Phylogeny and Fatty Acid Profiles of New Pinnularia (Bacillariophyta) Species from Soils of Vietnam

We studied the morphology, ultrastructure, and phylogeny of eight soil diatom strains assigned to the Pinnularia genus. Six of these strains, identified by us as new species, are described for the first time. We provide a comprehensive comparison with related species and include ecological data. Molecular phylogeny reconstruction using 18S rDNA and rbcL affiliates the new strains with different subclades within Pinnularia, including ‘borealis’, ‘grunowii’ and ‘stomatophora’. We also studied the fatty acid profiles in connection with the emerging biotechnological value of diatoms as a source of lipids. Stearic (36.0–64.4%), palmitic (20.1–30.4%), and palmitoleic (up to 20.8%) acids were the dominant fatty acids in the algae cultured on Waris-H + Si medium. High yields of saturated and monounsaturated fatty acids position the novel Pinnularia strains as a promising feedstock for biofuel production.


Introduction
Diatoms have been featured in multiple fields of technology and science, including paleoecological reconstructions [1], forensics [2], and biomedicine [3]. Strikingly, these microalgae produce about 20% of the primary biomass on Earth [4], while fixing about 25% of the global CO 2 . Moreover, diatoms accumulate enormous quantities of lipids-an estimated 10-fold compared with cultured terrestrial plants [5]. Despite the abundance of saturated and polyunsaturated long-chain fatty acids (FA), particularly eicosapentaenoic, docosahexaenoic, and arachidonic acids [4,6], FA profiles of diatoms remain understudied. The few reports are focused on marine species [7,8], notably Phaeodactylum tricornutum Bohlin rich in palmitic, palmitoleic, and eicosapentaenoic acids [7]. Another marine diatom, Thalassiosira weissflogii (Grunow) G.A. Fryxell et Hasle, is also rich in unsaturated palmitoleic and eicosapentaenoic acids [8]. Despite the eventual progress in FA studies on marine diatoms, similar studies focused on soil species are extremely rare. For instance, no FA profiles for such a ubiquitous genus as Pinnularia Ehrenberg 1843 can be found in the literature.
Pinnularia is one of the most numerous genera of biraphid diatoms [9]. A recent retrieval from Algaebase [10] includes 1432 specific and 1476 infraspecific epithets, of which 1403 have been flagged as accepted taxonomically. The Pinnularia algae are ubiquitously found in fresh waters and soils [11], reaching the highest diversity in the tropics (see the report on tropical diatoms of South America by Metzeltin and Lange-Bertalot [12]). Pinnularia is often mentioned as the most diverse group in soil algocenoses, as many of its species are cosmopolitan and common [13,14]. Numerous species of Pinnularia thrive in soils of spruce forests of the middle and southern taiga [15] and volcanic soils of Kamchatka, Russia [16]. Several Pinnularia species (P. borealis, P. obscura, P. schoenfelderi, and P. sinistra) have been identified as dominating taxa in the soils of Poland [17,18]. P. obscura and P. schoenfelderi also prevail in the Attert basin, Luxembourg [19], whereas P. borealis,

Sample Collection Procedure
The samples were collected as follows: the surface of the test site was examined to detect macrogrowth of algae, and then a composite sample was taken from an area of 10-30 m 2 . The composite sample consisted of 5-10 individual samples. For an individual sample, the topsoil was removed from an area of 5 to 20 cm 2 with a metal scoop or shovel. After sampling, the instruments were cleaned and sterilized with ethanol. The samples were placed in labelled plastic zip bags and carried to the laboratory. The hot-drying method measured the absolute humidity [42] and the samples were air-dried and packaged.
To measure pH, we mixed 30 g of soil with 150 mL of distilled water [43]. The suspension was poured into a clean glass beaker and the measurements were performed with a Hanna Combo (HI 98129) device (Hanna Instruments, Inc., Woonsocket, RI, USA).

Culturing
Gathered materials were processed in the Laboratory of Molecular Systematics of Aquatic Plants of the Institute of Plant Physiology of the Russian Academy of Sciences (IPP RAS). A sample of soil was thoroughly mixed and placed into a Petri dish, then saturated with distilled water up to 60-80% of full moisture capacity and placed into an illuminated climate chamber. After a 10-day incubation, the sample was diluted with a small amount of distilled water, mixed gently, and the suspension was transferred to a Petri dish for LM using a Zeiss Axio Vert A1 inverted microscope. Algal cells were extracted with a micropipette, washed in 3-5 drops of sterile distilled water and placed into a 300 µL well on a plate for enzyme-linked immunoassay with Waris-H + Si [44]. Non-axenic unialgal cultures were maintained at 22-25 • C in a growth chamber with a 12:12 h light: dark photoperiod and checked every 10-14 days for 5 months.
Cells 2022, 11, 2446 5 of 31 For fatty acid analysis cultures were maintained on Waris-H + Si in 250 mL Erlenmeyer glass flasks with 150 mL medium, under constant orbital shaking (150 rpm in ELMI Sky Line Shaker S-3L, ELMI Ltd., Riga, Latvia) for 25 days at 25 • C. The light intensity was 100 µmol photons m −2 s −1 with a 16:8 h light/dark photoperiod. All analyses were performed in triplicate. Tables show the mean values and standard errors.

Microscopy
A culture was treated with 10% hydrochloric acid to remove carbonates and washed several times with deionized water for 12 h. Afterwards, the sample was boiled in concentrated hydrogen peroxide (≈37%) to remove organic matter. It was washed again with deionized water four times at 12 h intervals. After decanting and filling with deionized water up to 100 mL, the suspension was pipetted onto coverslips and left to dry at room temperature. Permanent diatom preparations were mounted in Naphrax. Light microscopic (LM) observations were performed with a Zeiss Axio Scope A1 microscope (Carl Zeiss Microscopy GmbH, Gottingen, Germany) equipped with an oil immersion objective (×100, n.a. 1.4, differential interference contrast) and Axiocam ERc 5s camera (Carl Zeiss NTS Ltd., Oberkochen, Germany). Valve ultrastructure was examined using a scanning electron microscope JSM-6510LV (IBIW; Institute for Biology of Inland Waters RAS, Borok, Russia). For scanning electron microscopy (SEM), part of the suspensions were fixed on aluminum stubs after air-drying. The stubs were sputter-coated with 50 nm of Au using an Eiko IB 3 machine (Eiko Engineering Co. Ltd., Tokyo, Japan). The suspension and slides are deposited in the collection of Maxim Kulikovskiy at the Herbarium of the Institute of Plant Physiology Russian Academy of Sciences, Moscow, Russia.
