponnesus, Greece) ( Fig. 1). Its sedimentary succession ranges from Permian to Jurassic in age.
This paper seeks to improve our knowledge of the biostra1.igrapliy of the Eros Limestone by providing new important data on the distribution of some of the Triassic foraminifera, calibrated to the standard stages and substages of the Triassic, using mainly conodonts.

GEOLOGICAL SETTING
The island of Hydra belongs to the Subpelagonian Zone, defined as a passive continental margin succession deposited on the western part of the Pelagonian Block, according to Jacobshagen et al. (1977;1989 pers. comm.). Both Subpelagonian and Pelagonian zones are part of the Internal Hellenides (Fig. 1).
The sedimentary sequence of Hydra is arranged in four major south vergent thrust sheets, dissected by transcurrent faults (Fig. 3). The Eros Limestone constitutes the backbone of the southern thrust sheet and part of the 0. Zakoni and Kutali thrust sheets (nomenclature after Angiolini et al. 1992). The areal distribution of the Eros Limestone, together with the schematic tectonic setting of the island, are shown in Fig. 3.

THE EROS LIMESTONE
The Eros Limestone was named by Romermann (1 968) after the highest mountain of Hydra (Mount Eros, 589m). It constitutes a well-developed carbonate platform, 600-670 m thick, of Early to Middle Triassic age. It overlies the Aghios Nikolaos Formation (quarzarenites; Early Triassic) or directly the Episkopi Limestone (bioclastic limestone; Late Permian) and is bounded at the top by the Han Bulog Limestone (red nodular limestone; latest Pelsonian-latest Illyrian/Ladinian) or directly by the Adhami Limestone (grey cherty limestone; Early Ladinian-Late Triassic).
Angiolini et al. (1992) subdivided the Eros Limestone on the basis of microfacies and field analyses. The present work improves these analyses. Six stratigraphic sections are presented here (Fig.4): sections 1, 3, 4, 5 and 6 were sampled by the present authors, section 2 by M. Richards (University of Lausanne). We have divided the Eros Limestone into four subunits: the lower lithozone, the Eros Limestone sensu szricto, the dark member and the upper lithozone.
The lower lithozone (25-100 m thick) constitutes the base of the Eros platform (Fig. 2) and consists of grey oolitic limestone with a variable content of quartz grains; yellow siltstone interbeds with species of the bivalve genera Unionites and Eurnorphotis of Spathian age (determination by R. Posenato, University of Ferrara) are also present.
The dark member (0-200 m thick) interfingers with the Eros Limestone sensu stricto (Fig. 2) and consists of well-bedded platy beds (centimetres to decimetres thick) of dark limestone with chert. Massive beds of intraformational breccias are present and a later dolomitization is also evident.
The dark member and the Eros Limestone sensu stricto are overlain by the upper lithozone (10-30 m thick), made up of red matrix-bearing calcareous breccias with bioclastic clasts.

NE
The depositional environment of the Eros Limestone is characteristic of a carbonate platform developed on a marine terrigenous flat (Aghios Nikolaos Formation). Oolitic and bioclastic grainstones and packstones. probably forming sand bars, are widespread. ln places (Aghios Marina), these bars seem to have enclosed lower energy areas where mudstones and wackestones were deposited. A hard rim (bindstone) is present in the northeastern part of Hydra (0. Malies) and is apparently localized only at this site. The dark member probably represents not very deep and poorly oxygenated troughs dissecting the carbonate platform. Their areal distribution, in a narrow SW-NE alignment, which together with the presence of abundant intraformational breccias, seems to suggest a tectonic (extensional) origin.
The age of the Eros Limestone ranges from Spathian to the Pelsonian substage of the Anisian (Fig. 2), based on the presence of Spathian bivalves in the lower lithozone, Late Scythian conodonts (Neospathodus horneri (Bender) in sample MR225) in the dark member of the Eros Limestone and Pelsonian conodonts (Gondolella bulgarica (Budurov & Stefanov), G. bifurcata hifurcata (Budurov & Stepanov), G. bifiucata hanhulogi (Sudar & Budurov), and Gladigondolella tethydis (Huckriede)) in the upper lithozone [all the condononts listed in this paper and their stratigraphic ranges have been determined by A. Nicora, University of Milan, and are taken from Angiolini et al., 19921. The foraminiferal assemblages are in agreement with and support this age determination.
Of interest is the occurrence of Krikoiirnbilica pileiformis H e in the Scythian stage; this species, up until now, is known Lower and A4iddle Triassic Foraminifera, Greece     only in the Middle Triassic of southern Guizhou (China). Furthermore, in Hydra, the stratigraphic range of M. pusilla (Ho) also extends into the Anisian (see below), whereas M. cheni (Ho) appears to be restricted to the Scythian.

