European Journal of Phycology ISSN: 0967-0262 (Print) 1469-4433 (Online) Journal homepage: https://www.tandfonline.com/loi/tejp20 Auxosporulation in Chaetoceros acadianus sp. nov. (Bacillariophyceae), a new member of the Section Compressa Irena Kaczmarska, Brajogopal Samanta, James M. Ehrman & Ellen M. A. Porcher To cite this article: Irena Kaczmarska, Brajogopal Samanta, James M. Ehrman & Ellen M. A. Porcher (2019): Auxosporulation in Chaetoceros�acadianus sp. nov. (Bacillariophyceae), a new member of the Section Compressa, European Journal of Phycology, DOI: 10.1080/09670262.2018.1542031 To link to this article: https://doi.org/10.1080/09670262.2018.1542031 View supplementary material Published online: 21 Feb 2019. Submit your article to this journal View Crossmark data Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=tejp20 EUROPEAN JOURNAL OF PHYCOLOGY https://doi.org/10.1080/09670262.2018.1542031 Auxosporulation in Chaetoceros acadianus sp. nov. (Bacillariophyceae), a new member of the Section Compressa Irena Kaczmarskaa, Brajogopal Samantaa, James M. Ehrmanb and Ellen M. A. Porchera a Department of Biology, Mount Allison University, Sackville, New Brunswick E4L 1G7, Canada; bDigital Microscopy Facility, Mount Allison University, Sackville, New Brunswick E4L 1G7, Canada ABSTRACT We document the fine structure of auxospores in a Chaetoceros species isolated from the Acadian coast of New Brunswick, Canada. Auxospore development in this species occurs in a terminal rather than lateral position, a characteristic never before observed in this genus. Our observations suggest that auxosporulation was uniparental, probably an extreme form of autogamy with sister nuclei fusing following meiosis II. Mature auxospores were adze-shaped to sub-globular and contained both scales and transverse perizonia in their walls. The transverse perizonial band structure was similar to longitudinal perizonial bands found in other species of Chaetoceros and differed from the pinnate bands of pennate transverse perizonia, which consisted of a central rib and bilateral fimbria. Instead, the band structure in C. acadianus was more similar to unilateral fimbriate bands in cymatosiroids. We also propose that our diatom represents a species new to science and is a member of the Chaetoceros Section Compressa. We provide its morphological, molecular and reproductive characterization. ARTICLE HISTORY Received 8 June 2018 ; Revised 16 September 2018 ; Accepted 21 September 2018 KEYWORDS Autogamy; auxospore structure; compensatory base changes; ITS2 secondary structure; phylogeny; Section Compressa Introduction In most diatoms, sexuality is an obligatory life stage because it is an integral part of the process by which populations reverse the mean cell-size reduction that occurs during the vegetative phase of the life cycle. Large post-sexual cells of the new generation are then ready to propagate themselves vegetatively again (Round et al., 1990). Our understanding of sexual reproduction and auxospore development in centric, and polar centric diatoms in particular, has lagged behind that of the raphid pennates. Sexuality in these diatoms has not received much scientific attention since the seminal work of von Stosch (1982) and co-workers (von Stosch et al., 1973; Drebes, 1977), with some notable recent exceptions (Idei et al., 2015; Davidovich et al., 2017; Samanta et al., 2017, 2018). Among these, auxosporulation in several Chaetoceros Ehrenberg species was investigated in vitro and relatively frequently in natural samples. The sexual stages of at least 12 Chaetoceros species have been reported and/or illustrated thus far, and auxospores were found in a lateral position on the colonial chain of cells in all of them. Unfortunately, nearly all the reports have relied exclusively on light microscopy. Consequently, little is known about the structures involved beyond the typical position of the auxospores on the cell chain and their overall multipolar shape. The most recent report using scanning electron microscopy (SEM; Assmy et al., 2008) showed a few developmental CONTACT Irena Kaczmarska © 2019 British Phycological Society Published online 21 Feb 2019 iehrman@mta.ca stages of auxospore growth from a natural population of C. dichaeta Ehrenberg, further illustrating the point already made by von Stosch (1982): that the auxospore walls of Chaetoceros are challenging to examine. The fine structure and development of the auxospore wall is known from only two species of Chaetoceros (von Stosch, 1982). As in other polar centrics, these auxospores were neither spherical nor tubular when mature. They developed in three stages. In the first, the globular stage, auxospores grew isodiametrically and the siliceous elements of the wall consisted of incunabular scales. In contrast, the second growth stage was anisodiametric; it built upon and integrated into the auxospore wall the previously produced incunabular scaly wall (as a ventral part of the cell) but added a new structure, the perizonium, made of open and closed bands. The latter constituted the cover of the dorsal part of the auxospore. The fine structure was only resolvable on the final, longitudinal perizonial band circumscribing the widest region of the auxospore (separating the dorsal and ventral sides of the cell) and was found to bear fimbria. The third stage deposited the initial frustule inside the auxospore walls. The diatom genus Chaetoceros is relatively speciesrich and an abundant contributor to the meroplankton of coastal waters worldwide. The three-dimensionally complex nature of their colonial chains, lightly silicified valves, and relatively monotonous micro-architecture of the frustules make them challenging to identify. Nonetheless, several more recent taxonomic (Rines & 2 I. KACZMARSKA ET AL. Hargraves, 1988; Hernández-Becerril & Flores Granados, 1998; Rines, 1999; Li et al., 2017) and phylogenetic (Rines & Theriot, 2003; Kooistra et al., 2010; Li et al., 2015) works significantly contributed to bettering our understanding of the diversity of this genus, including that of the Section Compressa Ostenfeld (Chamnansinp et al., 2015). Similar to other challenging taxa, the application of a multipronged approach to their taxonomy revealed species diversity greater than previously anticipated. Chaetoceros Section Compressa contains species with two types of intercalary setae. Less frequent and nontypical setae differ from the others by being heavily silicified and undulated (or ‘contorted’ sensu Rines & Hargraves, 1988) among the straight and thinly silicified kind (Hustedt, 1962: 684). Here we present auxospore development unlike any other species of the genus Chaetoceros, in diatom isolates from the Acadian coast on New Brunswick, Canada. We place our findings in the context of current understanding of auxospore development among Chaetocerotales. Based on the evidence from colony, frustule and valve morphology, nuclear encoded Internal Transcribed Spacer 2 (ITS2) secondary structure of the clones, and molecular analyses, we propose that these diatoms represent a species new to science. Materials and methods Sampling and clonal culture establishment A coastal seawater sample was collected from Petit-Cap, New Brunswick, Canada (46.1996°N, 64.1647°W) in July 2017. Single clonal chains of Chaetoceros were isolated by micro-pipetting (Andersen, 2005) using a Zeiss Axiovert 200 inverted light microscope (Carl Zeiss, Oberkochen, Germany). Each chain was washed several times with sterile seawater until free from other biota, placed in a well of a 12-well culture plate (Corning Incorporated, Corning, New York, USA) with f/2 medium, and kept at 15°C with a photoperiod of 12:12 hours light:dark and an irradiance of ~ 26 µmol photons m–2 s–1 for up to 14 days for growth. Two monoclonal cultures (Chaeto1 and Chaeto2) of Chaetoceros were established and then scaled up to 50 ml volume flasks and grown under the same conditions. Cultures were