On Trochamijiella gollesstanehi, gen. nov et sp. nov. (Foraminiferida, Loftusiacea), an index for the Middle Eastern marine Late Bathonian

The new taxon Trochamijiella gollesstanehi is described; it is morphologically similar to, but phylogenetically distinct from, Amijiella, and is initially trochospiral; it is known from Iran, the United Arab Emirates and Oman, and is believed to characterise the Late Bathonian. Primary type specimens of Amijiella amiji and Haurania deserta are refigured for comparison and are believed to characterise Early Bathonian and older Jurassic strata. All three genera are reclassified and all are placed in the Biokovinidae Gušić, 1977.


I N T R O D U C T I O N
In the early 1960s, when micropalaeontological biostratigraphy was first being applied to the Jurassic sequences of shelf limestones being drilled off-shore of Abu-Dhabi, U.A.E., it was realised that many foraminifera, readily recognisable in thin-section and biostratigraphically valuable, were not identifiable with any known, published taxa. They were recorded by BP Exploration Co laboratory staff (including FTB) but were not formally named or described in publication. Contemporaneously, Ali Gollesstaneh was researching at University College London the micropalaeontological stratigraphy of the Jurassic-Cretaceous Kharni Group of Fars Province, Iran. One of us (lTB) examined his slides with him, while his work was in progress. It was recognised that one of the Jurassic Khami Group foraminifera was the same as one of the Jurassic Araej Formation foraminifera of off-shore Abu Dhabi; it was one of those which was potentially useful stratigraphically but it had no name. In consequence, Gollesstaneh named it "Zranica sliiigeri" in his ensuing thesis, but it remained unpublished until Gollesstanehs (1974) review of the Khami Group biostratigraphy; unfortunately, the species, although mentioned, was then undescribed, so it became 1ioriieri riiidiim. Following the revolution in Iran, Gollesstaneh's work in that country ceased. In this paper, we formally describe, name and classify the taxon and indicate its biostratigraphical value as a Bathonian index in the drilled sequences of marine, shelf limestones of the Middle East. Foraminiferida Eichwald, 1830Suborder Textulariina Delage & Herouard, 1896Superfamily Biokovinacea GuSiC, 1977Family Biokovinidae GuSiC, 1977 Trochamijiella gen. nov. Type species. Trocharnijiella gollesstanehi sp. nov.
Diagnosis. Test calc-agglutinating, septate, uniserial, initially trochospiral, later uncoiling and becoming rectilinear, with chambers approximately circular in section perpendicular to the long axis; in the coiled test, the aperture is initially interiomarginal and single, but it later becomes areal, multiple and cribrate; in the uncoiled, rectilinear test, the multiple, pore-like apertures are confined to the central, median part of each septum; the rectilinear chambers marginally possess vertical partitions which are arranged radially and extend inwards, from the lateral chamber walls, for about one-quarter to one-half of the chamber diameter (the apertural area of each septum overlaps with the innermost extent of many of these radial partitions); the radial, vertical partitions are usually simple and rarely bifurcate; the wall of the test is solid and is not known to become protocanaliculate or canaliculate.
amiji show the presence of centrally-situated pillars, between the apertures of successive septa (Banner, Gillmore et a/., 1991), whereasnosuchpillarsarefirmly known from transverse sections of Trocharnijiella.
Although the primary type specimens of Haurania deserta Henson (eg., PI. 2, figs 5,7,8) d o not show it, hypotypes (e.g. from Morocco, in Hottinger, 1967, P1. 8, fig. 7) which undoubtedly belong to that genus possess very numerous radial partitions which bifurcate along the long axis of the test, producing at least two stacks of radial partitions in each chamber. Vertical bifurcation of the radial partitions is rare in both Amijiella and Trochamijiella.