Amplifications were carried out using premade polymerase chain reaction (PCR) mastermixes (ScreenMix by Evrogen, Moscow, Russia). Amplification conditions for the 18S rDNA gene were as follows: initial denaturation for 5 min at 95 • C followed by 35 cycles of 30 s denaturation at 94 • C, 30 s annealing at 52 • C, and 50 s extension at 72 • C, with the final extension for 10 min at 72 • C. Amplification conditions for the rbcL gene were as follows: initial denaturation for 5 min at 95 • C followed by 45 cycles of 30 s denaturation at 94 • C, 30 s annealing at 59 • C, and 80 s extension at 72 • C, with the final extension for 10 min at 72 • C. PCR products were visualized by horizontal electrophoresis in 1.0% agarose gel stained with SYBR TM Safe (Life Technologies, Carlsbad, CA, USA). The products were purified with a mixture of FastAP, 10×FastAP Buffer, Exonuclease I (Thermo Fisher Scientific, Waltham, MA, USA), and water. The sequencing was performed using a Genetic Analyzer 3500 instrument (Applied Biosystems, Waltham, MA, USA).
Editing and assembling of the consensus sequences were carried out by processing the direct and reverse chromatograms in Ridom TraceEdit ver. 1.1.0 (Ridom GmbH, Münster, Germany) and Mega7 software [48]. The reads were included in the alignments along with corresponding sequences of 72 diatom species downloaded from GenBank (taxa names and Accession Numbers are given in Figure 2). Three centric diatom species were chosen as the outgroups.

USA).
Editing and assembling of the consensus sequences were carried out by processing the direct and reverse chromatograms in Ridom TraceEdit ver. 1.1.0 (Ridom GmbH, Münster, Germany) and Mega7 software [48]. The reads were included in the alignments along with corresponding sequences of 72 diatom species downloaded from GenBank (taxa names and Accession Numbers are given in Figure 2). Three centric diatom species were chosen as the outgroups. The nucleotide sequences of the 18S rDNA and rbcL genes were aligned separately using the Mafft ver. 7 software (RIMD, Osaka Japan) and the E-INS-i model [49]. The final alignments were then carried out: unpaired sites were visually determined and removed from the beginning and the end of the resulting matrices. For the protein-coding sequences of the rbcL gene, we checked that the beginning of the aligned matrix corresponds to the first position of the codon (triplet). The resulting alignments had lengths of 450 (18S rDNA) and 957 (rbcL) characters. After removal of the unpaired regions, the aligned 18S rDNA gene sequences were combined with the rbcL gene sequences into a single matrix Mega7 (Appendix S1).
The data set was analyzed using the Bayesian inference (BI) method implemented in Beast ver. 1.10.1 software (BEAST Developers, Auckland, New Zealand) [50] to construct a phylogeny. For the alignment partition the most appropriate substitution model, shape parameter α and a proportion of invariable sites (pinvar) were estimated using the Bayesian information criterion (BIC) as implemented in jModelTest ver. 2.1.10 (Vigo, Spain) [51]. This BIC-based model selection procedure selected the following models, shape parameter α and a proportion of invariable sites (pinvar): HKY + I + G, α = 0.6250 and pinvar = 0.5170 for 18S rDNA; TPM1uf + I + G, α = 1.1070, and pinvar = 0.7470 for the first codon position of the rbcL gene; JC + I + G, α = 0.3830 and pinvar = 0.7320 for the second codon position of the rbcL gene; TPM2uf + G and α = 0.5470 for the third codon position of the rbcL gene.
We used the F81 model of nucleotide substitution instead of TPM1uf, the HKY model instead of JC, and TPM2uf, given that they were the best matching model available for BI. A Yule process tree prior was used as a speciation model. The analysis ran for 5 million generations with chain sampling every 1000 generations. The parameters-estimated convergence, effective sample size (ESS), and burn-in period were checked using the Tracer ver. 1.7.1 software (MCMC Trace Analysis Tool, Edinburgh, United Kingdom). [50]. The initial 25% of the trees were removed, and the rest were retained to reconstruct a final phylogeny. The phylogenetic tree and posterior probabilities of its branching were obtained based on the remaining trees, having stable estimates of the parameter models of nucleotide substitutions and likelihood. The Maximum Likelihood (ML) analysis was performed using RAxML software [52]. The nonparametric bootstrap analysis with 1000 replicas was used. The phylogenetic tree topology is available online in Appendix S2. FigTree ver. 1.4.4 (University of Edinburgh, Edinburgh, United Kingdom) and Adobe Photoshop CC ver. 19.0 software (Adobe, San Jose, CA, USA) were used for viewing and editing the trees.

Fatty Acid Analysis
Biomass preparation for determining the fatty acid methyl ester (FAME) profiles was performed according to Maltsev et al. [53]. The diatom suspensions were conveyed to 15-50 mL tubes (depending on the volume). The cells were pelleted at room temperature for 3 min at 3600 g. The supernatant was removed, and the pelleted cells were resuspended in 10-15 mL (depending on the amount of biomass) of distilled water, quantitatively transferred to 15 mL centrifuge tubes, and pelleted again by centrifugation. The supernatant was removed, and samples were quantitatively transferred to a 50 mL round-bottom flask. Heptadecanoic acid (Sigma-Aldrich, St. Louis, MO, USA) was used as the internal standard for the fatty acid composition determination. To avoid the oxidation of unsaturated fatty acids, all samples were processed under an argon atmosphere. Ten milliliters of a 1 M solution of KOH in 80% aqueous ethanol was added to the dry residue, and the flask was sealed with a reflux condenser, and kept for 60 min at the boiling point of the mixture (~80 • C). After the time-lapse, the solvents were evaporated in vacuo to a volume of 3 mL and quantitatively transferred with distilled water to a 50 mL centrifuge tube to a total volume of 25 mL, followed by extracting the unsaponifiable components with 10 mL portions of n-hexane (Himmed, Moscow, Russia) 3 times. To accelerate the separation of the phases, the tube was centrifuged for 5 min at room temperature and 2022× g. After that, the aqueous phase was acidified to a slightly acidic reaction (on indicator paper) with a few drops of 20% sulfuric acid (Himmed, Moscow, Russia), and free fatty acids were extracted with 20 mL of n-hexane. The hexane solution of free fatty acids was transferred to a dry 50 mL round-bottom flask, and the solvent was evaporated to dryness using a rotary evaporator IKA RV-10 (IKA-WERKE, Staufen im Breisgau, Germany), after which 10 mL of absolute methanol (Sigma-Aldrich, St. Louis, MO, USA) and 1 mL of acetyl chloride (Sigma-Aldrich, St. Louis, MO, USA) were added to the dry residue. The flask, closed with a reflux condenser, was kept for one hour at 70 • C, then the solvents were evaporated to dryness, a few drops of distilled water were added to the dry residue, and FAMEs were extracted with n-hexane.