transferred to fresh growth media weekly to maintain exponential growth. Light microscopic observation of sexual stages To examine the nuclear behaviour during successive stages of sexual reproduction, cells were stained with Vectashield Mounting Medium with DAPI (Vector Laboratories, Burlingame, California, USA) as per manufacturer’s instructions. Wall silicification was visualized by tracing incorporation of PDMPO (2-(4-pyridyl)-5-((4-(2 dimethylaminoethylaminocarbamoyl)methoxy)phenyl)oxazole); Thermo Fisher Scientific, Waltham, Massachusetts, USA) into developing auxospores. PDMPO incorporated into silica fluoresces green (550 nm) when excited by violet (405 nm) light. The final concentration of PDMPO used was 0.125 µM (LeBlanc & Hutchins, 2005). Brightfield, DAPI, and PDMPO epifluorescence light microscopy were performed using a Zeiss AxioSkop 2 plus microscope equipped with an AxioCam HR colour camera and HBO 100 mercury vapour illumination source. Scanning electron microscopy (SEM) and energy dispersive x-ray spectroscopy (EDS) Vegetative frustules from each of the clonal cultures were prepared for SEM/EDS examination following Kaczmarska et al. (2005). Lightly and non-silicified cells were fixed for a minimum of 1 h with 2.5% glutaraldehyde in f/2 media, then rinsed 4× with distilled water (~100 ml) every 10 min, with gentle vacuum in a filtration tower onto a 25 mm diameter, 1 µm pore size polycarbonate filter (Sterlitech Corporation, Kent, Washington, USA). Filters were removed from the tower while still moist and stacked with stainless steel washers (OD 25 mm, ID 15.5 mm) and an additional filter in the order washer/specimen filter/washer/additional filter/washer and clamped together with a pair of 0.75 inch (19 mm) stainless steel foldover binder clips (Staples® Model 24169-CA) to contain the specimen side of the filter in a flow-through chamber. Binder clip handles were removed to allow the assembly to fit into the critical point dryer chamber (see below). Specimens were then post-fixed in 1% osmium tetroxide for 1 h, and dehydrated using 10 min changes of 20%, 50%, 70%, 85%, 95% ethanol:distilled water followed by 4 × 10 min changes of 100% anhydrous ethanol. Filter assemblies were dried with liquid CO2 in a Denton DCP-1 critical point dryer (Denton Vacuum, Moorestown, New Jersey, USA). For examination of auxospore wall components, specimens were prepared in two ways. In the first method, samples were rinsed of fixative onto filters as above, resuspended in ~1 ml distilled water and subjected to a brief (5 min) acid wash (15 ml 1:1 sulphuric: nitric acid) with no heating other than that produced by mixing the two acids. The acid solution was then diluted in 100 ml distilled water, transferred to the filtration tower containing a fresh filter, rinsed with 250 ml distilled water and air-dried. The second preparation method omitted the acid treatment and the sample was simply washed with 250 ml distilled water onto the filter and air-dried. Filters bearing specimens prepared by all methods were mounted on aluminium stubs with double-sided tape, rimmed with colloidal carbon, and coated with ~15 nm of gold using a Hummer 6.2 sputtering unit (Anatech Ltd, Union City, California, USA). EUROPEAN JOURNAL OF PHYCOLOGY Images were acquired using a Hitachi SU3500 SEM (Hitachi High Technologies, Toronto, Canada) at a working distance of 5 mm and 10 kV accelerating voltage. EDS was performed with the same instrument equipped with an Oxford AZtec/X-Max 20 EDS system (Oxford Instruments, High Wycombe, UK) at 10 mm working distance. Since the only element of interest in this study was silicon (Si-Kα, X-ray energy 1.74 keV), an accelerating voltage of 10 kV provided sufficient overvoltage for efficient X-ray excitation. Spectra were acquired for 100 s (dead time corrected) at 0.1 nA beam current, energy range 0–10 keV into 1024 channels. The EDS spectra were collected from intact and unobstructed structures and/or auxospores. Spectra from the polycarbonate support filter adjacent to the auxospores were also routinely taken and showed no remote excitation from neighbouring siliceous components (if present) at distances as close as 3 µm. Terminology Frustule morphology described here follows Brunel (1972) and Rines & Hargraves (1988). Sexual stages and structures use the terminology detailed in Kaczmarska et al. (2013). DNA extraction, PCR and Sequencing Clonal cultures were harvested from exponential growth phase and biomasses were concentrated by centrifugation. DNA was extracted using an UltraClean® Soil DNA Isolation Kit (Qiagen Sciences, Germantown, Maryland, USA (formerly Mo Bio Laboratories)) as per manufacturer’s instructions. About 700 bp of the nuclear-encoded ITS1-5.8S-ITS2 rDNA region was amplified using primers SR12cF and ITS4 (White et al., 1990; Moniz & Kaczmarska, 2010). A fragment (~1200 bp long) of plastid-encoded large subunit of ribulose 1,5-bisphosphate carboxylase/oxygenase (rbcL) was amplified using primers DtrbcL2F and DtrbcL2R (MacGillivary & Kaczmarska, 2011). The 25 µl volume of each PCR reaction contained 12.5 µl of GoTaq Mastermix (Promega, Madison, Wisconsin, USA), 0.25 µl each of forward and reverse primers (final concentration of 100 nM), 10 µl of DEPC treated water and 2 µl (~2–20 ng) of DNA template. The PCR conditions were as follows: initial denaturation at 95°C for 5 min followed by 35 cycles of denaturation at 95°C for 30 s, annealing at 49/50°C for 30 s, and elongation at 72°C for 1 min, and a final extension step at 72°C for 5 min. PCR products were visualized in a 1.2% agarose gel. Purification and sequencing of the PCR products were performed at Nanuq (McGill University 3 and Genome Québec, Montréal, Canada) following their standard protocols. Molecular analyses Chromatograms were checked using BioEdit v7.0 (Hall, 1999) for ambiguities and errors before undertaking further analysis. Subsequently, sequences of the clonal isolates were compared with published cultured sequences from GenBank [NCBI] (Benson et al., 2013) using the BLAST tool (http://blast.ncbi. nlm.nih.gov/Blast.cgi; Altschul et al., 1990). Fragments of the ~300 bp long 5.8S+ITS2 barcode region (Moniz & Kaczmarska, 2010) and 540 bp long rbcL region (MacGillivary & Kaczmarska, 2011) were used for phylogenetic analysis. Sequences of Chaetoceros from GenBank were included in the alignment and were aligned using MUSCLE (https:// www.ebi.ac.uk/Tools/msa/muscle/; Edgar, 2004). In addition, 5.8S+ITS2 and rbcL barcode regions of five additional Chaetoceros strains (C. affinis Lauder, C. decipiens Cleve, C. radicans Schütt, Chaetoceros sp. and C. contortus Schütt) were sequenced as part of this study and were also included in the alignment; GenBank and BOLD accession numbers for all are given in Supplementary table 1. The highly variable part of the 5.8S+ITS2 barcode region was annotated based on the Chaetoceros ITS2 rRNA secondary structure predicted for C. neogracilis van Landingham (Balzano et al., 2017). The final alignment of the 5.8S+ITS2 barcode region included 39 Chaetoceros sequences and 349 nucleotide positions. The appropriate model for the dataset was determined using jModelTest V2.1.2 (Darriba et al., 2012) according to Bayesian information criterion (BIC; Posada & Buckley, 2004). The HKY+G (−lnL = 4008.1265) and GTR+I = G (−lnL = 2766.3919) were the most optimal models for 5.8S+ITS2 and rbcL datasets, respectively. Bayesian inference (BI) was performed using MrBayes v3.2 (Ronquist et al., 2012). A four-chain run for 20 000 000 generations was used and trees were sampled every 5000 generations. Posterior probabilities (PP) were estimated with 50% burn-in, and a majority rule consensus tree was constructed. An online version of PHYML (Phylogenetic inferences using Maximum Likelihood; www.atgc-montpellier.fr/phyml/; Guindon et al., 2010) was used to construct the ML tree by selecting AIC (Akaike Information Criterion; Posada & Buckley, 2004) for the appropriate