Entogoniopsis stokesiana (Greville) J. Witkowski, P.A. Sims, N.I. Strelnikova & D.M. Williams, 2015

Witkowski, Jakub, Sims, Patricia A., Strelnikova, Nina I. & Williams, David M., 2015, Entogoniopsis gen. nov. and Trilamina gen. nov. (Bacillariophyta): a survey of multipolar pseudocellate diatoms with internal costae, including comments on the genus Sheshukovia Gleser, Phytotaxa 209 (1), pp. 448-450 : 448-450

11. Entogoniopsis stokesiana (Greville) J. Witkowski, P.A. Sims, N.I. Strelnikova & D.M. Williams , comb. nov.

(SEM: Figs 149–154 ; LM: Figs 155–160 )

BASIONYM: Triceratium stokesianum Greville (1866a , Transactions of the Microscopical Society of London, New Series 14: 8, pl. II, fig. 23).

TYPE:—‘ Moron deposit, Province of Seville’ .

Triceratium stokesianum var. moravica Grunow in A.W.F. Schmidt (1886b: taf. 112, fig. 19).

TYPE:—‘Brünn’.

Biddulphia stokesiana (Greville) Mann (1925: 46) .

Trigonium stokesianum (Greville) Hustedt (1959a : taf. 472, figs 1–3).

Valves tripolar, with straight or gently convex sides ( Figs 155–159 ), concave sides uncommon ( Fig. 160 ). Valve face undulate, with a central trifolium, and slightly raised poles ( Figs 149, 154 ). A conspicuous pseudocellus is located at each pole ( Figs 149, 153–154 ). Whole valve face perforated by cribrate poroid areolae ( Figs 151–152 ) arranged in radial rows ( Figs 155–160 ). Areolae are slightly variable in diameter, numerous pores and granules are scattered in between rows of areolae ( Fig. 151 ). On the trifolium, areolae are more loosely spaced than in the depressed sectors of the valve face ( Figs 149, 154 ). A group of rimoportulae, which in some cases form a poorly defined ring, is located in the central part of the trifolium ( Figs 149–152 ). Externally, rimoportulae are visible as short tubes with thick rims ( Fig. 151 ); internally as short slits between two barely raised lips ( Fig. 152 ). Mantle is steeply downturned, variable in depth ( Fig. 149 versus Figs 150 and 154 ), sometimes stepped ( Fig. 149 ), perforated by poroid areolae arranged in rows parallel to the pervalvar axis ( Fig. 149 ). Mantle margin smooth, hyaline, slightly expanded inwardly ( Fig. 150 ), and sometimes also outwardly ( Fig. 149 ). Series of short, robust costae are present on the valve interior, along each side ( Figs 150 , 155–160 ). The costae arise from the expanded mantle margin and protrude a short distance toward the valve face centre ( Figs 155–160 ). Raised sectors of the trifolium are usually bound by a costa on each side ( Figs 150 , 155–160 ). Girdle is composed of closed elements, bearing alternating rows of poroid areolae parallel to the pervalvar axis ( Fig. 154 ). Measurements (n =14): average side length: 106.3–221.8 µm; 1 costa in 10 µm measured along the valve face margin; 2–3 areolae in 10 µm; number of rimoportulae: 3–9.

Geographic and stratigraphic distribution ( Fig. 10, sites 18*, 22–25, 28, 33*; questionable records indicated with asterisks):

(a) specimens: Late Eocene-earliest Oligocene: Oamaru , Otago, New Zealand: BM coll. Adams GC3419. 2 Middle Miocene : Szakal, Hungary: BM63751 ( Figs 156–157 ), BM coll. Adams C921; Szurdokpüspöki (‘Castel’), Hungary:

BM stub P.1352 ( Figs 150–152 ), BM coll. Adams 741 ( Fig. 159 ); Pôtor, Slovakia (formerly Szent-Peter, Hungary): BM

stub E.9 ( Figs 149, 153–154 ), BM coll. Adams F1387. Middle-late Miocene: Morón de la Frontera, Spain: BM coll. Adams TS253.

