On Everticyclammina Redmond (Foraminifera), especially E. kelleri (Henson)

Everticyclammina kelleri (Henson) (= E. eccentrica Redmond and E. elegans Redmond), a Berriasian-Valanginian index in the Middle East, is descended from the Late Jurassic Tethyan E. praekelleri sp.nov, a species which probably had its ancestry in Late Jurassic Ammobaculites sp. From this ancestor also evolved E. virguliana, and, in the Middle East, its Early Cretaceous descendants E. hensoni Redmond, E. contorta Redmond and E. greigi (Henson), which themselves formed the ancestry of the Albian Hemicyclammina whitei (Henson) and the Tethyan Albian - Cenomanian index H. sigali (Maync). Another, independent lineage from Ammobaculites, Buccicrenata hedbergi (Maync) (= B. libyca Gohrbandt) to B. subgoodlandensis (Vanderpool), transglobal in the Barremian - Cenomanian, is distinguished.


INTRODUCTION
The type specimens of Everticyclammina kelleri (Henson) In his review of the Mesozoic lituolids distinguished by the Iraq Petroleum Co., its Chief Palaeontologist, F.R.S. Henson, described the new species Pseudocyclammina kelleri (Henson, 1948, p. 16, pl. 9, figs. 4 , 5 and 7), illustrating three of the 'many hundreds' of syntypes used by the company to define the species. All of the syntypes were random thin-sections, in limestone drilled by the British Oil Development Co. (later assimilated into the Mosul Petroleum Co., which was associated with I.P.C.) at 4726-4741 ft depth in the Awasil No. 5 Well, about 38 km west of Ramadi (and about 140 km west of Baghdad), Iraq. Henson (1948) originally believed that this limestone was of Argovian-Callovian age, but this age-determination was later revised. The beds in Awasil No. 5 which contained P. kelleri were firmly referred, by Wetzel and Dunnington, to the upper part of the Chia Gara formation, Berriasian, correlatable with the uppermost six beds of the stratotype Chia Gara of Kurdistan, the outcrop south of Amadia, north Iraq, which also contain P. kelleri (Dunnington ef ul., 1959, pp. 72-77). The rich ammonite faunas of the type Chia Gara show that this formation straddles the Jurassic-Cretaceous boundary, ranging in age from Middle Tithonian to Berriasian; the appearance and frequency of record of P . kellerr in the Berriasian part of the formation (loc. cit.) would suggest that potentially it is a basal Cretaceous index species. "It is now believed that this foraminifer is limited to rocks of Berriasian and very early Valanginian age in northern Iraq" (op.cit., p.76) and subsequent records, discussed below, confirm that this is true for at least 200 km to the south-east, to Saudi Arabia (Redmond, 1964) and adjacent areas. This is in spite of the Fact that the original description ofP. kelleri given by Henson ( 1948) was very limited by modem standards. All that was written was "Test thick-walled, arenaceous, with much cement; sub-epiderniul luver largely obscured; labyrinthic layer pronounced, with scattered sandgrains; the spiral part of the test is involute, lenticular with two whoi.Is and with twelve to rhirteen chambers in the last whorl; the serial part of the test is somewhat elliptical in section with up to four chambers." The three syntypes which were photographed (Hensoh, 1948, pl. 9) consisted of one-near equatorial section (pl. 9, fig.  5) of what may now be regarded probably as a microspheric specimen, one subaxial section of an uncoiling (and possibly also microspheric) specimen (pl. 9, fig. 4), and one slightly oblique, axial section of amegalospheric specimen (pl. 9, fig.  7). The thin-sections of limestone in which these figured syntypes occur (registered in the British Museum (Natural History), numbers P. 35968, P 35967 and P. 35969 respectively) contain many other syntypic specimens, including equatorial sections of megalospheric examples, which are figured here (Plate 1, figs. I , 2) for the first time. Almost all these syntypes (including those originally figured) are pyritised. In many, the wall-structure is not clear; the 'labyrinthic 1ayer"may be particularly obscure. Some of the specimens do not possess a "labyrinthic" wall at all, and may belong to the genera Ammobaculites, Lituola or similar forms. Others, like the syntypes figured by Henson (1 948), have narrowly alveolar walls; these and others have been photographed (Plates 1 , 2 ) and from them a lectotype is chosen (below). This restudy of the syntypes has enabled revision of the taxonomy and has also permitted an emendation of the description and diagnosis, so that the species may more readily and confidently be identified as a basal Cretaceous biostratigraphic index.