substitution model (Lefort et al., 2017) and NNI (NearestNeighbour Interchanges; Felsenstein, 2004) for branch swapping. Bootstrap supports (Felsenstein, 1985) were obtained based on 1000 replicates 4 I. KACZMARSKA ET AL. for ML analysis. Sequences of Attheya longicornis Crawford & Gardner, and Brockmanniella brockmannii (Hustedt) Hasle, Stosch & Syvertsen were used as outgroups for phylogenetic analyses. ITS2 secondary structure prediction and compensatory base substitution analyses To evaluate our Chaetoceros clones, we reconstructed the secondary structures of their ITS2 and compared them to closely related congeners. The ITS2 boundaries were annotated using Hidden Markov Models of the flanking 5.8S and 28S regions (Eddy, 1998; Keller et al., 2009). All ITS2 secondary structures were predicted by comparing RNA folding patterns of complete ITS2 sequences and, if necessary, of single helices, using the Mfold web server (http://unafold.rna.albany.edu/?q= mfold/RNA-Folding-Form2.3; Zuker, 2003) with default parameters except for temperature. The folding temperatures were set according to the culture growth condition of the isolates, 15°C for our clonal isolates and 27°C for other Chaetoceros species (Chamnansinp et al., 2015). The helices were labelled according to Mai & Coleman (1997) and also compared with the published secondary structure of C. neogracilis (Balzano et al., 2017). For comparative analysis of compensatory base changes (CBCs) and hemi-CBCs (hCBCs) among the investigated taxa, each sequence and their individual secondary structure was aligned using the 4SALE program (Seibel et al., 2006, 2008). The compensatory base substitutions were counted based on the sequence structure alignment using CBCAnalyzer as implemented in 4SALE (Wolf et al., 2005). Results Auxospore development and wall structure The first stages of auxospore development observed in light microscope preparations (LM) showed pervalvar expansion of the mother cells, bulging out in the girdle region (Figs 1, 2) and rounding up the entire cell contents (Figs 1–4). An acytokinetic nuclear division followed, resulting in one of the nuclei becoming pyknotic (Figs 3, 4). One or both thecae of these induced cells were then discarded, but most of the sexualized cells with two nuclei encountered were separated from their native chains and free in the growth medium. When free of parental thecae, these cells appeared spherical (14–19 µm in diameter) and showed the deposition of siliceous elements in their cell walls, as evidenced by PDMPO staining (Figs 5, 6), and also in EDS spectra. Next, the auxospores grew anisodiametrically (Figs 7–14). This was most clearly evident when viewed in perfectly lateral orientation which showed the ventral (older, ruptured sphere) part of the auxospore to be smaller in circumference compared with the dome-shaped dorsal (younger) part (Fig. 9). Observed from the dorsal side, the dome outline was widely elliptical (11–21 µm in shorter axis and 12–24 µm in longer axis). The PDMPO incorporation indicated that these cells had their dorsal siliceous wall components deposited after the basal, ventral part was in place (green in Fig. 14; initiation of perizonium deposition is shown in Fig. 6, a subsurface intensively green strip) because it can be seen located below it. The dorsal part of the auxospore wall was multilayered (Fig. 7). In some orientations, the asymmetry of larger auxospores was subtle, making the auxospore broadly ellipsoidal. The larger auxospores had two or three nuclei (Figs 11–14) when viewed in orientations such that multiple DAPI stained nuclei could be conclusively separated. In mature auxospores the protoplast pulled away from the dorsal wall, retracting to the equatorial part of the cell (Figs 13, 14). Following acytokinetic mitosis and pyknosis of the supernumerary nucleus (Fig. 16) the initial valve developed, with the setae (Figs 15, 16). As the setae grew, the dome-shaped dorsal part of the auxospore wall was lost in some cells. Post-sexual cells resumed vegetative propagation as large-cell chains (Fig. 17). The earliest, smallest auxospore developmental stages recognizable in SEM (Figs 18–29) were spherical cells whose surface was covered with incunabular scales (Fig. 18; compare to Fig. 3), some with, while others without, an annulus. Scales had radial rows of ribs (Figs 19–21). When partially damaged, the auxospores showed some internal elements normally obscured by the surface cover (note multilayered dorsal auxospore wall in Fig. 7). The second, non-spherical stage of auxospore development built upon and integrated into the auxospore wall previously produced incunabular elements (retained to cover the ventral part of the cell) but added a new and larger structure, the perizonium (a system of bands). The perizonium covered the dorsal part of the auxospore. Larger incunabular scales and elongated scales (Figs 22–25) were found interspersed with smaller scales (of the type that were common on spherical auxospores; Figs 19, 21); mostly on the ventral part of larger auxospores. Individual incunabular scales or elongated scales could also be found overlaying perizonial bands, particularly the bands near the border between ventral and dorsal parts of the auxospore. We found transverse perizonial bands (Figs 24–26; compare to Figs 7, 14) but in SEM preparations could not determine with full certainty their spatial relationships to each other except that there were groups of bands (Figs 24–26, 29). Most of the bands were about the same width, but 1–2 bands of greater width were found on the dorsal part of the nonspherical auxospore among the narrower bands. The layers of perizonial bands observed on auxospores that maintained some of their 3-D integrity criss-crossed the dorsal part of the auxospore in various directions EUROPEAN JOURNAL OF PHYCOLOGY 5 Figs 1–17. Nuclear behaviour and the progress of silica deposition in the auxospore traced by DAPI (blue nuclei) and deposition of new silica by PDMPO (green cell wall elements) staining; light microscopy. Red is natural autofluorescence of the chloroplasts. Figs 1–2. Brightfield and DAPI epifluorescence images of the same very young auxospore mother cell after completion of the first meiotic division while within the maternal frustule; note two nuclei after meiosis I. Figs 3–4. Brightfield and DAPI epifluorescence images of the same free young, globular auxospore after the first meiotic division. One nucleus is smaller due to pyknosis. Figs 5–6. Brightfield, DAPI and PDMPO epifluorescence images of the same auxospore showing deposition of incunabular silica (diffuse green) over the entire surface of the cell, and the earliest perizonial bands developing below the surface (brighter green strip, arrowhead). One blue stained nucleus located between red chloroplasts. Figs 7–8. Brightfield and DAPI epifluorescence images of the same older anisometric auxospore showing multilayered perizonium (arrowhead), one nucleus, and several chloroplasts. Figs 9–10. Brightfield and DAPI epifluorescence images of the same anisometric auxospore after the second meiotic division, with three blue stained nuclei (arrow) after completing meiosis II. Note asymmetry of auxospore wall (arrowhead points to dorsal side of auxospore wall). Figs 11–12. Brightfield and DAPI epifluorescence images of the same trinucleated (arrow) anisometric auxospore after completing meiosis II. Diffuse blue areas are artefacts due to overstaining in areas surrounding chloroplasts. Figs 13–14. Brightfield and PDMPO epifluorescence images of an older auxospore with retracting protoplast below clearly outlined dorsal (perizonial) part (arrowhead, strongest green) and opposite, lighter silicified ventral part (thinner green lines) of the siliceous wall. The initial epivalve will develop below the retracted cell membrane. Figs 15–16. Brightfield and DAPI epifluorescence images of the same mature auxospore with initial epivalve in position