The obtained FAMEs were analyzed using an Agilent 7890A gas-liquid chromatograph (Agilent Technologies, Santa Clara, CA, USA) with an Agilent 5975C mass spectrometric detector. A DB-23 capillary column 60 m long and 0.25 mm in diameter was used (Agilent Technologies, Santa Clara, CA, USA). The remaining conditions of the analysis were as follows: carrier gas was helium, flow rate of 1 mL min −1 , 1 µL volume of injected sample, 1:5 flow divider, and the evaporation temperature of 260 • C. Temperature gradient program: from 130 to 170 • C in 6.5 • C min −1 steps; from 170 to 215 • C in 2.5 • C min −1 increments, 215 • C for 25 min, from 215 to 240 • C in 40 • C min −1 increments, and the final stage lasting 50 min at 240 • C. The operating temperature of the mass spectrometric detector was 240 • C and the ionization energy was 70 eV.

Results and Discussion
Pinnularia specimens are ubiquitously found in soil samples collected in the Cát Tiên National Park. Based on the results of DNA sequencing, LM/SEM observations, and FA profiling of eight different strains of Pinnularia, we regard six of them as new species. For the sake of clarity, we address these strains by specific epithets ahead of the formal description.

Molecular Phylogeny
Comprehensive studies on Pinnularia phylogeny [24,25] confirm its monophyletic origin; the genus splits into three clades stably supported by the analysis. In a study by Souffreau et al. [25], who used five genetic markers-two nuclear (18S rDNA and 28S rDNA), two plastid (rbcL and psbA), and a mitochondrial cox1, these clades were designated A, B, and C. Each clade consists of several subclades characterized by morphological similarity about shapes (valves and apices, raphe endings linear or rounded, raphe fissures straight or undulate, chloroplasts H-shaped or elongated, with pyrenoids or not, etc.) and specific markings (ghost striae, fascia, wart-like bodies, etc.). Although our phylogenetic analysis involved only two markers (nuclear 18S rDNA and plastid rbcL), it perfectly preserved the tree topology and maintained high support to all clades and subclades introduced by Souffreau et al. [25]. The tree was expanded through the addition of the new taxa, which showed full morphological consistency with their parental subclades ( Figure 2).
Clade A includes two previously characterized subclades exemplified by Caloneis Cleve 1894 and P. cf. divergens. Here we complement the 'divergens' subclade with new species; a morphological feature is a fascia with rounded thickenings at the margin. Furthermore, the expansion of Pinnularia phylogeny revealed a new subclade of clade A, 'stomatophora', comprising P. ministomatophora, P. valida (strain VN305), and P. stomatophora (strain D11_014), all of them presenting with characteristic hollow markings on the external surface of the valve, crescent-shaped (P. stomatophora (p. 456, pl. 98, Figure 8 of [11]) or irregular (P. ministomatophora sp. nov.).
Clade B includes three subclades, 'grunowii', 'nodosa', and 'subgibba', of diminutive linear algae with bulbous apices and rounded external endings of the raphe [25]. Our phylogeny investigation recognizes these subclades with high support. However, the two species of 'nodosa' subclade, P. nodosa and P. acrosphaeria [25] split into distinct branches ( Figure 2). These species also show morphological distinctions: P. acrosphaeria display mottled, structured areas on both outer and inner surfaces of the valve (p. 296, pl. 19, Figures 1-6 of [11]) whereas in P. nodosa with the valve is smooth on the inside and its entire outer surface is heavily structured (p. 307, pl. 24, Figures 1-6

of [11]).
A morphological feature of the 'subgibba' subclade is ghost striae in the central area [25][26][27]. The term ghost striae was proposed by Cox [54] and refers to thinnings on the inner surface of the valve, corresponding in size and spacing to normal striae. It should be noted that there is no unified name for these structures defined as "large markings, differently structured on both sides and larger in the ventral side" [11], "deepenings on the inside of the valve" [11] or "depressions in the central area" [55]. Our phylogenetic study encompassed P. microstauron, P. parvulissima, P. cf. gibba, P. cf. subgibba var. sublinearis, P. subcapitata var. elongata, P. kattiensis, and several strains of Pinnularia sp. The newly identified soil species P. microgibba, P. minigibba, and P. vietnamogibba along with P. shivae form a separate branch within the 'subgibba' subclade ( Figure 2). Ultrastructural examination of P. microgibba, P. minigibba, and P. vietnamogibba specimens by SEM revealed ghost striae in the central area for all of them, consistently with their phylogenetic affiliation.
Clade C specimens have no markings in the central area but exhibit wide strokes on the valves. Our study generally preserved the topology for this clade introduced by Souffreau et al. [25], while the 'borealis' subclade was complemented by new strains and new species P. paradubitabilis. Sister lineages to the 'borealis' subclade are constituted by strains of P. amphisbaena and two related subclades, 'subcommutata' and 'viridiformis', comprising large specimens of linear-elliptical shape with almost parallel striae and small central areas, undulate external raphe fissures and with linear (for subcommutata) or rounded (for viridiformis) central raphe endings. In contrast with the original version [25], P. cf. microstauron, P. brebissonii, P. accuminata, Pinnularia sp. 4 (Wie)a, and strains morphologically similar to P. microstauron split into separate branches within clade C ( Figure 2).
It is interesting to observe that certain morphological distinctions, notably the presence of ultrastructural surface markings in the central area, consistently follow phylogenetic clades. Species presenting with markings in the central area can be reliably classified on their basis as 'divergens', 'stomatophora', 'subgibba', 'nodosa', and 'gronowii' groups.