ANISIAN FORAMINIFERAL ASSEMBLAGE
The Anisian foraminifera] assemblage of Hydra is very rich, diverse and well preserved in most samples. It occurs in the Eros Limestone S.S. mainly in microfacies 3 and 4. more rarely in microfacies 2 (Figs 4-5). These foraminifera belong to the Early and Middle Anisian time interval on the basis of the conodont association (listed above) recorded in the upper lithozone of the Eros Limestone, which directly overlies microfacies 3 and 4 of the Eros Limestone S.S. The typical Anisian foraminifera] assemblage is represented by

M e M r m p r a dinanca Koctransky
In the Eros Limestone (upper lithozone), the joint occurrence of Meandrospira pusillu and M. dinarica has been detected in sample G L 155 (Tsingri Bay, in front of Tsingry Island, see Fig. 3); the age of this assemblage is Pelsonian on the basis of conodonts. This points to a Scythian-Anisian distribution of M. pusilla.

Biostratigraphic considerations
Angiolini ~> t a/. (1992) described the stratigraphy of the Eros Limestone of Hydra, their dating being mainly with the aid of conodont assemblages.
In the present paper we have described the foraminifera1 fauna recorded in the same stratigraphic interval, providing further data on the occurrence of some important Early and Middle Triassic foraminifera (Table 2) in the Tethyan realm. Of particular importance are the following: (A) Pilamminu praedensa: the Scythian range reported by Urosevic (1988) is here confirmed on the evidence of the associated conodonts and bivalves. (B) Pilamminu densa: in Hydra, this species has been recorded only in the Pelsonian. (C) Meandrospira: in the Eros Limestone the distribution of this genus suggests the existence of a lineage (M. cheni-M. dinurica) evolving laterally from M . pusillu. The arguments supporting this interpretation are: first, the Scythian portion of the Eros Limestone hosts both M . pusilla and M . cheni; secondly, during the Early Anisian M. rlinarica appears, surviving until the Pelsonian, together with rare specimens of M. piisilla; and thirdly, M. cheni has been found only in the Spathian substage. (D) Krikoumhilica pileiformis: previously, this species has been known only in the Middle Triassic of China. Therefore, its Scythian occurrence in Hydra indicates that it must have appeared first in the Early Triassic. (E) Glomospirellu grandis: in Hydra, it has been found only in the Anisian (as high as the Pelsonian), in association with Meandrospira dinarica, thus confirming the restriction of this species to the Anisian. (F) Endoteba: the Upper Permian genus Endoteha is reported in the Triassic (Pelsonian) for the first time. A more detailed analysis of Triassic 'Endothyras' is needed to confirm this record. (G) Endothyranella wirzi: this species, also known in the Ladinian (Zaninetti, 1976), is here recorded in the Early Anisian, as high as the Pelsonian. We can confirm that elsewhere this species ranges through the Middle Triassic. (H) Palaeomiliolina judicariensis: appears to be restricted to Thin sections of forminifera. 6.1. Duostominidae: ?Duosfornina sp., sample: H186, stratigraphic section 6. 6.2. Earlandia arnplirnuralis (Pantic), samplc: H54, stratigraphic section 3. 6.3. Earlandia tinrinniforrnis (Misik), sample: GL342, stratigraphic section 4. Fig. 6.4. Duostominidae: ?Variostorna SP., sample: H128, stratiaraphic section 5. Fig. 6.5, 6.6. Ophrhalrnidiurn spp., sample: H55, stratigraphic section 3.
-. the Pelsonian (dating based on associated conodonts); this conclusion is confirmed by other authors (i.e. Premoli Silva, 1971;Trifonova, 1970;Oravecz-Scheffer, 1987). The stratigraphic ranges of the main foraminifera collected in Hydra are shown in Table 2; these results are integrated with data already known from the literature. We d o not propose any new foraminifera1 zonation but our data has been compared with the zonations proposed by Zaninetti et al. (1972a), Salaj et al. (19X8), and Trifonova (1992). In the schemes of Salaj et al. (198X) and Trifonova (1992), a subdivision of the Scythian consisting of a lower, Meandrospira cheni Interval Zone and an overlying, Meandrospiru pusilla Interval Zone, was proposed. In Hydra, M. cheni appears after the acme of M . pusilla in association with rare specimens of M . pusilla, disappearing at the end of the Scythian, whereas M . pusilla continues into the Anisian in association with M . dinarica.
In terms of systematics, it is noteworthy that we refer to M . pusilla the smallest forms belonging t o the genus Meandrospiru of the type-material from China described by Ho (1959). maintaining the name pusilla according to the  (1992) basically agrees with this zonation but differs in the detail of her subdivisions o f the Pelsonian. The M. clinaricu Subzont: was raised to a lull Zone by Zaninetti et al. (1972a).
In Hydra, M. dinarica appears at the base of the Anisian and continues to the top of the Pelsonian, whereas P. densa appears during [he Pelsonian. According to our data, the range of M. dinarica is therefore lower compared to previous zonations. Furthermore, we have found M. dinarica in associa tion with Aulotortus or Triadodisciis nsp. ( = Involiitina siriuosa prugsoides sensu Bronnimann et al., 1973a).