opposite to the ventral, smaller part of the auxospore wall (arrowhead). Note blue nuclei, a supernumerary nucleus pyknotizing following acytokinetic mitosis and production of initial epivalve. Fig. 17. Brightfield image of a chain of post-sexual, large vegetative cells with numerous chloroplasts. Scale bars: Figs 1–17, 25 µm. 6 I. KACZMARSKA ET AL. (Figs 25, 29). The transverse perizonial bands had an asymmetric and ill-defined rim and very short (if any) fimbria on one side, but long fimbria on the opposite side (Figs 24, 28, 29). These bands overlapped very slightly, nearly abutting one another. There was also an indication that at least two layers of bands were present over some part of the auxospore, with layers intersecting the other (Figs 25, 29). Initial valves were circular, 11–20 µm in diameter, with short, often mispositioned setae of varied length and unusual number (Fig. 27). Some of these valves did not have rimoportulae (Fig. 27) while other valves had the rimoportulae misplaced. Initial valves were more delicately ribbed (at least twice the number of striae in typical vegetative valves), radially extending from the valve pattern centre with small pores between the ribs (Figs 26, 27). Genetic and morphometric analyses Morphometric and genetic analyses identified our clones as distinct from currently known members of the Chaetoceros Section Compressa. Consequently, we propose that they represent a species new to science and name it C. acadianus, see below. Molecular phylogenies and ITS2 secondary structure analyses In the ITS region (5.8S+partial ITS2; Fig. 30) phylogeny, our new Chaetoceros species clustered with other members of Section Compressa. In this cluster, our new species was recovered as a sister species to C. contortus var. ornatus. In the rbcL tree, our new species was recovered as a sister species of C. contortus as this was the only available rbcL sequence in GenBank from the Section Compressa (Supplementary fig. 1). These results confirmed molecularly that C. acadianus belongs to the Chaetoceros Section Compressa. Furthermore, we predicted the secondary structure for four additional members of this species complex (C. contortus, C. contortus var. ornatus (Schütt) Chamnansinp, Moestrup & Lundholm, C. compressus Lauder, and C. hirtisetus (Rines & Hargraves) Chamnansinp, Moestrup & Lundholm) for which complete ITS2 sequences were available, for comparison (Table 1). The 250 nucleotide-long ITS2 sequence of C. acadianus folded into four helices (Fig. 31). Five 100% conserved paired regions (except the base pairs where CBCs and hCBCs were found in these regions) could be recognized in all five taxa. These regions were: (1) the first 10 base pairs of Helix 1, (2) the first 14 base pairs of Helix 2 including U-U mismatch, (3) the 5 base pairs at the base of Helix 3, (4) the apical part of Helix 3 excluding the terminal loop containing 21 base pairs including super-conserved motif UGGU, and (5) the basal part of Helix 4, containing 3 base pairs. Interspecific comparisons identified seven CBCs in these conserved regions (Fig. 31) between C. acadianus and the other four taxa: one in the case of C. compressus, two for C. contortus, three for C. contortus var. ornatus, and four for C. hirtisetus. There was a comparable number of hCBCs in the conserved region of the ITS2 secondary structure (Table 1) between our new species and the other four taxa, and additional CBCs and hCBCs were found in less conserved parts of the structure. Chaetoceros acadianus Samanta and Porcher, sp. nov. (Figs 32–41) DESCRIPTION: This is a chain forming diatom. Most chains in our cultures are short; with mean length of 7 cells/chain (range 4–14 cells, Figs 32–35), the chains becoming somewhat arched, but not contorted, when longer. Many chains contain at least one end-cell furnished with somewhat thicker terminal setae widely diverging from the valve face, mostly at an angle of ~30–70 degrees, forming a wide bowl-like shape (Figs 32–34, 36). Common intercalary setae may be very long and, at least initially, diverge more or less perpendicularly to the chain long axis (Figs 32, 35), with no consistent orientation pattern. In valve face view, setae diverge from the apical axis as in Brunel Group II (Fig. 39). Windows are relatively wide (1/3 up to 1/2 of the frustule pervalvar axis), rectangular to hexagonal and centrally constricted (Figs 36, 37). Small vegetative cells have two, and larger cells up to four chloroplasts, but a greater number is present in post-sexual cells (Fig. 17). Up to 10 chloroplasts can be counted with confidence in auxospores and post-sexual cells, but more could have been present in the largest cells. Frustules are square to rectangular in girdle orientation. Valves are broadly elliptical to circular, 5.5–17.0 µm in apical, and 5.0–11.0 µm long in transapical axes (Figs 39, 41; note that some of the largest valves were found in an orientation precluding measurements of the transapical axis). The apical to transapical axes ratio (AA:TA) varies from 1.04–1.4 (mean 1.1 and 1.24, for post-sexual and smaller vegetative valves, respectively; Table 2). Valve face has one central (lower), and two apical (higher) elevations; most apparent in critical-point dried SEM preparations (Fig. 37). Apical elevations form the bases of the setae, the central elevation harbours an annulus. One rimoportula is located in an annulus on the terminal valves (Fig. 36). The typical valve face shows a pattern of branching costae radiating from a central annulus and extending to the mantle when not obscured by a marginal ridge. The central EUROPEAN JOURNAL OF PHYCOLOGY 7 Figs 18–23. Wall components in globular auxospores; scanning electron microscopy. Fig. 18. Small, globular auxospore covered with relatively large, circular incunabular scales (is; only prominent scales labelled). Organic matter covering them obscures their structure. Fig. 19. Slightly larger auxospore with surface organic matter removed to expose ornamentation of scales (is; only prominent scales labelled), including annuli (arrows) and radial striations (arrowheads). Fig. 20. Relatively well preserved, intact anisometric auxospore, illustrating asymmetry in lateral view. Fig. 21. Globular auxospore with incunabular scales of various shapes and sizes. Prominent large elliptical scale (les) and small circular scale (scs) are indicated. Fig. 22. Part of the auxospore wall surface illustrating ribbed, long incunabular scales; one indicated by arrowheads. Fig. 23. Surface of an auxospore showing bands (arrowhead) overlapped by incunabular scales (is). Scale bars: Figs 18–23, 5 µm. annulus of the intercalary valve may have a loose network of a few costae but no portulae, consistent with the sub-genus Hyalochaete. The boundary between valve face and mantle is marked by a ridge (Fig. 37). The mantle is short, fortified by a distal thicker rim. The girdle bands vary in width and are ornamented by delicate costae, 9–12 costae in 1 µm (Figs 37, 40). The setae originate from a wide basal part within the valve face, near the margin. Sibling valve setae fuse when crossing at the chain margin. Setae are round and hollow in cross section. There are three types of setae: one terminal and two intercalary. Terminal setae may be somewhat more strongly silicified than common intercalary setae. The two intercalary setae are the common setae that are lightly silicified (albeit with a gradient in degree of silicification), and less frequent setae that are heavily silicified (Figs 33–36, 38, 39). All setae are perforated with spiral rows of pores and spines. Some terminal and heavy intercalary setae may appear undulated (or contorted, sensu Rines & Hargraves (1988), in LM), but this is an optical illusion in the case of our species due to the regular distribution of spines. Observed in 8 I. KACZMARSKA ET AL. Figs 24–29. Perizonial bands in auxospores; scanning electron microscopy. Fig. 24. Partially damaged auxospore with some