Comparative Morphology
Pinnularia minigibba sp. nov. is similar to several species of the P. gibba complex ( Table 2). The new species must not be confused with P. australogibba var. subcapitata [55]. The similarities include the outlines and proportions of the valves, characteristic shape of the central area with a broad fascia, and ghost striae. However, P. minigibba has wider valves (7-8 µm vs. 5.7-7.3 µm in P. australogibba var. subcapitata) with slightly concave margins (as opposed to convex sides of P. australogibba var. subcapitata). Other differences include stria density (respectively, 9-10 in 10 µm vs. 11-12 in 10 µm) and the shape of axial area (≤1/4 of the width and slightly wider towards the midportion in P. minigibba vs. lanceolate in P. australogibba var. subcapitata); also, in the new species, the stria is more tilted in the midportion. Another closely similar species, P. parvulissima (p. 397, pl. 69, Figures 6-11 of [11]), can be differentiated by larger size (34-70 µm length to 10-12 µm width vs., respectively, 40-43 to 7-8 in P. minigibba); besides, P. parvulissima have convex margins and wider axial area. The similarity of P. minigibba with certain strains of the polymorphic widespread P. microstauron (Ehrenberg) Cleve should be noted as well. For instance, it can be confused with P. microstauron var. angusta (p. 361, pl. 51, Figures 4-7 of [11]), which is a smaller variety with wider fascia, and a particularly challenging specimen shown in Figure 18, pl. 164 of Metzeltin et al. [57]. About these examples, P. minigibba can be differentiated by stria density (9-10 in 10 µm vs. 10-12 in 10 µm in P. microstauron var. angusta and 11 in 10 µm in P. microstauron sensu Metzeltin et al. [57], as well as concave margins (vs. parallel or subtly convex in P. microstauron), and characteristic presence of ghost striae in the central area. Despite the apparent absence of ghost striae in LM images of P. microstauron var. angusta and P. microstauron sensu Metzeltin et al. [57], ultrastructural examinations of archetypal P. microstauron specimens (e.g., AT_112Gel04, AT_113Gel11 cultures [55,56]) reveal hollowed markings on the inner surface of the valve confined to the central area, corresponding to ghost striae, consistently with the assignment to the 'subgibba' subclade (p. 371, Figure 3 of [26]).
Pinnularia vietnamogibba sp. nov. must not be confused with two closely similar taxa-P. gibba var. subsancta (pl. 13, Figure 7 [55]). According to Krammer [11] (P. 111), most specimens defined in literature as P. gibba var. sancta (Grunow) Meister arguably belong to a smaller variety described by Manguin (pl. 13, Figure 7 of [58]). According to Hustedt [60], this taxon is widespread in the tropics, whereas Pinnularia australogibba is found in ravines at Point Del Cano, Île Amsterdam in the southern Indian Ocean. The similarities of these species with the newly identified P. vietnamogibba include the shape of the central area with a broad fascia and ghost striae. However, P. gibba var. subsancta and P. australogibba specimens have lanceolate outlines, whereas P. vietnamogibba are linear, with the margins parallel or subtly bulging. P. vietnamogibba also has much broader fascia, which constitutes 17-20% of the valve length (cf. 4-10.3% of the length in P. gibba var. subsancta or 7.7-14.5% in P. australogibba) and lower stria density (10-11 in 10 µm vs. 13-15 in 10 µm and 12-13 in 10 µm in P. gibba var. subsancta and P. australogibba, respectively). P. vietnamogibba can also be confused with the already mentioned P. microstauron var. angusta (p. 361, pl. 51, Figure      * counted from published data. Pinnularia microgibba sp. nov. phylogenetically and morphologically closely related with 'subgibba'-group species [25,26], particularly with strains Pinnularia sp. 6 (Tor4)r and Pinnularia sp. 3 (Tor8)b (from littoral zones of freshwater bodies in Chile, see Table 2). These strains are virtually identical in size, shape, and outline; the only clue to their morphological distinction is stria density (12 in 10 µm. for (Tor8)b and 14 in 10 µm for (Tor4)r). Meanwhile, they are phylogenetically independent ( Figure 2). Morphometric data of P. microgibba almost totally match those of (Tor4)r and (Tor8)b, except for the shape the of central area: linear, broadening towards the valve margins in (Tor4)r and (Tor8)b, and rhombic in the new species. Phylogenetically, the strains of P. microgibba together with P. minigibba and P. vietnamogibba form a separate branch within the 'subgibba' subclade.
It should be kept in mind that ghost striae are often poorly distinguishable in light micrographs. For this reason, P. microgibba can be easily confused with certain diminutive species with narrow linear outlines and broad fascia, including the widespread, ubiquitous, and cosmopolitan P. sinistra [11,63]. However, P. microgibba has concave margins and a rhombic central area with ghost striae, whereas P. sinistra has a linear central area without markings (p. 265, Taf. 37, Figure 16 of [63]). Interestingly, a population of P. sinistra inhabiting the small oceanic island Île Amsterdam, situated in the southern part of the Indian Ocean (p. 225, Figures 194-209 of [55]), also has concave margins and rhombic central area; however, these taxa can be distinguished from both the type of P. sinistra and type of P. microgibba by their relatively wide lanceolate axial area (compared with the narrow-linear axial area in type populations of P. sinistra and P. microgibba). Several images for certain species found in the literature can be confused with P. microgibba as the distinctions are very subtle and demand close attention ( Table 2). For instance, the valves of P. microstauron var. angusta (p. 361, pl. 51, Figures 5 and 6 of [11]) are wider (6.5-8.0 µm) compared with P. microgibba (5.5-6 µm). P. subcapitata W. Gregory (given as P. hilseana Janisch 1861 (p. 757, pl. 205, Figure 9 Figures 3 and 4 of [57]) have a lower stria density (9.5-11.0 in 10 µm) compared with P. microgibba (11.0-12.0 in 10 µm). Pinnularia pisciculus found in low nutrient waters, mosses, and dry soils from India (pl. 73 [65]) display characteristic triundulate sides and capitate apices. Pinnularia similiformis var. koreana (p. 291, pl. 16, Figures 3-6 of [11]) is lengthy (40-60 µm), whereas P. microgibba is 36-40 µm. Pinnularia marchica (p. 283, pl. 12, Figures 11-17 of [11]) have elongated rostrate apices different from the subcapitate apices of P. microgibba. Ghost striae in the central area represent an important identifier; however, it should be applied with caution, as SEM data are not universally available and the sensitivity of LM about ghost striae is limited (Table 2).
