European Journal of Phycology
ISSN: 0967-0262 (Print) 1469-4433 (Online) Journal homepage: http://www.tandfonline.com/loi/tejp20
Thalassiosira species (Bacillariophyceae, Thalassiosirales) in the North Sea at Helgoland (German Bight) and Sylt (North Frisian Wadden Sea) – a first approach to assessing diversity Mona Hoppenrath , Bank Beszteri , Gerhard Drebes , Hannelore Halliger , Justus E. E. Van Beusekom , Silvia Janisch & Karen H. Wiltshire To cite this article: Mona Hoppenrath , Bank Beszteri , Gerhard Drebes , Hannelore Halliger , Justus E. E. Van Beusekom , Silvia Janisch & Karen H. Wiltshire (2007) Thalassiosira species (Bacillariophyceae, Thalassiosirales) in the North Sea at Helgoland (German Bight) and Sylt (North Frisian Wadden Sea) – a first approach to assessing diversity, European Journal of Phycology, 42:3, 271-288, DOI: 10.1080/09670260701352288 To link to this article: http://dx.doi.org/10.1080/09670260701352288
Published online: 15 Aug 2007.
Submit your article to this journal
Article views: 979
View related articles
Citing articles: 14 View citing articles
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=tejp20 Download by: [37.44.207.74]
Date: 29 January 2017, At: 04:21
Eur. J. Phycol., (2007), 42(3): 271–288
Thalassiosira species (Bacillariophyceae, Thalassiosirales) in the North Sea at Helgoland (German Bight) and Sylt (North Frisian Wadden Sea) – a first approach to assessing diversity
MONA HOPPENRATH1, BANK BESZTERI2, GERHARD DREBES3, HANNELORE HALLIGER3, JUSTUS E. E. VAN BEUSEKOM3, SILVIA JANISCH1 AND KAREN H. WILTSHIRE1 1
Biologische Anstalt Helgoland, Alfred Wegener Institute for Polar and Marine Research, Kurpromenade, D-27498 Helgoland, Germany 2 Alfred Wegener Institute for Polar and Marine Research, Am Handelshafen 12, D-27570 Bremerhaven, Germany 3 Wadden Sea Station Sylt, Alfred Wegener Institute for Polar and Marine Research, Hafenstraße 43, D-25992 List/Sylt, Germany (Received 8 June 2006; accepted 8 March 2007) Thalassiosira species are an important component of phytoplankton populations in pelagic waters around the islands of Helgoland and Sylt in the German North Sea. This taxonomic study revealed the presence of at least 27 species, 21 at Helgoland and 21 at Sylt. Fifteen species had not been recorded previously for Helgoland (T. aestivalis, T. angulata, T. concaviuscula, T. curviseriata, T. delicatula, T. diporocyclus, T. hendeyi, T. kuschirensis, T. minima, T. minuscula, T. oceanica, T. pacifica, T. proschkinae, T. tealata, T. tenera) and 16 for Sylt (T. aestivalis, T. angulata, T. concaviuscula, T. constricta, T. curviseriata, T. delicatula, T. guillardii, T. hendeyi, T. lundiana, T. mediterranea, T. minima, T. pacifica, T. partheneia, T. punctigera, T. tealata, T. tenera). We provide brief descriptions with illustrations and document the seasonal occurrence of these species at the two stations. We also discuss investigation methods used for different levels of species recognition, namely Utermo¨hl counts, light microscopical observations of living net-samples as well as scanning electron microscopy of netsamples and isolated cultures. Clonal cultures of some species were established to facilitate the taxonomic investigations and 18S ribosomal DNA sequencing was carried out to complement the morphological studies. A perspective for a routine specieslevel assessment strategy is given. Key words: biodiversity, diatoms, Helgoland Reede, microalgae, morphology, phytoplankton, seasonality, taxonomy