2. Questionable: labelled as T. stokesianum but is probably Entogoniopsis pseudonervata comb. nov. (see below).

22 • Phytotaxa 209 (1) © 2015 Magnolia Press

WITKOWSKI ET AL.

Late Miocene: Sendai, Japan: BM coll. Adams GC3422.

Age unspecified, presumably Miocene: Hungary: BM63749 ( Figs 155, 160 ), BM63750 ( Fig. 158 ).

(b) records:

Late Eocene-earliest Oligocene: Oamaru , Otago, New Zealand: Grove & Sturt (1887c: 137), De Lautour (1888: 310), Tempère & Peragallo (1915: 396). These records are questionable, as there are no other records of Entogoniopsis stokesiana from sediments older than the Miocene time interval. As E. stokesiana is easily confused with E. pseudonervata (see below), we suggest that the reports by Grove & Sturt (1887c), De Lautour (1888) and Tempère & Peragallo (1915) represent records of E. pseudonervata .

Early Miocene: Horné Strháre (formerly Felsö-Esztergály), Slovakia: Pantocsek (1886: 54); Modrý Kameň, Slovakia (formerly Kékkö, Hungary): Pantocsek (1886: 54), Tempère & Peragallo (1915: 149); Marmorito, Italy: Tempère & Peragallo (1915: 197).

Late early Miocene: Mogyoród, Hungary: Pantocsek (1886: 54).

Early-middle Miocene: Brno region, Moravia, Czech Republic (formerly Brünn, Mähren): Cleve (1885: 169, pl. 12, figs 5a,b), Schmidt (1886b: taf. 112, fig. 19), Hustedt (1959a: taf. 472, fig. 5); Přibice, Moravia, Czech Republic (formerly Pribitz, Mähren): Hustedt (1959a: taf. 472, figs 4, 6).

Middle Miocene : Szakal, Hungary: Pantocsek (1886: 54), Tempère & Peragallo (1915: 69), Hustedt (1959a, taf. 472, fig. 2); Bajtava (formerly Bajtha), Slovakia: Pantocsek (1886: 54), Veľký Krtíš, Slovakia (formerly Nagy-Kürtös, Hungary): Pantocsek (1889: 95), Hustedt (1959a, taf. 472, figs 1, 3); Szurdokpüspöki (‘Castel’), Hungary: Ross & Sims (1971: 160, pl. I, figs 5–7); Pôtor, Slovakia (formerly Szent-Peter, Hungary): Pantocsek (1886: 54, pl. X, figs 87a,b).

Middle-late Miocene: Morón de la Frontera, Spain: Greville (1866a: 8), Tempère & Peragallo (1915: 205); Montemayor, Spain: Tempère & Peragallo (1915: 325).

Late Miocene: Sendai, Japan: Schmidt (1890: taf. 159, fig. 2), Tempère & Peragallo (1915: 15).

Recent: Mann (1925: 46) recorded this species as extant in the Philippine Islands, transferring it to Biddulphia in the process. We consider Mann’s report as questionable, as there are no other records of E. stokesiana from sediments younger than the Miocene time interval. Similarly, we observed a specimen on slide BM coll. Adams GC3421, labelled as coming from Manila, Philippine Islands. This diatom has a large tripolar valve with high polar elevations and short, evenly spaced internal costae located along each margin. The central area of the valve face is conspicuously domed, hence it does not appear to be E. stokesiana . This supports the notion that the occurrences of E. stokesiana are restricted to the Miocene.