Pinnularia insolita sp. nov. can be recognized by its unusual outline with concave side margins and narrowed apices, as well as the characteristic shape of the central area with a prominent fascia broadening towards the margins. Nevertheless, certain varieties of other Pinnularia species are similar ( Table 2). Large specimens of P. pisciculus Ehrenberg (p. 429, pl. 85, Figures 25 and 26 of [11]) are similar to P. insolita (length 22-50 µm, width 6.0-8.3 µm in P. pisciculus vs. length 50-52 µm and width 7-7.5 µm in P. insolita), stria density (10.5-12.0 µm in 10 µm vs. 11-12 in P. insolita), and overall outline. Nevertheless, the algae can be reliably differentiated by their apices shape (capitate in P. pisciculus and subtly rostrate in P. insolita) and the shape of the central area (rhombic in P. pisciculus vs. broadening towards the margins in P. insolita). Moreover, larger specimens of P. pisciculus tend to have subtly triundulate side margins, whereas the sides of P. insolita are concave. Another similar species with P. insolita is P. brebissonii var. bicuneata, whose largest annotated specimen (p. 351, pl. 46, Figure 9 of [11]) resembles P. insolita by its size, shapes of the valve apices and central area, and stria density. However, in P. brebissonii var. bicuneata the sides are straight and parallel (vs. subtly concave in P. insolita), the valves are wider (8-11 µm vs. 7-7.5 µm in P. insolita), and the axial area is narrow (vs. moderate, about 1/3 width of the width, in P. insolita). Other confusing specimens are those of P. cavancinii (p. 411, pl. 127, Figures 1-3 of [57]), which also have narrowed apices, as well as valve dimensions, stria densities, and central/axial areas similar to P. insolita. However, these can also be differentiated by the outline (rhombic-lanceolate to elliptic-lanceolate in P. cavancinii vs. linear with subtly concave sides in P. insolita). Molecular phylogeny affiliates P. insolita with the subclade of P. obscura AT_70Gel12b and P. marchica Ecrins4_a. Voucher images available for these strains correspond to the species descriptions [24,25,56] and show considerable morphometric differences with P. insolita, including the smaller size (length ≤ 37 µm and width 6.3 µm vs. 50-52 µm and 7-7.5 µm in P. insolita) as well as distinctive outline and stria densities ( Table 2). It should be noticed, however, that a common morphological feature of this subclade is the absence of any markings on both sides of the valve within the central area ( Table 2).
Pinnularia paradubitabilis sp. nov. occupies a separate phylogenetic position within the 'borealis' subclade and at a first glance appears similar to archetypal specimens of this ambiguous taxonomic group. Morphology and phylogeny of the widespread polymorphic P. borealis have been extensively described in several studies, with multiple morphotypes illustrated [1, 11,28,29,67]. However, despite the overall semblance, the new species can be convincingly distinguished from the rest of 'borealis' by its broad fascia. In the opinion of Kollár et al. [26], the breadth of fascia represents a more stable indicator than valve outlines or aspect ratio. None of P. borealis morphotypes have fascia, other than the 1-2 stria missing in the central area. In addition, according to the comprehensive description by Krammer [11], P. borealis have wider valves (8.5-10.0 µm vs. 6-7 µm in P. paradubitabilis). The outlines are different as well: in P. borealis valves are linear or elliptic-linear with moderately convex margins, ends rounded, whereas in P. paradubitabilis the valves are linear with margins parallel or subtly concave, ends bluntly rounded. By morphometric features including size, outline, fascia, and stria density (Table 2), the newly identified species is most similar to P. dubitabilis Hustedt found in Java and Sumatra (p. 276, pl. 9, Figures 7-9 of [11]). On the other hand, P. dubitabilis have very short striae, confined to the narrow space along margins, and a wide axial area, whereas P. paradubitabilis show the opposite patterns with extensive striae and narrow axial area. Several other species from the section Distantes (Cleve) Patrick are similar to our new taxon, most notably those with fascia: P. angustiborealis Krammer et Lange-Bertalot and P. intermedia (Lagerstedt) Cleve (Table 2). However, P. angustiborealis have moderately convex side margins (cf. parallel or concave margins of P. paradubitabilis) and the width of 7.4-8.0 µm (cf. 6-7 µm of P. paradubitabilis). Comparison with P. intermedia reveals clear differences in stria density (7-10 in 10 µm vs. 5-6 in 10 µm in P. paradubitabilis) and apices shape (capitate in P. intermedia vs. obtusely rounded in P. paradubitabilis). The new species can also be confused with P. angulosa; however, P. angulosa have larger valves (42-53 µm length and 9.7-10.3 µm width vs., respectively, 39-41 µm and 6-7 µm in P. paradubitabilis) with the wide axial area (cf. the narrow axial area in P. paradubitabilis).
Altogether these lines of evidence suggest that FA repertoires in microalgae substantively depend on the habitat. In soil diatom strains, FA repertoires are dominated by saturated and monounsaturated acids. The algae isolated from brackish-water habitats present with higher content of long-chain polyunsaturated FA, whereas freshwater strains accumulate both saturated/monounsaturated and polyunsaturated FA. At the same time, algal taxa at different levels may show priorities toward the accumulation of certain types of fatty acids [6]. Proper assessment of the biochemical conservatism in species and strains of Pinnularia will require dedicated research on fatty acid composition for these algae from different habitats. FA profiles of the studied strains are heavily dominated by saturated acids (with a maximum of 92.7% in P. minigibba VP284) or saturated/monounsaturated acids (with the highest total in P. minigibba VP284 and P. vietnamogibba VP294). It might be sensible, therefore, to consider Pinnularia strains as potential producers of 16:0 palmitic and 16:1n-7 palmitoleic acids used in the production of biofuels [71].

Description of New Species
Pinnularia minigibba Kezlya, Maltsev, Krivova et Kulikovskiy sp. nov. (Figures 3 and 4). Diagnosis: Valves linear, sides slightly concave in the middle, length 40-43 µm, width 7-8 µm, length-to-width about 5.3-5.8, apices subcapitate, bluntly wedge-shaped to rounded width 4.8-5.5 µm (Figure 3). Raphe lateral, outer fissure straight or very weakly curved, central pores very small, slightly unilaterally deflected, drop-shaped, terminal fissures difficult to resolve, sickle-shaped, in the very small terminal area surrounded by striae at the poles. The axial area narrows, to 1/4 the width of the valve, continuously widening from the ends to the central area. Central area large, rhombic with a broad slightly asymmetric fascia, accompanied by ghost striae irregular in a shape, usually larger in the ventral side. Sometimes the ghost striae difficult to resolve in the LM. Striae radiate in the middle and convergent at the ends 9-10 in 10 µm.