Palaeogeographic considerations
Palaeogeo6raphi'cally, there is a particular similarity between the foraminiferal fauna of Hydra and that recorded by BrGnnirnann t't a/. (1973a,b) from the Anisian of southern Bosnia-Herzegovina, as well as in other localities of the Dinarides (Kochansky-DevidC 6i Pantic, 1966;Urosevic, IY77. 198 I. 19,K8). Lithological and faunal similarities suggest a possible correlation between the Dinarides and the Subpelagonian :Zone of the Internal Hellenides (Hydra Island).
In the M. Facito Formation auct. of the Lagonegro area (Southern Apennines), an Illyrian-Landinian foraminiferal fauna, calibrated with conodonts (Ciarapica er u/., 1990;Mietto et a/., 1991), has been studied. This foraminifera1 assemblage is similar to the Illyrian fauna recorded in Hydra; particularly they share, amongst others, the species Turriglornina magna, T. mesotriasica and Endothyranella wirzi. During the Pelsonian the carbonate platform of Hydra was characterized by the occurrence of Meandrospira dinurica, Pilammina densa, Palaeomiliolina judicariensis, and TriadodiscuslAulotortus nsp. This fauna is completely absent in the Lagonegro sequence: in fact, during this period Lagonegro is affected by terrigenous influxes, causing carbonate sedimentation to occur only from lllyrian time.
In conclusion, the two successions of Hydra and Lagonegro show affinities, testifying to a similar evolution of the passive continental margin, even if they were developed at slightly different times. In Hydra, the carbonate platform (Eros Limestone) which had developed from Scythian to Pelsonian time, becoming drowned during the Pelsonian, whereas the carbonate platform of the Lagonegro area is referable to the Illyrian-Longobardian and became drowned at several different times within the Ladinian.

SYSTEMATIC DESCRIPTIONS
Citations. post -1976, are given in full to show the wide distribution within Tethys of these important foraminifera and as a supplement to the detailed synonymies given by Zaninetti (1976). The reader is referred to Tables 1 and 2 for the distribution of the species in Hydra. For the position of the samples and for the lithologies quoted in the text below, see also Figs 4, 5.
G. grandis has been recorded in the Anisian throughout Tethys, often associated with Pilarnrnina densa and Meandrospira dinarica. The species was named, but not figured, by Ramovs (1972) from the Carnian of Dinarids.