Classification. Following Loeblich & Tappan (1985, both Arnijiella and Hauruizia belong to the Loftusiacea (a superfamily of the suborder Textulariina), but the former genus may be referred to the Cyclamminidae Marie (subfamily Choffatellinae Maync) while the latter genus should be placed in the Spirocyclinidae Munier-Chalmas. Although this may correctly imply that the two genera have different palaeobiological close affinities, we cannot accept the conclusions. As noted above, both genera may have centrally-situated, although sparse, interseptal pillars, and Ainijielln lacks the "continuous alveolar hypodermis" believed by Loeblich & Tappan (1988, p.101) to characterise the Choffatellinae, or the "inner alveolar layer" considered by these authors (1988. p.98) to characterise the Cyclamminidae asa whole. Conversely, thespirocyclinidae was said (Loeblich & Tappan, 1988, p.106) to be "planispiral to irregularly planispiral, becoming peneropliform to cyclic" and this hardly character ises Hn 11 ra I I ia . Simi 1 ar 1 y , Troclin in ijir[ In has characteristics which agree with those of neither family, as they were defined by Loeblich & Tappan (1988).
We consider that the genera Hniimiiin, Ainijirlla and Trochnrnijielln may be easily confused; a classification which groups them together makes their differences more apparent. Noneof these genera has an alveolar hypodermis, yet each has a test which is dominantly uniserial; these characteristics should exclude them immediately from the Cyclamminidae and the Spirocyclinidae. They are all very similar to the genus Biokovina Gusic (Loeblich & Tappan, 1988: p.91, P1.8, Figs 1-6), and differ principally in not being canaliculate. As has been pointed out elsewhere (Banner, Simmons & Whittaker, 1991) canaliculation can develop only in the later parts of some tests of some genera of the Chrysalidinidae, but may characterise different genera in the Textulariidae. Therefore, even within one superfamily (e.g. the Textulariacea), morphocharacters such as canaliculi can have different importance in different subgroups. In this case, we consider that the presence or absence of canaliculation may characterise certain genera which otherwise are all essentially morphologically similar; they are all calc-agglutinating, with no alveolar hypodermis, become uncoiled and rectilinear in adult ontogeny, but have radial partitions inside their adult chambers and an areal (often cribrate) aperture. We would place them all in the Biokovinidae GuSic, 1977 (modifying the definition given by Loeblich & Tappan, 1988, p.91). Incidentally, all of these four genera are of Jurassic age and could have had some palaeobiological affinity, even though it is probable that the immediate ancestors of each were distinct. Material. Many hundreds of specimens in random thin section from the same stratigraphic levels and the same area of provenance as the holotype; also other specimens from Wadi Milaha and other areas of Oman. Paratypes deposited in the British Museum (Natural History) are in six limestone thin-sections registered as P. 52652-52657. Description. The initial trochospire consists of about three whorls of low chambers, eight or nine in each whorl; the septa arecurved and the septa1 apertures areinitially interiomarginal and single but become areal and multiple in the last half of the last whorl. The uniserial, rectilinear part of the test has an initial breadth only slightly less than the maximum diameter of the last whorl of the initial coil, but the succeeding rectilinear chambers slowly increase in breadth and height. Their sutures are weakly or not at all depressed. The uniserial chambers are virtually circular in transverse section; they are five or six times as broad as high, and their height is only about 150% greater than the thickness of the septa which separate them.  Figs 1-7,9-ll,and 13-15areof specimensfrom the Unim Shaif field, and aredeposited in theBritis11 Museum (Natural History); the other specimen (Figs 8, 12) is from the Zakum field, also off-shore Abu Dhabi, and is deposited in the BP collection, Natural History Museum, London.  Fig. 1 shows the last whorl o f the initial trochospire, in near-equatorial section, follcnved by the mcdian part of the rectilinear chambers and finally by their more peripheral parts with the radial partitions. The off-centred axial section of the initial trochosphirc is seen i n Fig. 8 (enlarged in Fig. 12); a more off-centred, near-axial section is seen at the base of Fig. 10. Even when the initial spirc cannot be seen in random section, the species is rccognisable and is clearly distinct from other taxa (scc Plate 2). become corrugated where the radial partitions form (Text- fig.  2 and PI. 1, figs 1,3,4,8,12). There are about 20 radial partitions in the perimeters of each of the later chambers, and these partitions are of variable length (Pl. 1, figs 2, 5, 13), some reaching to the middle zone of each chamber but others remaining only at the perimeter; occasionally, these partitions bifurcate sporadically to develop incomplete double-stacks of partitions (Pl. 1,figs 1,lO). Theseptal aperturesof the rectilinear chambers are multiple and areal (PI. 1,figs 1,4,5,6,8,10), in the median zone of each septum. Dimensions. Length of holotype (Pl. 1, fig. 1): 1.4mm. Paratypes have greatest observed length of 2.8mm (Pl. 1, fig. 4). Greatest observed breadth of uniserial chambers is 0.9mm (Pl. 1, Fig.  13). Locality and horizon of holotype and paratypes. Offshore Abu Dhabi, U.A.E., subsurface, in the middle and lower parts of the Uweinat Formation, Bathonian. The holotype and the paratypes whicharedeposited in the British Museum (Natural History) were all obtained from the Umm Shaif field.