Introduction After Chaetoceros Ehrenberg, Thalassiosira Cleve is the diatom genus with the highest species diversity in the pelagic, marine, temperate phytoplankton community. It includes approximately 100 species (Round et al., 1990; Hasle & Syvertsen, 1996). Thalassiosira species contribute significantly to the diatom spring bloom (Hagmeier, 1971; Bauerfeind et al., 1990; Du¨rselen & Rick, 1999; Ehrenhauss et al., 2004) in the North Sea. Because of their ecological importance in temperate regions (Hasle & Smayda, 1960; Karentz & Smayda, 1984; Haigh et al., 1992; Lange et al., 1992), the assessment of species diversity is an essential task in order to understand the different ecological ‘roles’ played by each single species. Thalassiosira species frequently occur in temperate seas, and
their profile in the literature has been raised by recent publications on their potential toxicity to zooplankton (e.g. Ianora et al., 2004). The most important characters for species recognition are the shape, position and number of strutted (fultoportulae) and labiate processes (rimoportulae) on the valve face, the areolation pattern or other ornamentation, and the girdle band morphology (Hasle & Heimdal, 1970; Fryxell & Hasle, 1972, 1977; Hasle, 1972, 1973a, b, 1978a, b; Somers, 1972; Fryxell, 1975; Hasle & Fryxell, 1977). Useful features for species identification under the light microscope (LM) include cell shape and their occurrence as single cells, as chains or in mucilage colonies (Fryxell et al., 1984). However, it is important to keep in mind that Thalassiosira species can vary considerably during their life-cycle, not only in size (e.g. Fig. 53), but
Correspondence to: Mona Hoppenrath. e-mail:
[email protected] Present address: Department of Botany, University of British Columbia, 6270 University Blvd., Vancouver, BC V6T 1Z4, Canada. ISSN 0967-0262 print/ISSN 1469-4433 online/07/030271–288 ß 2007 British Phycological Society DOI: 10.1080/09670260701352288
M. Hoppenrath et al. also in valve geometry and micromorphology (e.g. initial cells). Often, only large vegetative valves can be identified unambiguously. Detailed descriptions of Thalassiosira species diversity for different geographic regions are given by Hallegraeff (1984), Herzig & Fryxell (1986), Mahood et al. (1986), Fukuyo et al. (1990), Harris et al. (1995), Muylaert & Sabbe (1996) and Sar et al. (2002). Grøntved (1952) listed four Thalassiosira species in his phytoplankton investigations of the southern North Sea, and Braarud et al. (1953) listed eight for the North Sea and adjacent waters. A detailed account of Thalassiosira species from the Elbe estuary is also available (Muylaert & Sabbe, 1996). Molecular methods, especially the use of ribosomal DNA (rDNA), are increasingly being used for diversity assessment in marine eukaryotic plankton (Dı´ ez et al., 2001; Moon-van der Stay et al., 2001). The success of such investigations for species inventory, i.e. to link the obtained DNA sequence information to classical taxonomic information, critically depends on the availability of rDNA sequence information from wellcharacterized cultures. rDNA sequences provide a useful basis for monitoring the presence of taxa using oligonucleotide probes (Simon et al., 2000; Biegala et al., 2003; John et al., 2003). In this study the species diversity of Thalassiosira at the North Sea islands of Helgoland and Sylt was investigated. 18S rDNA, the molecule most commonly used in molecular diversity studies, was sequenced from some of the cultures obtained for our study. We compared these sequences with those already available for other Thalassiosira species using phylogenetic analyses. The results were compared with earlier findings and a future species-level monitoring strategy was recommended.