Remarks.
The specimens from Hydra are comparable with Pantic's original material, except for the high number of convolutions in the latter (40-60 in Pantic, 1965). P. densa is known from the Anisian of many Tethyan localities (i.e. Dinarides, Hellenides, W. Carpathians, Giudicarie Alps, Northern Calcareous Alps, Swiss Prealps, Kocaeli Peninsula (Turkey), Qinghai and Guizhou provinces of China). The species is frequently mentioned in association with Meandrospira dinarica and Glornospirella grandis. We have found P. densa, together with the same Involutinacea (Triadodiscidae or Aulotortidae) recorded by Bronnimann et al. (1973a,b) and assigned by these authors to Inuolutina sinuosa pragsoides (Oberhauser), but we have never found P. densa either with M . dinarica or Glomospirella grandis.

Pilammina praedensa
Pilarnrnina praedensa is known from the Scythian of the Inner Belt of eastern Serbia. The age of P. praedensa is confirmed in Hydra by the presence of Spathian bivalves (Unionites sp. and Eurnorphotis sp.) and Scythian conodonts ( N . homeri).
In our opinion, the species E. malayensis Gazdzicki & Smit, from the (probably Ladinian) Kodiang Limestone (Kedah, North West Malaysia) must be synonymized with E. salaji as it is not possible to recognize significant morphological, dimensional and stratigraphical differences to justify a separate species.
As already pointed out by Koehn-Zaninetti (1969), the presence of tectum and diaphanotheca in the wall of Triassic endothyras is never clear; this is related to an evolutionary simplification of the wall or to diagnetic processes. Vachard & Razgallah (1988) suggested that most of the Triassic endothyras and endothyranellas are directly linked with their new genus Endoteba, described from the Permian of Jebel Tebaga, Tunisia.
Endothyra salaji was recorded by Gazdzicki et al. (1975) in the Fassanian Muschelkalk of Poland; Oravecz-Scheffer (1987) found, but did not figure the species in the Anisian Falsoor Limestone of Hungary; E. salaji also occurs in the Pelsonian-Illyrian of the West Carpathians  and possibly also in the Carnian-Norian of the same region ; it was also recorded in the Middle Triassic of Guizhou (China) (He, 1984). In Hydra, E. aff. E. salaji is Early Anisian to Pelsonian in age. Occurrence. 0. Purgos, Eros Limestone s.s., section 5, samples H 130, G L 65.

Remarks.
Only one specimen, an equatorial cross-section, corresponds to the morphology of the genus Endoteba as described by Vachard & Razgallah, 1988 (type species Endoteba controuersa, found in the Permian of Jebel Tebaga, Tunisia). Previously, Endoteha was known only from the Permian, but here it is recorded in the Triassic. However, Vachard & Razgallah suggest that many Triassic endothyras are morphologically very close to Endoteba and propose a phylogenetic trend between the two genera.
Remarks. All the specimens assigned to E. wirzi have characteristics comparable with those mentioned in the original description by Koehn-Zaninetti (1969), although it was not always possible to observe the coiled stage. The globular embracing chambers, the terminal aperture produced on a neck in the uniserial part and the thin microgranular or finely agglutinating wall are diagnostic characteristics of the species.
The specimen figures in our PI. 2, fig. 12 and others from the literature show a change of coiling plane in the spiral part, this character confirming the attribution of the species to Endothyranella Galloway & Harlton, rather than to Arnmobaculites Cushman, as originally described.
Endothyranella wirzi is known in the Tethyan Middle Triassic, occurring in the Anisian (Koehn- Zaninetti, 1969) and in the Ladinian (Zaninetti, 1976). In Hydra we have recorded the species in the Early Anisian (as high as the Pelsonian  Loeblich & Tappan. 1946 Meundrospira pusilla  (PI. 1, fig. 1 1990 Meandrospira cheni (Ho): Baroz et a/.: 28,pl. 4,figs Remarks. In association with rare specimens of M . pusilla in the Scythian of Hydra, we have recorded a few specimens distinguishable from M. pusilla on the basis of a reduced number of zigzag bends, together with a smaller size of test and deuteroloculus. Morphologically these specimens are comparable with part of the type-material of Meandrospira cheni (= Trochamminoides cheni  from China. This species has long been considered a synonym of M. pusillu. Recently, Baroz et a/. (1990) asserted the validity of M . cheni, recorded together with M. pidsillu, in the Lower Triassic of the Oreokastro mountain belt (Greece) and we agree with them in considering M. cheni a transitional form between M . pusillu and M . dinarica. A phylogenetic lineage within this group had already been commented upon by Farabegoli et ul. (1976) and by Oravecz-Scheffer (1987). These authors proposed, however, a link through the informal taxon 'Meandrospira forma gigantea' of Farabegoli et (11. (1976), recorded in association with both M. pusilla and M. rfinarica. As already noted by Baroz et al. (1990), Farabegoli et a/.'s specimens of ' M . forma gigantea', found in association with M . pusilla, show affinities with M. cheni, whereas those associated with M . dinarica fall within the intraspecific variation of M. dinarica. Specimens of ' M . gigantea' figured also by Oravecz-Scheffer (1987) (pl. 13, figs 1-4) seem to belong to dinarica.
In conclusion, we can confirm that in Hydra M. cheni occurs together with rare specimens of M . pusilla at the base of the Eros Limestone s.s., in the Scythian (probably Spathian), and it is never found above this level. In contrast, M. pusillu is present also in the later, Pelsonian portion of Eros Limestone (upper lithozone) together with M. dinarica. Occurrence. 0. Purgos, Eros Limestone s.s., section 5, sample H 105.  1966: 27). The same specimen was considered by Salaj et al. (1983) to be a new species belonging to the genus Meandrospiranellu Salaj, but they did not elaborate further in the text. In our view, this specimen is an axial cross-section of M . dinarica whose dimensions and morphology can be considered as intraspecific variation.