transverse perizonial bands (tpb) exposed. Note that these bands do not overlap, they disassociate along abutting edges. Fig. 25. Dorsal part of large auxospore covered with transverse perizonial bands, one or two of them (possibly two ends of the same; arrowheads) are twice as wide as the others (thin arrows). Note at least two layers of intersecting bands; an overlapping (thick arrows) and underlying (thin arrows) set of bands. Fig. 26. Rounded initial valve (delineated by 5 arrowheads) in a partially disarticulated auxospore showing radial ribs and pores between them; transverse perizonial bands (tpb) overlay the valve while incunabular scales (is) overlay the band in this location. Fig. 27. Nearly circular initial epivalve; note misaligned setae (arrowheads; seta on right is draped by two small scales), lack of rimoportula and striation much finer than that on a typical vegetative valve. A large incunabular scale (is) is visible below. Fig. 28. Incunabular scales (is; one delineated by three arrowheads) showing their fine ribbing. Fig. 29. Five transverse perizonial bands (pb 1–5) showing their fine structure, lack of overlap and multilayered organization. Bands 4 and 5 lay over bands 1–3 most clearly at the arrowhead. Scale bars: Figs 24–29, 5 µm. SEM, most of these setae are nearly straight or gently waved at best (Figs 36, 38, 39). Heavy intercalary setae are directed toward the chain end. HOLOTYPE: DMF SEM stub 299-10 deposited as Botanischer Garten und Botanisches Museum (BGBM), Berlin, preparation B 40 0043102 and illustrated in Fig. 36; in culture as clone Chaeto2; fixed, non-cleaned material from that cultured clone as BGBM preparation B 40 0043103. Sequence fragments of 5.8S rRNA, ITS2 and rbcL genes are deposited in GenBank and the BOLD System. GenBank accession numbers are MH114070–71 (5.8S rRNA+ITS2), MH114063–64 (rbcL); BOLD accession numbers are CHAC0006-18 and CHAC0007-18. TYPE LOCALITY: Petit-Cap, Northumberland Strait, New Brunswick, Canada (46.1996°N, 64.1647°W). COMPARISON TO OTHER SPECIES: A number of Chaetoceros species are known to possess two kinds of intercalary setae, but only the members of the Section Compressa have common setae that are relatively straight, smooth and thin while less common heavily silicified setae are undulating, unbranched and/or spinose. Moreover, species in this Section harbour two or more chloroplasts. EUROPEAN JOURNAL OF PHYCOLOGY 9 Fig. 30. 5.8S+ITS2 phylogenetic tree inferred from Bayesian inferences (BI). Support values at nodes from left to right are posterior probabilities (BI) and bootstrap percentages (ML). Scale bar indicates substitutions per site. These characteristics distinguish members of the Section Compressa from others (Cleve-Euler, 1951; Hustedt, 1962; Rines & Hargraves, 1988). C. acadianus demonstrates these characters and so belongs to Section Compressa morphologically. Currently known taxa that are well documented morphologically and molecularly comprising this Section are C. compressus, C. contortus with one variety, and C. hirtisetus. A summary of morphological comparisons for these taxa is given in Table 2 and Figs 42–44; genetic identities in Figs 30, 31 and Table 1. In addition, several other taxa previously synonymized with C. compressus (Hustedt, 1962) I. KACZMARSKA ET AL. 10 Table 1. Compensatory Base Changes (CBCs) and hemiCompensatory Base Changes (hCBCs) matrix based on the conserved regions of the ITS2 secondary structures (green in Fig. 31) among the five Chaetoceros species. 1. 2. 3. 4. 5. C. C. C. C. C. hirtisetus contortus contortus var. ornatus acadianus compressus 1. 0/0 4/5 3/7 4/4 3/3 2. 3. 4. 5. 0/0 1/9 2/2 1/2 0/0 3/7 1/7 0/0 1/2 0/0 Bold numbers indicate CBCs in the conserved regions, non-bold numbers are hCBCs. together with C. contortus, differ from C. acadianus in window shape (long rectangle in C. kelleri Brun) and the form of heavy intercalary setae (short, very strongly silicified in C. subcompressus Schröder and C. ciliatus Lauder). Lastly, Chaetoceros medium Schütt has thus far been only reported from the Baltic (Schütt, 1895) but was only described generically and illustrated with no heavy setae, terminal setae or valve outline. The single metric datum provided in the species diagnosis for C. medium was the length of the apical axis (13 µm). This would place C. medium equally well in the size range of all members of the Section (Table 2; Rines & Hargraves, 1990; Rines, 1999; Chamnansinp et al., 2015). The overlap in size between C. contortus and C. compressus was not known to Schütt in 1895. With no distinct characters indicated, it is impossible to determine the taxonomic affinity of C. medium in the context of presently known species diversity in this Section. Several taxa of this Section have been recently reappraised and amended using modern taxonomic tools (Chamnansinp et al., 2015). In this context, all of these species possess frustules that differ morphologically from our new species by having strongly undulated and/or heavier silicified intercalary setae. The ellipsoidal-to-rounded vegetative and post-sexual valve outline of our species is more similar to C. contortus-like than to C. compressus-like species; metric comparison of these taxa to C. acadianus is shown in Figs 42–44. The morphology of terminal setae and the way they diverge differs between C. compressus and our species and may be the best character to use in differentiating between them. In C. compressus, terminal setae are very thin, initially diverging perpendicularly to the chain axis, but very soon bending so they run parallel to the chain axis, similar to their heavy intercalary setae. In contrast, in C. acadianus and C. contortus-like taxa, terminal setae diverge in the form of a cup. Compared to these, our species terminal setae diverge wider still, forming a wide bowl or wide arch outline. Genetically, our species appears close to C. compressus (Table 1, Fig. 31) and C. contortus var. ornatus (Fig. 30). There is only one CBC in the conserved regions of the ITS2 structure (Helix 3, Fig. 31) between our species and C. compressus, but two to four between C. acadianus and C. contortus-like taxa examined here (Table 1). C. acadianus joins Fig. 31. Diagram of the secondary structure of the ITS2 transcript of Chaetoceros acadianus. Regions in the structure backbone marked in green indicate the regions 100% conserved except the CBCs and hCBCs in all taxa examined. The base pairs marked in red rectangular boxes indicate compensatory base changes (CBCs) in our Chaetoceros species compared to C. hirtisetus, C. contortus, C. contortus var. ornatus and C. compressus. The orientation of the base pairs in red rectangular boxes is [(5´ base)-(3´ base)]. Asterisk indicates non-canonical base pair. EUROPEAN JOURNAL OF PHYCOLOGY 11 Discussion Figs 32–35. Chain and setae types; light microscopy. Fig. 32. Typical length chain with the bowed, somewhat thicker end setae. Fig. 33. Shorter chain of cells showing a pair of heavy intercalary setae. Fig. 34. Short chain showing the end and typical, thin intercalary setae. Fig. 35. Short chain showing variation in orientation of heavy intercalary setae. Scale bars: Figs 32–35, 25 µm. C. contortus var. ornatus as a sister branch in the ITS region (5.8S+partial ITS2) phylogeny. Altogether, the morphological differences between our new species and known members of Section Compressa (albeit subtle in the case of C. contortus-like taxa), the CBC matrix and phylogenetic analyses suggest that our clones represent a species previously unknown to science. Unfortunately, our clones did not sporulate during this study, and so this part of comparative analysis and species life history remains unknown at this time. It is also not possible to compare species environments, because historical twists-and-turns of the taxonomic status of species from the Section Compressa preclude much confidence in the reports when only names are provided. For example, it is likely that