Ultrastructure: In external views raphe branches straight ( Figure 4A-C), slightly unilaterally deflected in the proximal part, terminated small, drop-shaped ends. The distal raphe ends are sickle-shaped and extend to the valve margin ( Figure 4C). The striae are composed of one large alveolus, each alveolus is composed of five rows of small areolae. Diagnosis: Valves linear, sides slightly concave in the middle, length 40-43 µm, width 7-8 µm, length-to-width about 5.3-5.8, apices subcapitate, bluntly wedge-shaped to rounded width 4.8-5.5 µm (Figure 3). Raphe lateral, outer fissure straight or very weakly curved, central pores very small, slightly unilaterally deflected, drop-shaped, terminal fissures difficult to resolve, sickle-shaped, in the very small terminal area surrounded by striae at the poles. The axial area narrows, to 1/4 the width of the valve, continuously widening from the ends to the central area. Central area large, rhombic with a broad slightly asymmetric fascia, accompanied by ghost striae irregular in a shape, usually larger in the ventral side. Sometimes the ghost striae difficult to resolve in the LM. Striae radiate in the middle and convergent at the ends 9-10 in 10 µm. Ultrastructure: In external views raphe branches straight ( Figure 4A-C), slightly unilaterally deflected in the proximal part, terminated small, drop-shaped ends. The distal raphe ends are sickle-shaped and extend to the valve margin ( Figure 4C). The striae are composed of one large alveolus, each alveolus is composed of five rows of small areolae.  Sequence data: partial 18S rDNA gene sequence comprising V4 domain sequence (GenBank accession number OL739454) and partial rbcL sequence (GenBank accession number OL704397) for the strain VP284. Holotype: Slide no. 07042 (Holotype represented by Figure 3A) (Figures 5 and 6). Diagnosis: Valves linear to linear elliptical with slightly convex or parallel sides, tapering to the broadly rounded apices, length 34-54 µm, width 7-8 µm, apices width 5 µm, length-to-width in small size valves about 4.7, in large size valves 6.4-6.75 ( Figure 5). Raphe lateral, outer fissure straight, central pores very small, drop-shaped, slightly unilaterally deflected, terminal pores sickle-shaped. The axial area is moderately broad about 1/4 the width of the valve, linear or widening from the end to the central part of the valve, in small size valves lanceolate. Central area is large, rhombic with a broad slightly asymmetric fascia, accompanied by four ghost striae irregular in a shape, usually larger on the ventral side. Often the ghost striae are difficult to resolve in the LM. Striae radiate in the middle and strongly convergent at the ends 10-11 in 10 µm.  5 and 6) Diagnosis: Valves linear to linear elliptical with slightly convex or parallel sides, tapering to the broadly rounded apices, length 34-54 µm, width 7-8 µm, apices width 5 µm, length-to-width in small size valves about 4.7, in large size valves 6.4-6.75 ( Figure 5). Raphe lateral, outer fissure straight, central pores very small, drop-shaped, slightly unilaterally deflected, terminal pores sickle-shaped. The axial area is moderately broad about 1/4 the width of the valve, linear or widening from the end to the central part of the valve, in small size valves lanceolate. Central area is large, rhombic with a broad slightly asymmetric fascia, accompanied by four ghost striae irregular in a shape, usually larger on the ventral side. Often the ghost striae are difficult to resolve in the LM. Striae radiate in the middle and strongly convergent at the ends 10-11 in 10 µm.  Ultrastructure: In external views raphe branches straight ( Figure 6A-C), slightly unilaterally deflected in the proximal part, terminated small, drop-shaped ends. The distal raphe ends are sickle-shaped and extend to the valve margin ( Figure 6C). The striae are composed of one large alveolus, each alveolus is composed of five rows of small areolae. In internal views ( Figure 6D-F), the raphe is straight, proximal raphe endings are connected and represent a continuous slit, in the middle of the central area near a raphe is a well-developed unilaterally inflated central nodule ( Figure 6E). On either side of the central nodule are ghost striae unequal, irregular in the shape, larger on the ventral side. Distal raphe ends straight, terminate on small helictoglossae ( Figure 6F). The alveoli are open.
Holotype: Slide no. 07052 (Holotype represented by Figure 5A  Ultrastructure: In external views raphe branches straight ( Figure 6A-C), slightly unilaterally deflected in the proximal part, terminated small, drop-shaped ends. The distal raphe ends are sickle-shaped and extend to the valve margin ( Figure 6C). The striae are composed of one large alveolus, each alveolus is composed of five rows of small areolae.
In internal views ( Figure 6D-F), the raphe is straight, proximal raphe endings are connected and represent a continuous slit, in the middle of the central area near a raphe is a well-developed unilaterally inflated central nodule ( Figure 6E). On either side of the central nodule are ghost striae unequal, irregular in the shape, larger on the ventral side. Distal raphe ends straight, terminate on small helictoglossae ( Figure 6F Isotype: Slide no. 07052a, deposited in the collection of MHA, Main Botanical Garden RAS. Sequence data: partial 18S rDNA gene sequences comprising V4 domain sequence (GenBank accession numbers: OL739456 for VP290, OL739458 for VP294) and partial rbcL sequences (GenBank accession numbers: OL704399 for VP290, OL704401 for VP294).
Etymology: The specific epithet refers to the name of the country (Vietnam) where this species comes from and the name of a related complex species Pinnularia gibba.
Distribution: As yet known only from the type locality. Pinnularia microgibba Kezlya, Maltsev, Krivova et Kulikovskiy sp. nov. (Figures 7 and 8). Diagnosis: Valves narrow-linear, sides slightly concave in the middle, length 35-40 µm, width 5.5-6.0 µm, length-to-width about 6.5-6.6, apices subcapitate, bluntly wedge-shaped to broadly rounded, slightly smaller valve width 3.8-4.5 µm (Figure 7). Raphe lateral, outer fissure straight or very weakly curved, central pores very small, slightly unilaterally deflected, drop-shaped, terminal fissures sickle-shaped. Axial area narrow, linear, slightly widening to the central part of the valve. Central area large, rhombic with a broad slightly asymmetric fascia, accompanied by four ghost striae irregular in a shape, usually larger on the ventral side. Sometimes the ghost striae are difficult to resolve in the LM. Striae parallel or slightly radiate in the middle and convergent at the ends 11-12 in 10 µm. Isotype: Slide no. 07052a, deposited in the collection of MHA, Main Botanical Garden RAS.