The species of Everticyclammina Redmond
The genus Everticyclammina was proposed primarily to accommodate a group of species which had been used by the micropalaeontologists of the Arabian American Oil Company (Aramco) in their stratigraphy of the Early Cretaceous of Saudi Arabia, and which could not satisfactorily be referred to Pseudocyclammina. However, the original description of the genus Everticyclammina given by Redmond (1964, pp. 407-8) was, and is, difficult to cqmprehend precisely. It was interpreted by Maync (1 965) to "include all those transitional forms of 'pseudocyclamminid' Ammohaculites" in which the septa were "simple, not pierced", but he did not clarify the nature of the aperture, although this has become one of the premier diagnostic characters of the genus.
The originally designated type species of Everticyclammina was E . hensoni Redmond, 1964; the sectioned paratypes of this species (Redmond, 1964, pl. 1 , figs. 22-25) show that it possessed a narrowly but regularly alveolar chamber wall but solid, non-alveolar septa, and that its aperture was simple, single, areal and "ammobaculitid". The primary types came from the Buwaib Formation drilled in Aramco well Dammam-16, Saudi Arabia (Redmond, 1964, p. 409), which was dated by Aramco as of Hauterivian age (Powers, 1968, p. 53) although it contains no fossils other than benthic foraminifera and is overlain by the totally unfossiliferous Biyadh Sandstone (Powers,1oc.cit.). The Buwaib Formation is currently regarded as being of Late Hauterivian to Early Barremian age, the Early Hauterivian being missing in its disconformity with the underlying, Valanginian Yamama Formation (Hughes Clarke, 1988). The type specimens of E. hensoni must also be considered to have belonged to the Late Hauterivian or Early Barremian or both.
The late Hauterivian/Early Barremian Buwaib Formation, drilled in the Aramco Saudi Arabian well Abqaiq-62, yielded the primary types of E. contorta Redmond (1964, pl. 1, figs. 12-15). Hottinger (1 967) illustrated equatorially-sectioned topotypes (1967, pl. 9, figs. 17-18), which he believed to be synonymous with E. greigi. However, this species is clearly distinguishable from the contemporaneous E . hensoni and the younger E. greigi, using the equatorial sections of all of them. In E. hensoni (Plate 3, fig. 6), not only are the lower parts of the septa separated, meeting the preceding spiral suture nearly perpendicularly, but the upper part of the septa meet the chamber-peripheries also almost at right-angles; consequently, in equatorial section the adult chambers are of quadrilateral shape, almost parallelograms. In E. contorta (Plate 3, fig. 3), the lower parts of the septa, while still separated, meet the preceding spiral suture at about 45" and the chamber peripheries at a similar angle; the chambers are reniform in equatorial section. In E. greigi, as noted above, the lower parts of the septa are coalescent, being thickened on meeting the preceding spiral suture tangentially, while the upper parts of the septa meet the chamber-peripheries obliquely (Plate 5, fig. 4, as in E. contorta).