the European specimens which Hustedt (1962) reported and illustrated as C. compressus are in fact C. contortus (fig. 388) which he synonymized with the former. If this particular source is used to identify either of the species, it will likely continue to generate misinformation about species ecology and distribution. Characteristic lateral auxospores of Chaetocerotales are one of the most frequently illustrated auxospores from natural waters (Meunier, 1910; Cupp, 1943; Drebes, 1974; Assmy et al., 2008; Hoppenrath et al., 2009; Li et al., 2017), with about 20 different species thus far reported. In some (Bacteriastrum; Drebes, 1972), special openings have evolved in the frustule to accommodate zygote exit from the female gametangium without compromising the integrity of the clonal chain. Similarly, laboratory-based studies have shown lateral auxospores and that Chaetoceros species can reproduce homothallically and oogamously (e.g. Hargraves, 1972; von Stosch et al., 1973; French & Hargraves, 1985; Furnas, 1985). However, the preponderance of homothallic sexuality in natural plankton remains unknown. In contrast, the auxospores of our species are terminal. Two close relatives of C. acadianus (C. compressus and C. contortus) have been shown to produce typical lateral auxospores in natural waters (Cupp, 1943 and Meunier, 1910, respectively), and to engage in homothallic allogamy in culture (Rines, 1999). Nonetheless, we never observed spermatogenesis in our cultures. Instead, a single auxospore mother cell is produced by the sexualized parent in whom the first meiotic division occurs before the auxospore mother cell frustule dehisces. Following this, auxospores were most commonly free and multinucleated (mostly binucleated, but some trinucleated were also present). We interpret trinucleated auxospores as the result of the second meiotic division taking place in only one of the two nuclei that are products of meiosis I. This would indicate that one of these nuclei becomes non-functional, even though it did not always appear pyknotic. Several centric diatoms have been reported to be autogamous, although the nuclear behaviour during meiosis was not always fully illustrated (von Stosch, 1967, 1982; Roshchin, 1975; Drebes, 1977), except for two thalassiosiroids: Cyclotella meneghiniana Kützing (Iyengar & Subrahmanyan, 1944) and Thalassiosira angulata (Mills & Kaczmarska, 2006). The behaviour of the nucleus during meiotic division in the auxospore mother cells of the two thalassiosiroids differed substantially. In C. meneghiniana, the meiotic nucleus went through the full course of divisions and produced four apparently viable haploid nuclei of equal size. Two of these nuclei fused and one diploid auxospore eventually developed. In T. angulata, the first meiotic division was followed by division of only one of the nuclei (the other pyknotized) leading to a pair of normal, haploid sister and one pyknotic nuclei following meiosis II. The sister nuclei eventually fused producing one diploid auxospore. Trinucleated auxospores in C. acadianus suggest the second type of autogamy, similar to T. angulata. 12 I. KACZMARSKA ET AL. Figs 36–41. Chain and setae types; scanning electron microscopy. Fig. 36. Terminal cell in a chain with a rimoportula (arrowhead) and somewhat thicker end setae, a specimen from the holotype preparation No. B 40 0043102. Fig. 37. Intercalary frustules showing hexagonal, constricted windows and epicingulae; central sibling-cells, after mitotic division. Note lack of rimoportulae on the intercalary valves. Fig. 38. Sibling valves joined by fused heavy intercalary setae. Note that these setae are not undulated. Fig. 39. Two sibling valves showing Brunel Group II type of setal divergence. Fig. 40. Close up of copulae (cp) showing their ornamentation. Sperm-like object on surface of band (and visible in other figures) is most likely a Caulobacter-like bacterium commonly found in marine waters and diatom cultures. Fig. 41. Circular post-sexual valve with imperfect setae and misplaced rimoportula (shown as enlarged insert; arrowhead). Scale bars: Figs 36, 38, 39, 41, 40 µm; Figs 37, 40, 5 µm. Little is known about auxospore fine structure and development in members of the Order Chaetocerotales, despite indications of considerable diversity among the related genera; compare Attheya decora West (Drebes, 1977), Bacteriastrum hyalinum Lauder (Drebes, 1972) and Chaetoceros didymus Ehrenberg. Von Stosch (1982) and co-workers (von Stosch et al., 1973) detailed all key stages of auxospore ontogeny in C. didymus, including electron microscopy imagery. The mature auxospore of this species was diagrammatically summarized (von Stosch, 1982; fig. 2). It shows the spatial relationship between the primary wall (scaly from spherical stage) and secondary wall components (a series of dorsally located perizonial bands underlapping the scaly portion) and their contribution to the overall covering of a mature polar auxospore. The bands were arranged concentrically in flattened hoops when viewed dorsally, but in a fan-like pattern in lateral view. Fanning perizonial bands allow greater expansion in the dorsal area of the auxospore where the large initial valve is expected to be deposited. The fine structure of all perizonial bands was not determined, but the last longitudinal perizonial band was shown to carry long fimbriae (von Stosch, 1982). The overall pattern of auxospore development seen in C. acadianus indicates that in general it is similar to C. didymus. Auxospores start with a spherical stage covered with incunabular scales, but subsequently grow anisodiametrically with EUROPEAN JOURNAL OF PHYCOLOGY 13 Table 2. Summary of valve characters helpful in distinguishing examined members of the Chaetoceros Section Compressa. Valves Seta Terminal Speciessource C. hirtisetusb AA (µm) 8–28 AA:TA ratio 1.66 # Chloroplasts 4–12 C. compressusa,c,d 10–42 1.60 4–40 C. acadianuse 5.5–17 ve: 1.24 in: 1.10 2–10+ C. contortus var. ornatusd 9–15 1.42 7–9 C. contortusa,c 6–22 ~1.00 7–10 Rimoportula #/Location 2–7 (3 usual) terminal valve 1 terminal valve 1 terminal valve 1 terminal valve 1–3 terminal valve Windows elliptical; constricted Form thin; nonundulated Divergence widely bowed to parallel elliptical; constricted thin; nonundulated thin to heavier; nonundulated ~parallel thin; nonundulated thin; nonundulated widely bowed hexagonal; constricted hexagonal; constricted wide hexagonal; constricted widely bowed narrowly bowed to ~V-shaped Intercalary Common Heavy thin; long, heavy; capillate undulated thin; long, heavy; nonstrongly capillate undulated thin; long, nonheavier; capillate some slightly wavy thin; long, heavy; nonstrongly capillate undulated thin; long, heavy; nonstrongly capillate undulated AA, apical axis; TA, transapical axis; ve, vegetative cell; in, initial cell. aRines & Hargraves, 1988; bRines & Hargraves, 1990; cRines, 1999; d Chamnansinp et al., 2015; ethis study deposition of the perizonium. However, the position and the structure of perizonial bands in C. acadianus differ from that of C. didymus. In general, Chaetoceros species valves and other individual, single layer components are best imaged using transmission electron microscopy (TEM). The TEM approach was inappropriate for our investigation due to the multilayered nature of the walls in C. acadianus, which would superimpose on each other in transmitted electron images. It is also difficult to disassociate delicate siliceous wall elements from the closely associated (likely embedded) amorphous matrix obscuring them gently enough to preserve the natural spatial relationships of all. Boiling in strong acids weakens and damages components of the auxospore cell wall, amorphous and siliceous, leading to flaccid walls with the wall components completely disarticulated or superposed on each other, already noted by earlier investigators. Consequently, some of the spatial relationships of the structures of their auxospore