Observations:—Based on material from Morón, Spain, Greville (1866a: 8) noted that in his specimens of Triceratium stokesianum , the ‘veins’ protruding from the two adjacent sides of the valve are connected. Further, Greville suggested some morphological variation in the number of costae, and in the distance that they reach toward the valve face centre. From the Brno region of Moravia, Czech Republic, Cleve (1885: 169) reported specimens with internal costae that do not fuse. The description in Cleve (1885) emphasizes the variability in the pattern formed by internal costae. Grunow (1887a: 66), however, suggested that the specimen reported by Cleve (1885) differs from that of Greville (1866a), such that the former should be treated as new variety: T. stokesianum var. moravica Grunow (in Schmidt 1886b: taf. 112, fig. 19). Grunow (1887a; see also Schmidt 1886b, taf. 112, fig. 19) did not specify the morphological features that distinguish this variety from the species, but presumably he meant the variations in the number and length of the internal costae. Hustedt (1959a: taf. 472) found these taxonomic concepts confusing and proposed the following distinction: (1) T. stokesianum has undulate valves, with one raised sector per side bound by a pair of divergent internal costae; (2) T. stokesianum var. moravica has a circumferential depression around a raised valve face centre and the internal costae are generally convergent. In addition to valve view figures that support this distinction (taf. 472: figs 1–2, 5–6), Hustedt figured both in oblique or girdle view (1959a: taf. 472, figs 3–4). The oblique view of his T. stokesianum shows a clear trifolium in the centre (taf. 472, fig. 3); the girdle view of T. stokesianum var. moravica shows a raised area in the middle of the side facing the observer, which also suggests a presence of a trifolium. Therefore, Hustedt (1959a) may have mislabelled his girdle view figure of T. stokesianum var. moravica .

Our examination of specimens from Morón, Sendai, and a number of Central European localities shows that the arrangement of internal costae in E. stokesiana is highly variable. A single specimen from Morón examined in this study (BM coll. Adams TS253) shows faint connections between pairs of internal costae arising from adjacent valve margins. This specimen is also convex-sided, which is another character that agrees well with Greville’s description. However, SEM examination of the valve interior in a specimen from Szurdokpüspöki (‘Castel’) ( Fig. 150 ) shows that such connections are found also in specimens from other localities, and that they are merely local thickenings of the valve face. Therefore, whether the internal costae appear to fuse or not depends primarily on toward which side these costae are deflected. Specimen in Fig. 156 shows the presence of such thickened areas

ENTOGONIOPSIS GEN. NOV. AND TRILAMINA GEN. NOV. (BACILLARIOPHYTA)

Phytotaxa 209 (1) © 2015 Magnolia Press • 23 between a pair of costae adjacent to one pole, and between pairs of neighbouring costae. Therefore, given the high degree of morphological variation in E. stokesiana , we see no use in defining infraspecific taxa based on the pattern formed by the internal costae alone. None of the specimens examined in this study showed a circumferential depression comparable to that reported by Hustedt (1959a). As in the case of Trilamina wittiana (see below), we suggest that the morphological variation in E. stokesiana may be further complicated if this species formed heterovalvar frustules. Until intact frustules have been examined, we propose to synonymize the species and the variety.

Figures

FIGURES 149–154. Scanning electron micrographs of Entogoniopsis stokesiana from Pôtor (formerly Szent-Péter), Slovakia (Figs 149, 153–154) and Szurdokpüspöki (‘Castel’), Hungary (150–152). Fig. 149: Oblique external view, showing a trifolium in the centre of the valve face. The central ring of rimoportulae is indicated with an arrowhead. Note the deep, stepped mantle (arrow). Fig. 150: Oblique internal view of a valve with four internal costae per side. Note that the more prominent costae are associated with the edges of the trifolium (arrows). The location of the rimoportulae is indicated with an arrowhead. Fig. 151: Detail of the central area, showing the external tubes of the rimoportulae bearing thickened rims, and granules scattered on the valve face (arrow). Note the well-preserved slightly sunken cribra, and the diminished size of areolae next to the rimoportulae. Fig. 152: Detail of the specimen in Fig. 150, showing the internal openings of the rimoportulae, in the form of a slit between slightly raised lips. Note the smaller areolae located next to the rimoportulae, and the well-preserved cribra. Fig. 153: Detail of a pseudocellus. Fig. 154: Oblique external view of a specimen with a detached girdle element, probably a copula (C?).

FIGURES 155–160. Light micrographs showing the range of morphological variation in Entogoniopsis stokesiana. Figs 155 and 160: BM63749, Hungary. Figs 156–157: BM63751, Szakal, Hungary. Note valve face thickenings indicated by arrowheads in Fig. 156. Fig. 158: BM63750, Hungary. Fig. 159: BM coll. Adams TS741, Castel, Hungary.