The stratigraphic range of this group of species of Everticyclammina, in this area of the Mid-East Gulf, is now believed to be from the Berriasian-Valanginian (Upper Chia Gara and Yamama Formations and their equivalents), through Hauterivian and Barremian (Simmons & Hart, 1987, pp. 189-190) to probable Aptian (Banner, 1970). There is a clear stratigraphical succession in this plexus of forms which, when separated taxonomically, can be used both biostratigraphically and phylogenetically. For the former, it should be remembered that the taxa were originally employed stratigraphically, by both Iraq Petroleum and its associated companies and by Aramco; as Redmond wrote (1 965, p. 185), they "were selected for publication because of their importance in a zonation set up over a more than fifteen year period of work in Saudi Arabia." To regard them as wholly conspecific. as was done by Brun and Rey ( 1 9 7 3 , has no justification.

The phylogeny of Everticyclammina
The species Ever.ticyclamminu virgulianu (Koechlin), E . kelleri (Henson),E. hensoni Redmond, E. contot?a Redmond and E. greigi (Henson) are not only congeneric, they represent stages in a phylogenetic plexus of evolving communities bridging the Kimmeridgian to Aptian interval. The story of the evolution, shown on text-figure I , may be outlined as follows.
In the MiddleLate Jurassic (probably Callovian/Oxfordian), a descendant of a species of Ammohuculites developed an alveolar wall. This was the first representative of the new genus Ei~er.tic.~c.ltrninli,ltr.
In conglomeratic, sparitic limestone outcropping at Broumana (about 20 miles east of Beirut), Lebanon, are abundant specimens of Ei.ci.tic.?.(./Nniniiiiu with coarsely alveolar and irregularly labyrinthic, thick hypodermae. This deposil, of an inner neritic palaeoenvironment. is probably of Kimmeridgian-Tithonian age, for this, heavily block-faul ted area is mapped (Dubertret, 1963) as exposing thin limestone beds of this age and the succeeding Neocomian sandstones, lignites and contemporary basalts (Dubertret, 1963, pp. 38-45, 7 1-77). This species, here regarded as the new species Evertic.ychimmina pr-aekeller-i (it was regarded by Henson, 1948, p. 17, as an untypical form of "P.sc~udoc:yc~lunimitiu keileri") differs from all other known forms of Ever.ticyclammina by the large size and coarseness of both the alveolae and the dividing wall-elements of its thick hypodermis. The broad alveolae were closed in the latero-posterior parts of chamber walls, as the hypodermis there fused to the epidermis of the wall of the preceding chamber (Plate 3, fig.  5.) On the innermost surfaces of the posterior parts of the hypodermae, this produced very large alveolae, of irregular size and (polygonal) shape, with ramifying walls (Plate I , fig.   1).
Its Kimmeridgian deeper water contemporary, with narrow alveolae and a regularly labyrinthic hypodermis, E. virguliancr, ranged from Portugal (where its oldest occurrence may be at the summit of the Oxfordian, see Ramalho, 1985), through north Africa and southern Europe to the Middle East. E. virgulianu ranged from Kimmeridgian to "Portlandian" (Tithonian) and occupied a deeper-water palaeoenvironment (outer neritic to bathyal) even than the contemporaneous Alveoseptu (outer to inner neritic) (see PelissiC et ul., 1984) and one which was much deeper than that occupied by the shallow-neritic E. praeke/leri. E. vir-,quliunu was axially biconcave, with depressions at each umbilical area, and (like E. praekelleri) in equatorial section it had reniform chambers, with solid septa which met the chamber periphery and the inner spiral suture obliquely. The wall had alveoles which in the Kimmeridgian were simple, not bifurcating, and relatively broad, but by the end of the Jurassic ("Port1andian"orTithonian equivalent) had narrowed and had begun to bifurcate. The posterior hypodermal alveoles of E. vir-,quliuna were larger than the rest, and formed a distinctive row of subcircular spaces in the latero-posterior walls of each chamber (Koechlin, pl. I , fig. 