272 Microscopic observations and species identification Net samples from surface water of the Helgoland Reede Station (54 11.30’N; 7 54.00’E; Fig. 1C) were collected twice a week in 2001 and 2002 and once a week in 2003–2004. Net samples from surface water in the North Friesian Wadden Sea at List/Sylt (55 01.30’N; 08 27.10’E; Fig. 1B) have been collected once a week since 1987 (ongoing). Nets of different mesh size, 20 mm and 80 mm, were used. The samples were brought to the laboratory and living organisms in small Petri dishes were identified with an inverted and a normal LM equipped with seawater-immersion objectives. All species observed and identified were listed (for Helgoland see Hoppenrath, 2004) and seasonality documented for those identified under the LM. For species identifiable only under the scanning electron microscope (SEM) no seasonal occurrence is given, only their isolation/ observation dates are mentioned. The frequency of observation was calculated as the total number of observations of a certain species per month divided by the total number of samples investigated per month. The results from Sylt are based on a much longer dataset (since 1987) than for Helgoland (2001–2003). Thalassiosira cells were also isolated and cultured, and some mixed samples and cultures were fixed with Lugol’s solution for further investigation with SEM. Specimens were filter-mounted, rinsed with distilled water and subsequently dehydrated with 30% ethanol and with dimethoxypropane (Merck). The filter was air dried. Diatom frustules were also cleaned with hot acid, washed with distilled water and mounted on a cover slip. Preparations were sputter-coated with gold-palladium for three minutes at 45 mA (Bal-Tec SCD 050) and examined by SEM with a Zeiss DSM 940A. All identifications were based on published illustrations and descriptions (see the literature cited in the Introduction and Discussion). For difficult specimens,
Materials and methods Area description Helgoland is situated in the German Bight about 70 km from the main coast (Fig. 1A). Salinity ranges from 28–33 psu and temperature ranges from 2 to 22 C (Wiltshire & Manly, 2004). Planktonic primary production in this part of the German Bight is about 250 g C m2 year1 (Joint & Pomroy, 1993). Sylt is situated in the northern part of the Wadden Sea (Fig. 1a), a shallow tidally influenced coastal sea along the Dutch, German and Danish North Sea coast. Tides are semidiurnal with a mean range of 2 m. Mean depth is about 2–3 m. Average salinity ranges between 27.5 psu in February and 31 psu in August. Long-term average temperatures range from 2.7 C in February to 18.1 C in August. Planktonic primary production is about 160 g C m2 year1 (Asmus et al., 1998).
Fig. 1. Maps showing the sampling sites. A – Locations of Sylt and Helgoland in the German Bight, North Sea. The only other nearby study of Thalassiosira species was from the Elbe estuary. B – Sampling site (arrow) at Sylt. C – Sampling site (arrow) at Helgoland.
Thalassiosira species of the German North Sea expert advice was obtained (see Acknowledgements). Cell sizes were measured from living cells and/or from valves in the SEM. Size ranges given in brackets are from the literature, mainly Hasle & Syvertsen (1996) and Drebes (1974). The quantitative data provided here are from the Helgoland long-term series, one of the longest aquatic data sets in history (Hickel et al., 1992; Wiltshire & Du¨rselen, 2004). Phytoplankton species have been counted 5 days a week from 1962 to the present. Using a bucket, surface water samples are taken (usually before 9 am) on working days at the Kabel Tonne site (54 11.3’N; 7 54.0’E; Fig. 1C) between the two islands at Helgoland. This sample is mixed well and subsampled into a plastic bottle for future analyses of nutrients and phytoplankton. The phytoplankton samples are preserved in brown glass bottles using Lugol’s solution. The samples are counted daily under an inverted microscope, to species level when possible, or in size classes (as sp.) using the Utermo¨hl method (25 ml settled out). The general counting protocol is described in Lund et al. (1958), the species protocol is after Berg 1992 (see Wiltshire & Du¨rselen, 2004), and size classes are in accordance with Hillebrand et al. (1999).