Etymology: The specific epithet refers to the name of the country (Vietnam) where this species comes from and the name of a related complex species Pinnularia gibba.
Distribution: As yet known only from the type locality. Pinnularia microgibba Kezlya, Maltsev, Krivova et Kulikovskiy sp. nov. (Figures 7  and 8) Diagnosis: Valves narrow-linear, sides slightly concave in the middle, length 35-40 µm, width 5.5-6.0 µm, length-to-width about 6.5-6.6, apices subcapitate, bluntly wedgeshaped to broadly rounded, slightly smaller valve width 3.8-4.5 µm (Figure 7). Raphe lateral, outer fissure straight or very weakly curved, central pores very small, slightly unilaterally deflected, drop-shaped, terminal fissures sickle-shaped. Axial area narrow, linear, slightly widening to the central part of the valve. Central area large, rhombic with a broad slightly asymmetric fascia, accompanied by four ghost striae irregular in a shape, usually larger on the ventral side. Sometimes the ghost striae are difficult to resolve in the LM. Striae parallel or slightly radiate in the middle and convergent at the ends 11-12 in 10 µm.  Ultrastructure: In external views raphe branches straight ( Figure 8A-C), slightly unilaterally deflected in the proximal part, terminated small, drop-shaped ends. The distal raphe ends are sickle-shaped and extend to the valve margin ( Figure 8C). The striae are composed of one large alveolus, each alveolus is composed of five rows of small areolae. In internal views ( Figure 8D-F), the raphe is straight, proximal raphe endings are connected and represent a continuous slit, with a well-developed unilaterally inflated central nodule in the middle of the central area. On either side of the central nodule are hollowed areas unequal, irregular in the shape, larger on the ventral side ( Figure 8E). Distal raphe ends straight, terminate on small helictoglossae ( Figure 8F). The alveoli are open.
Holotype: Slide no. 07047 (Holotype represented by Figure 7A  Ultrastructure: In external views raphe branches straight ( Figure 8A-C), slightly unilaterally deflected in the proximal part, terminated small, drop-shaped ends. The distal raphe ends are sickle-shaped and extend to the valve margin ( Figure 8C). The striae are composed of one large alveolus, each alveolus is composed of five rows of small areolae.
In internal views ( Figure 8D-F), the raphe is straight, proximal raphe endings are connected and represent a continuous slit, with a well-developed unilaterally inflated central nodule in the middle of the central area. On either side of the central nodule are hollowed areas unequal, irregular in the shape, larger on the ventral side ( Figure 8E). Distal raphe ends straight, terminate on small helictoglossae ( Figure 8F Isotype: Slide no. 07047a, deposited in the collection of MHA, Main Botanical Garden RAS. Sequence data: partial 18S rDNA gene sequences comprising V4 domain sequence (GenBank accession numbers: OL739455 for VP289, OL739457 for VP292) and partial rbcL sequences (GenBank accession numbers: OL704398 for VP289, OL704400 for VP292).
Etymology: The specific epithet refers to particularly small dimensions of the valves (micro-) and the name of a similar complex species Pinnularia gibba.
Distribution: Known from the type locality and samples KT39, KT80 (Table 1). Pinnularia insolita Kezlya, Maltsev, Krivova et Kulikovskiy sp. nov. (Figures 9 and 10). Diagnosis: Valves linear, sides slightly concave in the middle, length 50-52 µm, width 7-7.5 µm, length-to-width about 6.8-7.3, apices relatively long rostrate, finally broadly rounded width 4 µm (Figure 9). Raphe lateral, outer fissure straight, central pores relatively large, slightly unilaterally deflected, drop-shaped, terminal fissures not distinct, in the very small terminal area surrounded by striae at the poles. Axial area moderately broad, to 1/3 the width of the valve, widening from the end to the central part of the valve. The central area is very large with a broad slightly asymmetric fascia widening towards the valve margin, always bigger than the width of the valve. Striae radiate in the middle and strongly convergent at the ends 11-12 in 10 µm.
Ultrastructure: In external views raphe branches straight ( Figure 10A-C). Proximal raphe ends are relatively long, drop-shaped, slightly unilaterally deflected. The distal raphe ends are sickle-shaped and extend to the valve margin ( Figure 10C). The striae are very closely spaced and composed of one large alveolus; each alveolus is composed of six to seven rows of small areolae.
In internal views ( Figure 10D-F) the raphe is straight, proximal raphe endings are connected and represent a continuous slit, terminating near a base well-developed unilaterally inflated central nodule. Distal raphe ends slightly deflected to one side and terminate on small helictoglossae ( Figure 10F) Diagnosis: Valves linear, sides slightly concave in the middle, length 50-52 µm, width 7-7.5 µm, length-to-width about 6.8-7.3, apices relatively long rostrate, finally broadly rounded width 4 µm (Figure 9). Raphe lateral, outer fissure straight, central pores relatively large, slightly unilaterally deflected, drop-shaped, terminal fissures not distinct, in the very small terminal area surrounded by striae at the poles. Axial area moderately broad, to 1/3 the width of the valve, widening from the end to the central part of the valve. The central area is very large with a broad slightly asymmetric fascia widening towards the valve margin, always bigger than the width of the valve. Striae radiate in the middle and strongly convergent at the ends 11-12 in 10 µm. Ultrastructure: In external views raphe branches straight ( Figure 10A-C). Proximal raphe ends are relatively long, drop-shaped, slightly unilaterally deflected. The distal raphe ends are sickle-shaped and extend to the valve margin ( Figure 10C). The striae are very closely spaced and composed of one large alveolus; each alveolus is composed of six  In internal views ( Figure 10D-F) the raphe is straight, proximal raphe endings are connected and represent a continuous slit, terminating near a base well-developed unilaterally inflated central nodule. Distal raphe ends slightly deflected to one side and terminate on small helictoglossae ( Figure 10F). The alveoli are open.
Holotype: Slide no. 07038 (Holotype represented by Figure 9A)   Holotype: Slide no. 07038 (Holotype represented by Figure 9A)  Etymology: The specific epithet reflects the unusual shapes of the valves with narrowed apices and the central area.