7; Plate 3, fig. 4), smaller and more regularly formed than those of E. praekeller-i. The microspheric generation of E. virguliunu had 8 to I2 chambers in an adult whorl, while the megalospheric had 6 to 9 (Plate 2, fig. 5), more than in E. pi-uekelleri (Plate 1, fig. I ) In the earliest Cretaceous (Berriasian -Valanginian), the Iraqi-Arabian area saw the evolution of Ever-tic,yclummina keller-i, which, in equatorial section, was virtually indistinguishable from E. vir-guliuna, and which had much finer and regularly formed alveolae than its ancestral E. pruekelleri. The latter had a coarsely labyrinthic hypodermis; the new E. ke1ler.i had a smooth, regularly structured hypodermis. In axial section, E. kelleri had no trace of umbilical biconcavity; the umbilical areas were now very convex, and axially the test was diamond-shaped when in section (Plate 1, figs. 4-6; Plate 2, figs. 2,4). The latero-posterior alveolae of the hypodermae of the chamber-walls were very large compared to those of E . i9ir;quliunu, but smaller and more regularly formed than those 5 of the ancestral E . pt~aekelleri, and lacked undulating, ramifying margins; they formed a distinctive row of subcircular to subquadrilateral spaces at the base of each chamber, when they were cut tangentially in axial (Plate I , figs. 3a, 3b) or oblique-equatorial sections (Plate 2, figs. 3a, 3b; Plate 3, figs. I , 2). Some Valanginian representatives of this species (from the basal Yamama formation and called E. elegans by Redmond, 1964Redmond, , 1965 often were relatively small (Plate 2, fig. 4), but the later Arabian representatives (from the top of the Yamama, and called E. eccentrics by Redmond, 1964) were nearly double this size (Plate 2, fig. 2) and were similar to the Iraqi Berriasian syntypes of E. kelleri (Plate I , figs. 4-6). Like E. i~i~-g u / i u /~a , the microspheric E. kelleri had 9-12 chambers in adult whorls (Plate 2, fig. la), while the megalospheric had about 6 (Plate 1, figs. 1, 2a; Plate 2, fig. 3a). However, E.
kellori became extinct at the end of the Valanginian, and left no descendants.
The other Early Cretaceous species formed part of a continuous lineage. E. ~~otztorta (Plate 3, fig. 3), from the HauterivianEarly Barremian, closely resembled ancestral E. i~irguliunu but had reduced the number of chambers in its microspheric adult whorls to 7-8 per whorl and the alveolae were regularly and tightly bifurcating: the latero-posterior alveolae of each chamber hypodermis were no longer relatively enlarged. It is clearly intermediate betweenE. i>ft;qu/iana and its descendant, the Barremian-Aptian E. greigi, a species in which the lower parts of the adult septa become oblique, then tangential. to the spiral suture, and thicken and coalesce to form an imperforate "basal later" to the chambers (Plate 5 . fig. 4). The reduction of the distinct, lower parts of the septa, and the smoothing out of the "basal layer", produced the Albian -Cenomanian genus Heniii:vc~lunimina, in which the upper parts of the septa are thinned and meet the test periphery almost at right-angles, so that the chambers become almost quadrangular in equatorial section (as in H . Mhitei, Plate 5 , fig.  3). There was an earlier attempt to develop such quadrangular chambers but without the development of a "basal layer"; here, in the HauterivianEarly Barremian, the septa meet both the chamber periphery and the spiral suture almost perpendicularly. This was in E. heruoni (Plate 3, fig. 6). the type lamminu, which apparently also left no direct descendants. with agglutinated, scattered, fine-silt size, quartz and other exotic grains; planispirally coiled, involute, with umbilical overlap of successive chamber-walls, so that the umbilical areas become thickened and convex and the rest becomes subrhombic in axial section; megalospheric tests may consist of about 2 whorls of chambers and the microspheric tests of about 3, and the last few (up to 3 or 4) chambers may uncoil and become oval in cross-section and rectilinear in microspheric forms; megalospheric