Molecular genetic methods Liquid cultures were harvested by filtration and DNA was extracted using the Invisorb Spin Plant Mini Kit (Invitek, Germany). For the amplification of the 18S rDNA PCR protocols and primers described in Medlin et al. (1988) were used. PCR products were purified with the QIAQuick PCR Product Purification Kit (QIAGEN, Germany) and sent for direct sequencing to QIAGEN (Germany). Sequencing primers are listed below, they were based on Elwood et al. (1985). Some of them were used with slight modifications: 1F 528F (50 -AACCTGGTTGATCCTGCCAGT-30 ), (50 -GCGGTAATTCCAGCTCCAA-30 ), 1055F (50 -GG TGGTGCATGGCCGTTCTT-30 ), 536R (50 -AATTAC CGCGGCKGCTGGCA-30 ), 1055R (50 -ACGGCCATG CACCACCACCCAT-30 ) and 1528R (50 -TGATCCTTC TGCAGGTTCACCTAC-30 ). The sequences obtained for this study were aligned with other 18S rDNA sequences available in GenBank from Thalassiosiraceae (Table 1). The sequence designated T. antarctica (accession: AF374482) was excluded from our data set because it grouped outside the Thalassiosiraceae. Kaczmarska et al. (2006) note that this is probably an 18S rDNA sequence from a raphid diatom. Multiple alignments were prepared using ClustalX (Thompson et al., 1997). Phylogenetic trees were calculated using PAUP 4.0b10 (Swofford, 1998). The program Modeltest (Posada & Crandall, 1998) was used to determine the nucleotide substitution model best fitting the data. The best-fit model according to the hierarchical likelihood ratio tests was used to estimate phylogenetic trees. Neighbour joining trees were calculated using maximum likelihood distances with parameters of the model thus chosen. Maximum parsimony trees were calculated in heuristic searches
273 with random taxa addition. To get an indication of the confidence in nodes of the trees obtained, bootstrap analyses were performed in 1000 (for neighbour joining) and 500 (maximum parsimony) replicates.
Results Twenty-seven identified Thalassiosira species are described briefly below and their occurrence at the two sampling sites is given. Further images, especially LM pictures, of these species can be found in Hoppenrath et al. (2007). Species are ordered alphabetically, not indicating any systematic relationships. The analysis of Thalassiosira counts from the Helgoland Reede data set is presented before the molecular data (18S rDNA).
Table 1. Genbank accession numbers and species names of sequences used in the phylogenetic analyses Accession number AY485494 AY485473 AY485472 AY485469 AY485452 AY485445 X85398 AJ535171 AJ535172 AJ535170 AJ535169 AJ535168 AJ535166 AJ535165 AJ536450 AY496213 AY496212 AY496211 AY496210 AY496209 AY496208 AY496207 AY496206 AF525672 AF374482 AF374481 AF374480 AF374479 AF374478 AF374477 AF462060 AF462059 AF462058 X85397 X85396 X85393 X85394 X85395 X52006 M54988 AY485469 X85398 AY188181
Species name Skeletonema subsalsum Skeletonema costatum Minidiscus trioculatus Porosira pseudodelicatula Thalassiosira pseudonana Thalassiosira weissflogii Porosira pseudodenticulata Thalassiosira sp. Cyclotella meneghiniana Thalassiosira weissflogii Thalassiosira pseudonana Skeletonema menzelii Skeletonema subsalsum Skeletonema sp. Skeletonema menzelii Cyclotella meneghiniana Cyclotella meneghiniana Cyclotella meneghiniana Cyclotella meneghiniana Cyclotella cf. scaldensis Cyclotella cf. scaldensis Cyclotella meneghiniana Cyclotella meneghiniana Detonula confervacea Thalassiosira antarctica Thalassiosira pseudonana Thalassiosira rotula Thalassiosira oceanica Thalassiosira guillardii Thalassiosira weissflogii Skeletonema pseudocostatum Thalassiosira rotula Thalassiosira rotula Thalassiosira rotula Thalassiosira eccentrica Skeletonema pseudocostatum Skeletonema pseudocostatum Skeletonema costatum Skeletonema costatum Skeletonema costatum Porosira pseudodelicatula Porosira pseudodenticulata Ditylum brightwellii
M. Hoppenrath et al. T. aestivalis Gran et Angst (Figs 2, 3) Cells 18.8–20.8 (14.0–56.0) mm in diameter. Valve with fine areolae arranged in sectors; more than 20 areolae in 10 mm. One central strutted process adjacent to a large areola and one marginal ring of strutted processes – relatively short external tubes. Four marginal processes in 10 mm. One labiate process takes the place of a strutted process. Valve flat in girdle view. LIVE HABIT: Usually occurs as single cells, but also forms chains. Areolation not visible in LM. DISTRIBUTION: Neritic, in temperate to warm water. SEASONALITY: Isolated from Sylt and currently being monitored as a species complex; present with T. concaviuscula throughout the year, occurring mainly from October to April.