walls remain undetermined. Nonetheless, we can confirm that there are incunabular scales in C. acadianus auxospore walls, but they are larger and less obviously concentrated in one area of the cell compared to C. didymus (von Stosch, 1982). Furthermore, we also found transverse perizonial bands, but they are not organized in a fan-like manner about the primary band. Rather, the bands are more parallel in C. acadianus and the groups of them intersect with an underlying layer of similar bands. We cannot confirm the existence of a longitudinal perizonial band, or which (if any) of the transverse bands are open or closed. Assmy et al. (2008) used SEM to examine Chaetoceros auxospores but the images shown by these authors hint at a perizonial composition possibly more similar to B. hyalinum; that is a perizonium made of parallel rings of increasing diameter stacked up into an inverted pyramid (Drebes, 1972). In B. hyalinum, the entire perizonium remains enclosed within the stretched primary wall of the auxospore (presumably containing incunabular scales) until the initial frustule is ready to exit the auxospore (Drebes, 1972). The fine structure of the auxospore wall in A. decora has the transverse perizonium constructed in a manner more similar to that of C. didymus (von Stosch, 1982) than to Bacteriastrum (Drebes, 1972), but it is not lateral. In summary, the autogamous, non-lateral auxospore of C. acadianus, with a more complex, multilayered perizonium differs from those currently known in the order Chaetocerotales. Morphologically, the genus Chaetoceros has been divided into two sub-genera Phaeoceros and Hyalochaete (Cupp, 1943; Rines & Hargraves, 1988). In our 5.8S+ITS2 tree topology, the species belonging to Phaeoceros forms a monophyletic basal clade, whereas a larger cluster of selected taxa represents all the Hyalochaete, including Section Compressa. The Section, including C. acadianus, forms the basal sub-clade in Hyalochaete. In an LSU rDNA tree topology, the sub-genus Hyalochaete is recovered as a paraphyletic group comprising all of the ingroup taxa except for the taxa belonging to the Section Compressa (Li et al., 2015). These phylogenetic analyses suggest that the Section Compressa is monophyletic, but more work is needed before the natural 14 I. KACZMARSKA ET AL. Acknowledgements We thank M.L. MacGillivary for permission to use his unpublished sequences, Drs Lundholm and Chamnansinp for sharing their sequences and raw data included in Figs 42–44, and G. Hannach from King County Environmental Laboratory, Seattle, WA, USA for samples of local plankton. We acknowledge two anonymous reviewers for their helpful comments. Disclosure statement No potential conflict of interest was reported by the author(s). Funding This work was supported by a Natural Sciences and Engineering Research Council of Canada (NSERC) Discovery Grant to IK. Supplementary information The following supplementary material is accessible via the Supplementary Content tab on the article’s online page at http://doi.org/10.1080/09670262.2018.1542031 Supplementary fig. 1. Bayesian inference (BI) rbcL phylogenetic tree. Support values at nodes from left to right are posterior probabilities (BI) and bootstrap percentages (ML). Scale bar indicates substitutions per site. Supplementary table 1. List of Chaetoceros strains used in this study and their accession numbers. Figs 42–44. Comparative summary of the valve metric characters for closely related Chaetoceros species. Fig. 42. Apical axis. Fig. 43. Transapical axis. Fig. 44. Ratio of apical to transapical axes. Upper/lower boundary of box = 75th/25th percentile; upper/lower error bar = 90th/10th percentile; upper/lower dot = 95th/5th percentile; line within box = median. Metrics for C. acadianus includes postsexual cells. Raw data for C. compressus, C. hirtisetus and C. contortus var. ornatus are the same as in Chamnansinp et al. (2015) with permission by the authors. relationships between the members of Hyalochaete can be established. In summary, we document for the first time nonlateral formation of auxospores in a member of the genus Chaetoceros. In our species, these auxospores are produced autogamously – a process that is infrequently reported among diatoms. Based on molecular, reproductive and to a lesser extent morphological characters of the heavy intercalary setae and the valve, we propose that the subject of our investigation, Chaetoceros acadianus, is a species new to science. We also demonstrate the successful application of the CBC species concept in recognizing diatom species in the Section Compressa using comparative phylogenetic analysis of their ITS2 secondary structure models. Author contributions I. Kaczmarska: original concept, light and electron microscopy, culture work, drafting and editing manuscript; B. Samanta: molecular analysis, culture work, light microscopy, drafting and editing manuscript; J. Ehrman: light and electron microscopy, EDS, statistical analysis, drafting and editing manuscript; E. Porcher: cell isolation and culture work, light microscopy, DNA extraction and PCR, editing manuscript. References Altschul, S.F., Gish, W., Miller, M., Myers, E.W. & Lipman, D.J. (1990). Basic local alignment search tool. Journal of Molecular Biology, 215: 403–410. Andersen, R.A. (2005). Algal Culturing Techniques. Elsevier Academic Press, San Diego. Assmy, P., Hernández-Becerril, D.U. & Montresor, M. (2008). Morphological variability and life cycle traits of the type species of the diatom genus Chaetoceros, C. dichaeta. Journal of Phycology, 44: 152–163. Balzano, S., Percopo, I., Siano, R., Gourvil, P., Chanoine, M., Marie, D., Vaulot, D. & Sarno, D. (2017). Morphological and genetic diversity of Beaufort Sea diatoms with high contributions from the Chaetoceros neogracilis species complex. Journal of Phycology, 53: 161–187. Benson, D.A., Cavanaugh, M., Clark, K., Karsch-Mizrachi, I., Lipman, D.J., Ostell, J. & Sayers, E.W. (2013). GenBank. Nucleic Acids Research, 41: D36–D42. EUROPEAN JOURNAL OF PHYCOLOGY Brunel, J. (1972). Orientation of setae in the genus Chaetoceros, in regard to the apical axis. Journal of the Marine Biological Association of India, 14: 315–327. Chamnansinp, A., Moestrup, Ø. & Lundholm, N. (2015). Diversity in the marine diatom Chaetoceros (Bacillariophyceae) in Thai waters – revisiting Chaetoceros compressus and Chaetoceros contortus. Phycologia, 54: 161–175. Cleve-Euler, A. (1951). Die Diatomeen von Schweden und Finnland. Kungliga Svenska Vetenskapsakademiens Handlingar, Fjärde Serien, 2: 1–163. Cupp, E.E. (1943). Marine Plankton Diatoms of the West Coast of North America. University of California Press, Berkeley. Darriba, D., Taboada, G.L., Doallo, R. & Posada, D. (2012). jModeltest 2: more models, new heuristics and parallel computing. Nature Methods, 9: 772. Davidovich, N.A., Davidovich, O.I., Podunay, Y.A., Gastineau, R., Kaczmarska, I., Poulíčková, A. & Witkowski, A. (2017). Ardissonea crystallina has a type of sexual reproduction that is unusual for centric diatoms. Scientific Reports, 7: 14670. Drebes, G. (1972). The life history of the centric diatom Bacteriastrum hyalinum Lauder. Nova Hedwigia, 39: 95–110. Drebes, G. (1974). Marines Phytoplankton. Eine Auswahl der Helgoländer planktonalgen (Diatomeen, Peridineen). Thieme, Stuttgart. Drebes, G. (1977). Cell structure, cell division, and sexual reproduction of Attheya decora West (Bacillariophyceae, Biddulphiineae). Nova Hedwigia, 54: 167–178. Eddy, S.R. (1998). Profile hidden Markov models. Bioinformatics, 14: 755–763. Edgar, R.C. (2004). MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Research, 32: 1792–1797. Felsenstein, J. (1985). Confidence limits on phylogenies: an approach using the bootstrap. Evolution, 39: 1993–2002. Felsenstein, J. (2004). Inferring Phylogenies. Sinauer Associates, Sunderland. French, F.W. III. & Hargraves, P.E. (1985). Spore formation in the life cycle of diatoms Chaetoceros diadema and Leptocylindrus danicus. Journal of Phycology, 21: 477–483. Furnas, M.J. (1985). Diel synchronization of sperm formation in the diatom Chaetoceros curvisetum Cleve. Journal of Phycology, 21: 667–671. Guindon, S., Dufayard, J.F., Lefort, V., Anisimova, M., Hordijk, W. & Gascuel, O. (2010). New algorithms and methods to estimate maximum-likelihood phylogenies: assessing the performance of PhyML 3.0. Systematic Biology, 59: 307–321. Hall, T.A. (1999). BioEdit: a user-friendly biology sequence alignment editor and analysis program for Windows95/ 98/NT. Nucleic Acids Symposium Series, 41: 95–98. Hargraves, P.E. (1972). Studies on marine plankton diatoms. I. Chaetoceros diadema (Ehr.) Gran: life cycle, structural morphology, and regional distribution. Phycologia, 11: 247–257. Hernández-Becerril, D.U. & Flores Granados, C. (1998). Species of the diatom genus Chaetoceros (Bacillariophyceae) in the plankton from the southern Gulf of Mexico. Botanica Marina, 41: 505–520. Hoppenrath, M., Elbrächter, M. & Drebes, G. (2009). Marine Phytoplankton: Selected Microphytoplankton Species from the North Sea Around Helgoland and Sylt. Schweizerbart Science Publishers, Stuttgart. 