Distribution: As yet known only from the type locality. Pinnularia ministomatophora Kezlya, Maltsev, Krivova et Kulikovskiy sp. nov. (Figures 11 and 12). length-to-width about 5.9-6.7 ( Figure 11). Raphe lateral, outer fissure straight or slightly broadly curved, central pores small, drop-shaped, straight or slightly curved on one side, terminal fissures long, bayonet shaped, lying in an elongate-elliptic terminal area. Axial area narrow, up to ¼ the width of the valve, widened from the ends to the rhombic central area which is differentiated in the valve middle with a broad symmetric or slightly asymmetric fascia. On either side of the central nodule in the central area are hollows in the valve surfaсe irregular in the shape (Figures 11A-G,I-K and 12A), in some valves, the hollows are difficult to resolve in LM. Striae 10-11 in 10 µm, strongly radiate in the middle and strongly convergent at the ends. Ultrastructure: In external views the raphe is straight. Proximal raphe ends are dropshaped, slightly unilaterally deflected. The distal raphe ends are sickle-shaped and extend to the valve margin ( Figure 12A). The striae are very closely spaced and composed of one large alveolus; each alveolus is composed of five to seven rows of small areolae. In the central area on either side of the central nodule are hollows in the valve surfaсe irregular in the shape. In internal views ( Figure 12B,D,E) the raphe is straight. The raphe branches are straight with short, bent, proximal raphe endings, terminating on a bad-developed unilaterally inflated central nodule. Distal raphe ends slightly deflected to one side and terminate on small helictoglossae ( Figure 12D). The alveoli are open.
Holotype: Slide no. 07116 (Holotype represented by Figure 11A)   Diagnosis: Valves are linear in outline, sides slightly convex to parallel, apices broadly rostrate to broadly rounded, subcapitate, length 44-57 µm, width 7-9.5 µm, length-to-width about 5.9-6.7 ( Figure 11). Raphe lateral, outer fissure straight or slightly broadly curved, central pores small, drop-shaped, straight or slightly curved on one side, terminal fissures long, bayonet shaped, lying in an elongate-elliptic terminal area. Axial area narrow, up to 1/4 the width of the valve, widened from the ends to the rhombic central area which is differentiated in the valve middle with a broad symmetric or slightly asymmetric fascia. On either side of the central nodule in the central area are hollows in the valve surface irregular in the shape ( Figure 11A-G,I-K and Figure 12A), in some valves, the hollows are difficult to resolve in LM. Striae 10-11 in 10 µm, strongly radiate in the middle and strongly convergent at the ends.
Ultrastructure: In external views the raphe is straight. Proximal raphe ends are dropshaped, slightly unilaterally deflected. The distal raphe ends are sickle-shaped and extend to the valve margin ( Figure 12A). The striae are very closely spaced and composed of one large alveolus; each alveolus is composed of five to seven rows of small areolae. In the central area on either side of the central nodule are hollows in the valve surface irregular in the shape.
In internal views ( Figure 12B,D,E) the raphe is straight. The raphe branches are straight with short, bent, proximal raphe endings, terminating on a bad-developed unilaterally inflated central nodule. Distal raphe ends slightly deflected to one side and terminate on small helictoglossae ( Figure 12D). The alveoli are open.
Holotype: Slide no. 07116 (Holotype represented by Figure 11A)  Etymology: The specific epithet refers to the small dimensions of the valves (mini-) and the name of a similar species Pinnularia stomatophora.
Etymology: The specific epithet refers to the small dimensions of the valves (mini-) and the name of a similar species Pinnularia stomatophora.
Distribution: Known from the type locality and sample KT70 (Table 1). Pinnularia paradubitabilis Kezlya, Maltsev, Krivova et Kulikovskiy sp. nov. (Figures 13 and 14) Diagnosis: Valves outline linear, margins parallel or slightly concave, apices bluntly rounded, length 39-41 µm, width 6-7 µm ( Figure 13). Raphe moderately lateral the outer fissures straight to weakly curved, central pores small, drop-shaped, slightly curved on one side, terminal fissures long, sickle-shaped. Axial area narrow, linear, central area broad, more than the width of valve forming a broad fascia. Striae 5-6 in 10 µm, radiate or parallel in the middle, convergent (or sometimes parallel) at the ends. Ultrastructure: In external views, the raphe branches are straight or broadly rounded, clearly unilaterally deflected in the proximal part ( Figure 14A-C). Proximal raphe ends are drop-shaped. The distal raphe ends are sickle-shaped and extend to the valve margin ( Figure 14C). The striae are composed of one large alveolus; each alveolus is composed of 12 to 15 rows of small areolae. The striae continued shortly on the valve margins. In internal views ( Figure 14D-F) the raphe straight, proximal raphe branches are straight or short bent, terminating on a well-developed unilaterally inflated central nodule. Distal raphe ends slightly deflected to one side and terminate on small helictoglossae ( Figure 14F). The  Sequence data: partial 18S rDNA gene sequence comprising V4 domain sequence (GenBank accession number OL739452) and partial rbcL sequence (GenBank accession number OL704395) for the strain VP236.
Etymology: The specific epithet refers to the resemblance to Pinnularia dubitabilis Hustedt. and name of country, when the species were found. Distribution: Known from the type locality and samples KT19, KT 40,974,965 (Table  1). This species was identified by Niels Foged (p. 355, pl. XII, Figures 18 and 19 of [72]) as Diagnosis: Valves outline linear, margins parallel or slightly concave, apices bluntly rounded, length 39-41 µm, width 6-7 µm (Figure 13). Raphe moderately lateral the outer fissures straight to weakly curved, central pores small, drop-shaped, slightly curved on one side, terminal fissures long, sickle-shaped. Axial area narrow, linear, central area broad, more than the width of valve forming a broad fascia. Striae 5-6 in 10 µm, radiate or parallel in the middle, convergent (or sometimes parallel) at the ends.
Ultrastructure: In external views, the raphe branches are straight or broadly rounded, clearly unilaterally deflected in the proximal part ( Figure 14A-C). Proximal raphe ends are drop-shaped. The distal raphe ends are sickle-shaped and extend to the valve margin ( Figure 14C). The striae are composed of one large alveolus; each alveolus is composed of 12 to 15 rows of small areolae. The striae continued shortly on the valve margins. In internal views ( Figure 14D-F) the raphe straight, proximal raphe branches are straight or short bent, terminating on a well-developed unilaterally inflated central nodule. Distal raphe ends slightly deflected to one side and terminate on small helictoglossae ( Figure 14F).
The alveoli are open.