tests have 6 to 7 chambers in each adult whorl, while microspheric tests have up to about 12 or 13 before uncoiling starts; in thickness, the chamber walls are about one-quarter of the total chamber height when the test is seen in equatorial thin section; the walls of the chambers of the first whorls may be solid, but in subsequent whorls the chamber walls have an alveolar hypodermis, with anteroperipheral areas containing bifurcating alveoles, each in total about three times as long as broad and spaced apart at distances approximately equal to their diameters, but with postero-lateral areas with broadening alveoles which cease to ~~

Explanation of Plate 2
Figs. I a, I b, and 3a. 3b. Ei,c,r.fic,~c.luninrirlu kelleri (Henson), paralectotypes, in slide P. 35968, Berriasian-Valanginian limestone cored in well Awasil-5. Iraq. Figs. la, Ib, equatorial section of microspheric specimen; fig. la, x 42. 5 ; fig. 1 Redmond (1964, pl. I) . Fig. 2, "E. ccwntricu" Redmond, paratype, x 38, from Aramco well D;mman-16, 3485-90 ft, (upper?) Yamama Formation, Saudi Arabia, Valanginian. Fig. 4, "E e1egari.s" Redmond, paratype, x 38, from Aramco well Ain Dar-60,4980-82 ft, (lower?) Yamama Formation, Saudi Arabia, Valanginian. Both are synonyms o f E . kclleri (Henson). bifurcate, which increase their diameters by 4 to 5 times, yet which are still separated by thin hypodermal walls (approximately as thick as those which separate the antero-peripheral alveoles); the alveolar hypodermis is covered by a very thin, imperforate epidermis, which seals the hypodermal alveoles; the septa are curved, meeting both the chamber peripheral walls and the inner spiral suture obliquely; the septa are solid, non-alveolar, and are each pierced medially by a single, areal, suboval or slit-like aperture; in thickness, the septa may be about equal to the chamber walls, but the upper part of some septa may be thinned relative to both the adjacent chamber walls and the inner (lower) parts of the same septa. Remarks. The lectotype (like the paralectotype illustrated in Plate 2, figs. 3a, 3b) contains flecks of iron pyrites in its test wall; the other figured paralectotypes are more heavily pyritised.
The lectotype, which is an off-centred, axial, microspheric specimen, shows (Plate 1, fig. 3b) the solid wall of the first whorl and the alveolar walls of subsequent whorls, plus (Plate 1, figs. 3a, 3b) the marked and progressive broadening of the posterio-lateral alveoles of the hypodermis. These broader alveoles are also seen in the axial sections of Plate 1, fig. 4, and Plate 2, fig. 2 (which is a paratype of "E. eccentrica Redmond"); in equatorial and oblique-equatorial sections, they may be clearly seen on Plate 2, figs. 3a, 3b and Plate 3, figs 2 and 2. The thick-walled spaces on the left of the umbilical area of the subaxial section of Plate I, fig. 6, are parts of chambers of the last whorl (compare the thin-walled, broadened alveolae of Plate 1, fig. 3b); the umbilical overlap of such chambers produced the biconvex, subrhombic or "diamond-shaped" axial section, which, with the greatly broadened posteriolateral hypodermal alveolae, enable this species to be readily distinguished from all other known species of Everticyclammina.
The occasional presence of an apparently many-holed aperture ("a narrow slit or a series of pores along the bottom of a long narrow depression") noted by Redmond ( 1 964, pp. 408-9) must be due to its partial infilling with matrix in the extracted specimen; no axial, equatorial or oblique thin section is known which shows anything other than the single, areal, "ammobaculitid" aperture now generally agreed (e.g. Brun & Rey, 1975;Loeblich & Tappan, 1988) to be charac-teristic of Everticyclammina. However, the "retral-processlike parallel ridges crossing sutures" noted by Redmond (loc. cit.) may be external, epidermal covers for the enlarged posterio-lateral alveoles of the hypodermis, which are largest adjacent to the posterior intercameral sutures; only re-examination of extracted specimens can resolve this.
