A new species of Somalina (Somalina hottingeri) with partially vacuolate lateral walls from the Middle Eocene of Oman

Somalina hottingeri, a new species with partially vacuolate lateral walls is described from the Middle Eocene of Oman. It is distinguished from the only previously recorded species with this wall structure (S. transitorius (Hottinger)) by having equatorial chambers that are relatively low throughout the test. On the basis of this character, it is suggested that S. hottingeri evolved from the Opertorbitolites douvillei Group (redefined here) and that it gave rise to the true somalines. S. transitorius is regarded as arising from O. latimarginalis (ex. O. latimarginalis Group, introduced in this study) but not to have given rise to any other species. Since forms of Somalina with only partially vacuolate walls appear to be confined to the late Early Eocene to early Middle Eocene, it is concluded that the presence of this wall structure provides a useful stratigraphic marker.

Opertorbitolites sensu stricto first appeared in the Late Palaeocene (Alveolina cucumiformis Zone, see Opertorbitolites gracilis (Lehmann), in Lehmann, 1961) and Somalina sensu stricto is not known from rocks older than the Early Eocene (probably late Alveolina violae Zone). Since the oldest known occurrence of individuals with an intermediate morphology is also Early Eocene (A. trempina Zone ;Hottinger & Krusat, 1972), they undoubtedly represent a transitional evolutionary stage. Hottinger (in Hottinger & Krusat, 1972) assigned the only previously recorded species with partially vacuolate lateral walls to Opertorbitolites transitorius (PI. 2, fig. 3). However, vacuolate lateral walls are characteristic of Somalina, to which genus these transitional forms are here assigned.

SYSTEMATIC DESCRIPTION
The classification of Loeblich & Tappan (1987) is followed and localities mentioned in the text are shown on Fig. 1. Range charts for each of these localities are given in White (1994), to which the reader is referred for more stratigraphical information. All specimens examined are deposited in the palaeontological collections of the Natural History Museum, London.    (F 12, 16, 17, 18, 19, 20); Jebel Hafit (HN 1157(HN , 1155(HN , 1154; Wadi Bani Khalid (WB 9, WBK 2,4, 5, 10); Wadi Rusayl (WR 33a,46,49a,51,52,56,59,60) and Wadi Tiwi (HN 1189, 1188). Type description. Megalospheric test medium to large, lenticular, occasionally slightly to strongly undulate. Surface smooth. Embryonic apparatus comprising a small proloculus and a flexostyle. Post embryonic chambers arcuate, not increasing in height significantly from the centre to the periphery of the test. The lateral walls attain a thickness of up to 0.8mm over the umbo and are partially vacuolate.
Microspheric test of medium size with partially vacuolate lateral walls (see Remarks below  fig. l), occurs together with specimens that also appear to belong to this generation (e.g. P1. 2, fig. 2). These have smooth surfaces and post embryonic chambers that do not increase significantly in height from the centre to the periphery of the test. Microspheric and megalospheric forms have not been recorded together. However, the vacuolate nature of the lateral wall, and the low chambers apparently present in both generations leaves little doubt that they belong to the same species.
Although S. hottingeri is characterized by having partially vacuolate walls, this feature is best seen in tangential or near equatorial sections (e.g. P1. 1, fig. 5; P1. 2, fig. 1 White, 1994, figs 14.7, 14.9, 14.11, 14.12 and 14.15). Lehmann (1961) distinguished two groups of Opertorbitolites with relatively low equatorial chambers throughout the test, and separated 0. latimarginalis (Lehmann), in which the equatorial chambers increase significantly in height towards the periphery. He regarded biplanar forms with only slightly thickened lateral walls as belonging to his biplanus Group and placed those with lenticular tests and strongly thickened lateral walls in that of 0. douvillei. Forms with relatively unthickened lateral walls, however, may also be lenticular (e.g. 0. sp. cf. 0. gracilis Lehmann,P1. 2, fig. 4), and those with thick lateral walls are not always so (e.g. 0. sp. cf. 0. douvillei, P1. 2, fig. 5). Consequently, the present author does not recognize the distinction between these two groups. Instead, the 0. douvillei Group of Lehmann is redefined here to comprise species (including those originally placed by Lehmann in his douvillei and biplanus groups) characterized by relatively low equatorial chambers. The 0. latimarginalis Group is introduced for those species with equatorial chambers that increase significantly in height towards the periphery of the test.

DISCUSSION
The low equatorial chambers in S. hottingeri, combined with its partially vacuolate lateral walls and its first appearence in the late Early Eocene strongly suggest that it evolved from the 0. douvillei Group, probably 0. douvillei itself which is known from the Laki Series (Early Eocene) in Pakistan (Nuttall, 1925). In contrast, the relatively rapid increase in height of the equatorial chambers in S. transitorius implies that this arose from 0. latimarginalis (Alveolina trempina Zone, see Lehmann,196 l), the only known species belonging to the 0. latimarginalis Group.
From the above, it seems certain that S. hottingeri and S. transitorius belong to separate lineages (Fig. 2). Of these, the latimarginalis-transitorius line does not appear to have evolved horringeri and Somalina sensu stricto. Vertical solid and dashed lines denote definite and inferred ranges, respectively. Both S . hottingeri and Somalina S.S. are known from the Early Eocene. Although the zone has not been identified, they must at least range into that of Alveolina violae. The A . corbarica occurrence of the 0. douvillei Group is that of 0. biplanus, recorded by Lehmann (1961). The range of S. transitorius is taken from Hottinger & Krusat (1972), while that of the 0. latimarginalis Group (is. 0. latimarginalis) is from Lehmann (1961). further. On the other hand, the close similarity between S. hottingeri and Somalina sensu stricto (particularly their low equatorial chambers) leaves little doubt that they belong to a single evolutionary lineage. The later appearance of 0. latimarginalis compared with the 0. douvillei Group is tentatively interpreted as the former having evolved from the latter. It is of course possible that the two Opertorbitolites groups have separate origins.
The relatively short combined stratigraphical range of S. transitorius and S. hottingeri (i.e. Alveolina trempina --early Alveolina stipes zones) makes their characteristic wall structure a useful age indicator even when the species cannot be identified (e.g. from fragments in thin section).