15 Hustedt, F. (1962). Die Kieselalgen Deutschlands, Österreichs und der Schweiz unter Berücksichtigung der übrigen Länder Europas sowie der angrenzenden Meeresgebiete. Dr. L. Rabenhorst‘s Kryptogamen-Flora von Deutschlands, Österreichs und der Schweiz, 7: 217–920. Idei, M., Sato, S., Nagasato, C., Motomura, T., Toyoda, K., Nagumo, T. & Mann, D.G. (2015). Spermatogenesis and auxospore structure in the multipolar centric diatom Hydrosera. Journal of Phycology, 51: 144–158. Iyengar, M.O. & Subrahmanyan, R. (1944). Reduction division and auxospore formation in Cyclotella meneghiniana Kütz. Journal of the Indian Botanical Society, 23: 125–149. Kaczmarska, I., Martin, J.L., LeGresley, M.M. & Ehrman, J.M. (2005). Diversity of the diatom genus Pseudo-nitzschia Peragallo in the Quoddy Region of the Bay of Fundy, Canada. Harmful Algae, 4: 1–19. Kaczmarska, I., Poulíčková, A., Sato, S., Edlund, M.B., Idei, M., Watanabe, T. & Mann, D.G. (2013). Proposals for a terminology for diatom sexual reproduction, auxospores and resting stages. Diatom Research, 28: 263–294. Keller, A., Schleicher, T., Schultz, J., Müller, T., Dandekar, T. & Wolf, M. (2009). 5.8S-28S rRNA interaction and HMM-based ITS2 annotation. Gene, 430: 50–57. Kooistra, W.H.C.F., Sarno, D., Hernández-Becerril, D.U., Assmy, P., Prisco, C.D. & Montresor, M. (2010). Comparative molecular and morphological phylogenetic analyses of the taxa in the Chaetocerotaceae (Bacillariophyta). Phycologia, 49: 471–500. LeBlanc, K. & Hutchins, D.A. (2005). New applications of a biogenic silica deposition fluorophore in the study of oceanic diatoms. Limnology and Oceanography: Methods, 3: 462–467. Lefort, V., Longueville, J.E. & Gascuel, O. (2017). SMS: Smart Model Selection in PhyML. Molecular Biology and Evolution, 34: 2422–2424. Li, Y., Zhu, S., Lundholm, N. & Lü, S. (2015). Morphology and molecular phylogeny of Chaetoceros dayaensis sp. nov. (Bacillariophyceae), characterized by two 90° rotations of the resting spore during maturation. Journal of Phycology, 51: 469–479. Li, Y., Boonprakob, A., Gaonkar, C.C., Kooistra, W.H.C.F., Lange, C.B., Hernández-Becerril, D., Chen, Z., Moestrup, Ø. & Lundholm, N. (2017). Diversity in the globally distributed diatom genus Chaetoceros (Bacillariophyceae): three new species from warm-temperature waters. PLoS ONE, 12: e0168887. MacGillivary, M.L. & Kaczmarska, I. (2011). Survey of the efficacy of a short fragment of the rbcL gene as a supplemental DNA barcode for diatoms. Journal of Eukaryotic Microbiology, 58: 529–536. Mai, J.C. & Coleman, A.W. (1997). The internal transcribed spacer 2 exhibits a common secondary structure in green algae and flowering plants. Journal of Molecular Evolution, 44: 258–271. Meunier, A. (1910). Microplancton des Mers de Barents et de Kara. Duc d’Orléans Campagne Arctique de 1907. Charles Bulens, Brussels. Mills, K.E. & Kaczmarska, I. (2006). Autogamic reproductive behavior and sex cell structure in Thalassiosira angulata (Bacillariophyta). Botanica Marina, 49: 417–430. Moniz, M.B.J. & Kaczmarska, I. (2010). Barcoding of diatoms: nuclear encoded ITS revisited. Protist, 161: 7–34. 16 I. KACZMARSKA ET AL. Posada, D. & Buckley, T.R. (2004). Model selection and model averaging in phylogenetics: advantages of Akaike information criterion and Bayesian approaches over likelihood ratio tests. Systematic Biology, 53: 793–808. Rines, J.E.B. (1999). Morphology and taxonomy of Chaetoceros contortus, with preliminary observations on Chaetoceros compressus (subgenus Hyalochaete, section Compressa). Botanica Marina, 42: 539–551. Rines, J.E.B. & Hargraves, P.E. (1988). The Chaetoceros Ehrenberg (Bacillariophyceae) flora of Narragansett Bay, Rhode Island, U.S.A. Bibliotheca Phycologica, 79: 1–196. Rines, J.E.B. & Hargraves, P.E. (1990). Morphology and taxonomy of Chaetoceros compressus Lauder var. hirtisetus var. nova, with preliminary consideration of closely related taxa. Diatom Research, 5: 113–127. Rines, J.E.B. & Theriot, E.C. (2003). Systematics of Chaetocerotaceae (Bacillariophyceae). I. A phylogenetic analysis of the family. Phycological Research, 51: 83–98. Ronquist, F., Teslenko, M., Mark, P.V., Ayres, D.L., Darling, A., Höhna, S., Larget, B., Liu, L., Suchard, M.A. & Huelsenbeck, J.P. (2012). MrBayes 3.2: efficient Bayesian inference and model choice across a large model space. Systematic Biology, 61: 539–542. Roshchin, A.M. (1975). Osobiennosti ontogeneza morskich centritcheskich diatomowych wodoroslei w klonowych culturach. Nautchne Doklady Wyschei Shkoly, Biologitcheskie Nauki, 3: 47–51. Round, F.E., Crawford, R.M. & Mann, D.G. (1990). The Diatoms. Biology & Morphology of the Genera. Cambridge University Press, Cambridge. Samanta, B., Kinney, M.E., Heffell, Q., Ehrman, J.M. & Kaczmarska, I. (2017). Gametogenesis and auxospore development in the bipolar centric diatom Brockmanniella brockmannii (Family Cymatosiraceae). Protist, 168: 527–545. Samanta, B., Heffell, Q., Ehrman, J.M. & Kaczmarska, I. (2018). Spermatogenesis in the bipolar centric diatom Plagiogrammopsis vanheurckii (Mediophyceae). Phycologia, 57: 354–359. Schütt, F. (1895). Arten von Chaetoceros und Peragallia. Ein Beitrag zur Hochseeflora. Berichte der Deutschen Botanischen Gesellschaft, 13: 35–48. Seibel, P.N., Müller, T., Dandekar, T., Schultz, J. & Wolf, M. (2006). 4SALE – a tool for synchronous RNA sequence and secondary structure alignment and editing. BMC Bioinformatics, 7: 498. Seibel, P.N., Müller, T., Dandekar, T. & Wolf, M. (2008). Synchronous visual analysis and editing of RNA sequence and secondary structure alignments using 4SALE. BMC Research Notes, 1: 91. von Stosch, H.A. (1967). Diatomeen. In Handbuch der Pflanzenphysiologie, vol. 18: Vegetative Fortpfllanzung, parthenogenese und Apogamie bei Algen (Ettl, H., Müller, D.G., Neumann, K., von Stosch, H.A. & Weber, W., editors), 657–681. Springer-Verlag, Berlin. von Stosch, H.A. (1982). On auxospore envelopes in diatoms. Bacillaria, 5: 127–156. von Stosch, H.A., Theil, G. & Kowallik, K. (1973). Entwicklungsgeschichtliche Untersuchungen an zentrischen Diatomeen V. Bau und Lebenszyklus von Chaetoceros didymum, mit Beobachtungen uber einige anderen Arten der Gattung. Helgoländer wissenschaflicher Meeresuntersuchungen, 25: 384–445. White, T.J., Bruns, T.D., Lee, S.B. & Taylor, J.W. (1990). Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In PCR Protocols: A Guide to Methods and Applications (Innis, M.A., Gelfand, D.H., Sninsky, J.J. & White, T.J., editors), 315–322. Academic Press, London. Wolf, M., Friedrich, J., Dandekar, T. & Müller, T. (2005). CBCAnalyzer: inferring phylogenies based on compensatory base changes in RNA secondary structures. In Silico Biology, 5: 291–294. Zuker, M. (2003). Mfold web server for nucleic acid folding and hybridization prediction. Nucleic Acids Research, 31: 3406–3415.