Everticyclammina praekelleri sp.nov. (Pl. 1, fig. 1; PI. 3 , fig. 5; P1.4, figs. 1-1 1) Derivation of Name. From its stratigraphically earlier occurrence than E. kelleri. Diagnosis. A species of Everticyclammina with enlarged postero-lateral alveolae and irregularly alveolate wall. Type Specimens. The holotype is that specimen figured here as Plate 1 , fig. 1 ; all the other specimens figured here (Plate 3, fig. 5; Plate 4, figs. 1-11) are paratypes. All are random sections in a thin-section of limestone, registered in the British Museum (Natural History) as P. 52255. Locality and Horizon. The rock-sample was collected from Broumana, near Beirut, Lebanon, from an outcrop on the road to Mar Mousa, between Baabdat and Bhannb, near a spring (Henson, 1948, p. 17), from limestones judged by macrofossils to be Kimmeridgian-Tithonian in age (Dubertret, 1963) Description. Test made of microgranular, micritic, imperforate calcite; planispirally coiled, involute, chambers touching or minimally overlapping in the umbilical regions, so that the umbilical areas are flattened or slightly convex; megalospheric tests may consist of about one to two whorls of chambers with about six in the final whorl; in thickness, the chamber walls are about one-quarter of the total chamber height when seen in equatorial section, but they appear to vary from about onethird to one-sixth depending on preservation and angle of cut; the walls of the chambers of the first whorl may be solid, but in the last whorl coarse alveolae develop irregularly in the hypodermis; in the anterio-peripheral areas of the chamber walls, the alveoles may be about as broad, or about half as broad as long, simple (not bifurcating along their length) and spaced irregularly but at distances equal to or greater than their breadth; in the postero-lateral areas of the hypodermae, the irregular alveoles become much broader and closely spaced, so that (when sectioned in the plane of the innermost Explanation of Plate 3 Figs. 1, 2. Everticyclarnmina kelleri (Henson), paralectotypes from Iraq, well Awasil-5, 4728 ft., core of Berriasian-Valanginian limestone; oblique-equatorial sections showing postero-lateral hypodermal enlarged alveolae. Fig. 1, in slice P. 52257, x 95. Fig.   2, in slide P. 35969, x 98. Fig. 3. E . contorta Redmond, equatorially sectioned, megalospheric paratype, enlarged from Redmond, 1964, (pi. 1, fig. 12), x 38; from Aramco well Abqaiq-62, core at 5263 ft., Buwaib Formation (Late HauterivianEarly Barremian), Saudi Arabia. Fig. 4. E. virguliana (Koechlin), syntype x 61, enlarged from Koechlin, 1942 (pl. 6, fig. 7) from Virgula-Mergel, la Chaux, Switzerland, Upper Kimmeridgian; the enlarged postero-lateral alveolae of the hypodermis are particularly well seen in the chamber which is fourth from the last. Fig. 5. E praekelleri spnov., obliquely off-centred paratype x 41; in slide P 52255, limestone from Broumana, near Beirut, Lebanon: showing large, irregular hypodermal alveolae, and their enlargement in the postero-lateral region with their fusion in this region to the preceding epidemis. Fig. 6. E . hensoni Redmond, equatorially sectioned, megalospheric paratype, enlarged from Redmond, 1964 (pl. 1, fig. 22), x 37.5; from Aramco well Dammam-16, ditch sample at 3315-20 ft., Buwaib Formation (Late Hauterivian/Early Barremian), Saudi Arabia. surface of the hypodermis) they possess thinner and irregularly ramifying walls (when sectioned in an outer plane, just below the epidermis, the alveolus-walls are thicker, and can be seen to fuse against the epidermis of the preceding chamber); when cut equatorially, the large, postero-lateral alveoles are subpolygonal in shape; when cut obliquely, they give the appearance of ridge-like, ramifying but subparallel structures, but when cut axially these alveoles are irregularly subquadrilateral in shape; the hypodermis is covered by an imperforate epidermis of variable thickness, which externally seals the hypodermal alveoles; the septa are curved, meeting both the chamber peripheral walls and the inner spiral suture obliquely; the septa are solid and non-alveolar, and each is pierced medially by a single, area, suboval aperture; the septa are about as thick as the chamber walls.