STRATIGRAPHY
Trochaniijiella gollesstaizehi was first recorded by Gollesstaneh (1965Gollesstaneh ( , 1974, under the iiorneii tzudurn "lranica slingeri", from the limestones of the Khami Group in Fars Province, southern Iran. He noted (1974, p.187) that it was an excellent marker for his "Zone VIII", the Pfendrina Zone, which he also found to contain P. trochoidea Smout & Sugden, P. salernitarra Sartoni & Crescenti and other microfossils. He considered (1974, p.188) that the zone was probably Bathonian in its lower part and Callovian in its upper part. "lranica slingeri" was not mentioned by Kalantari (1986) in his review of the microfacies of Iranian limestones, but this author illustrated it (1986, p.113, P1. 51.1) from the Surmeh Formation, Ahmadi anticline, of the Interior Fars area; it was dated as Bathonian, but misnamed. "Haurania amiji".
T. gollesstanehi has often been recorded b y BP biostratigraphers (in unpublished internal reports) as occurring in the middle and lower parts of the Uweinat Member of the Araej Formation, limestones drilled subsurface in off-shore and on-shore Abu Dhabi, U.A.E. Here, T. gollesstanehi occurs with Pfenderiiia trochoiden and P. salernitnna, and it has a similar microfossil association in equivalent strata of Oman.
On the other hand, Haurania deserta Henson and Amijiella amiji (Henson) (also referred to Haurania by Henson, 19481, in the U.A.E. -Oman area, occur in beds correlated with the Lower Araej Formation, stratigraphically below the Uweinat Member (and the youngest Trochamijiella gollesstanehi). The type specimens of both these species were obtained by Henson (1948) from the same depth in the Wadi Amij government water-well of western central Iraq, from bedsoriginally thought to belong to the Mulussa limestone, a formation which is possibly as old as Triassic, in part at least. Consequently, Henson (1948) considered the types of these species to be of "Jurassic or Triassic age". Later (Dunnington et a/., 1959, pp. 192-194) referred these same beds, and their microfossils, to the Bathonian Muhaiwir Formation, which has its type locality in Wadi Hauran, on the road to Qasr Amij, west Iraq, and is probably early Bathonian at youngest (it may extend down into the Bajocian). The U.A.E. -Oman occurrences are probably also of this age. Kalantari (1986, pp. 112-1 13) has also retrieved H. deserta from the Bathonian Surmeh Formation of Iran. Both H . deserta and A. amiji have also been reported to occur in Italy in beds as young as the Bathonian (Sartorio & Venturini, 1988, p.66) even although they also occur in older Jurassic limestones. We believe that the stratigraphically highest Haurania deserta and Amijiella ainiji are useful zonal markers, in the marine, shelf carbonate rocks of the Middle East at least, for beds of early Bathonian age.
Consequently, we consider that Trochamijiella gollesstanehi and the Haurania deserta -Amijiella ainiji association enable ready biostratigraphic recognition of two successive zones in Middle Eastern marine, shelf limestones, denoting the late Bathonian and the early Bathonian-Bajocian, respectively.