Meandrospira dinarica
M . dinaricu has been recorded in the Anisian of all Tethyan regions. In Hydra, M . dinarica occurs in the Early Anisian as high as the Pelsonian. Remarks. In sample G L 334 we have found some specimens consisting of a globular proloculus and an undivided, tubular second chamber irregularly coiled; the wall is porcelaneous. These forms are very close to those figured by Trifonova (1972, pl. 2, figs 4, 5) and Salaj et a/. (1983, pl. 54, figs 1-6) and included in the taxon Meandrospira deformata Salaj, 1967 by them. Similar specimens have also been identified by Gazdzicki et al. (1975, pl. 7, figs 9-16) as Meandrospira? deformata Salaj.
In our opinion all these specimens, including those from Hydra, d o not correspond in shape and dimension to the holotype of Meandrospira deformata as figured originally by Salaj et al. (1967, pl. 2, fig. 3d) and by Salaj et al. (1983, pl. 53, fig. 4 Salaj et al., 1967) in having a larger test (0.55-0.6 mm) and deuterolocular (0.055-0.065 mm) diameter (cf. also type material figured by Salaj, Borza & Samuel, 1983, pl. 57, figs 1-5 (Urosevic, 1981) [ = G'lomoturitellellu rnagnu (Urosevic, 1981(Urosevic, , 1988, originally recorded from the Carnian-Norian? limestones of the Balkans. Urosevic (1981) claimed to have observed the existence of dimorphism in this species, but the morphological differences of the two supposed generations are not clear. Neither we nor Zaninetti et a/. (1990) could recognize dimorphism in our material, but can confirm the occurrence of the species in the Middle Triassic of Hydra and the Southern Apennines (M. Facito Formation auct.) Trifonova (1979) referred this species to the genus Palueomiliolina Antonova, 1959, because her new analysis demonstrated that it has chambers with a quinqueloculine coiling which sometimes have a tendency to a sigmoiline arrangement in the last whorls. The chamber arrangement of our specimens is in agreement with the generic placement of judicariensis within Palaeomiliolina as proposed by Trifonova (1979). According to Premoli Silva (1971), P. judicariensis appears in the Pelsonian at about the same time as the disappearance of Meandrospira (Pl. 3,figs 5 ,9) figs 1, 2, 3, 5, 6.
The calcareous hyaline wall of T. prealpina is not well preserved in our material, because of the presence of an encrusted surface.