(b) broader, wider anterio-peripheral hypodermal alveoles, which are less regularly and more widely spaced (as in all figures), (c) bigger and more irregular postero-lateral hypodermal alveoles, which may be very variable in size and in subpolygonal shape in equatorial section (Plate I , fig. 1). and which adopt a subparallel, elongate appearance when sectioned obliquely (Plate 4,figs. 3,4). These enlarged posterior alveoles clearly arise by their fusion on to the preceding epidermis (Plate 3, fig. 5; Plate 4, fig. 11) and do not form a discrete, isolated posterior alveolae row, but are part of an irregular series of enlarged postero-lateral alveoles (Plate I , fig. 1; Plate 4, figs. 1, 3, 4,), (d) less convex umbilical areas, (e) broader, less slit-like aperture. The microspheric form has not been seen in equatorial section, but two subaxial sections (Plate 4, figs. I , 10) show more whorls than are known in megalospheric specimens and have short, uncoiling terminal growth stages; they may be sections of microspheric specimens and are strictly comparable to the lectotype of E. kelleri (Plate 1, fig. 3a, 3b).
Everticyclammina praekelleri is believed to be the Late Jurassic (Tithonian and possible Kimmeridgian) immediate ancestor of the earliest Cretaceous (Berriasian-Valanginian) E. kelleri. E. praekelleri is not yet known from the Iraqi-Arabian region, but, in those areas, the Hith and Gotnia anhydrites (and associated dolomitisation) make preservation of Late Kimmeridgian-Tithonian foraminifera very rare (Dunnington et al., 1959).
The very enlarged postero-lateral alveolae and the irregularly alveolate wall (together with the tendency towards fewer chambers per whorl) distinguish E. praekelleri from members of the lineage of E. virguliana -E. greigi.

COMPARISON WITH THE LINEAGE OF B UCCICR ENATA
In the evolution of Everticyclammina, the early species of the main lineage (Text-figure 1) ranged from Iberia to the Middle East (see, e.g., E. contorta, as figured by Brun & Rey, 1975, pl. 2, figs. 1,2,7,8 and pl. 3, figs. 1-5, from the Hauterivian of Portugal), while the younger and the side-shoot members (E. greigi, E. kelleri, E. hensoni) are, as yet, known only from the Iraq -Arabia -Oman area. With the final straightening of the upper parts of the septa and total loss of the lower parts, the evolution of the new, (Aptian?) -Albian -Cenomanian genus Hemicyclammina saw not only new species confined to the Middle East (IikeH. white;, Plate 5, fig. 3) but also the appearance of western to central Tethyan species ( H . sigali Maync, 1953b). No member of this phylogeny has yet been recorded from the Americas.
In contrast, the shorter (?Barremian -Aptian -Cenomanian) phylogeny of taxa which can be related to the genus Buccicrenata Loeblich & Tappan (1 949;emended 1985, p. 100) seems to have either ranged from America to the Middle Easter (Buccicrenata itself) or to be confined to central America (Alveocyclammina Hillebrandt,197 1 ). Like Everticyclummina, Buccicrenatu has an alveolar hypodermis and a single, areal, "ammobaculitid" septal/terminal aperture; unlike Everticyctammina, the septa are also alveolar and are extensions of the lateral chamber walls.
Like Hemicyclammina, the descendant Alveocyclammina has lost the lower part of its septa, so that the apertures are now basal, and no longer areal.