274 Cells 20.0–70.0 (14.0–78.0) mm in diameter. Valve with eccentric areolation; 25 to 30 areolae in 10 mm. Central strutted processes in clusters (one to three) in a subcentral ring. One marginal ring of conspicuous strutted processes whose tubes decrease in diameter towards their distal ends; three to six in 10 mm. One labiate process located between two marginal strutted processes, just inside the marginal ring of strutted processes. Cells rectangular in girdle view. Connecting threads arranged in three to six groups at some distance from the centre. LIVE HABIT: Occurs in chains; connecting threads visible. DISTRIBUTION: Worldwide occurrence. SEASONALITY: Occurring especially in spring. Isolated from Helgoland and Sylt; recorded in April and May at Helgoland and all year round at Sylt; highest frequency from February to April.
T. angulata (Gregory) Hasle (Figs 4, 5) BASIONYM: Orthosira angulata Gregory 1857 SYNONYM: T. decipiens (Grunow) Jørgensen non T. decipiens (Grunow) Jørgensen in Hasle 1979 Cells 19.1–38.2 (11.0–39.0) mm in diameter. Valve with hexagonal areolae in curved rows in sectors; 14 to 24 areolae in 10 mm. One central strutted process and one marginal ring of strutted processes with widely separated, long external tubes; three marginal processes in 10 mm. Marginal strutted processes a double tube, with the outer tube projecting as a flange from the tip. One large, tube-like labiate process located close to a marginal strutted process, just inside the marginal ring of strutted processes. One central areola slightly larger than the others. Cell quadrangular in girdle view with smoothly rounded margins. LIVE HABIT: Occurs as single cells, sometimes coated with fine detritus, but also forming chains, mainly during the spring bloom. Areolation visible in LM. DISTRIBUTION: Occurs in cold to temperate waters. SEASONALITY: Isolated from Helgoland and Sylt, and present throughout the year, but reaching maximum values between November and April (Fig. 55). Its bloom period is longer around Helgoland than Sylt.
T. concaviuscula Makarova (Figs 8–10) Cells 19.1–27.0 (14.0–56.0) mm in diameter. Valve with fine areolae arranged in sectors; more than 20 areolae in 10 mm. One central strutted process adjacent to a large areola and one marginal ring of strutted processes with relatively short external tubes. Four marginal processes in 10 mm. One labiate process lies between two marginal strutted processes. Valve flat in girdle view. LIVE HABIT: Cells usually occur singly, but can also form chains. Areolation not visible in LM. DISTRIBUTION: Neritic, occurring in cold to temperate waters. SEASONALITY: Isolated from Helgoland and Sylt. It has not been distinguished from T. aestivalis (see above); present all year round at Helgoland, mainly from October to April. T. constricta Gaarder (Figs 11, 12)
T. anguste-lineata (A. Schmidt) Fryxell et Hasle (Figs 6, 7)
Cells 18.8 mm (12.0–32.0) mm in diameter. Valve with fine siliceous ribs, areolation (40 to 60 areolae in 10 mm) only in the area around the marginal strutted processes. Central strutted processes in a cluster. One marginal ring of strutted processes with tubes only slightly raised above the surface; three to five processes in 10 mm. One small labiate process midway between two marginal strutted process. Cells rectangular in girdle view. Pervalvar axis often longer than the cell diameter.
BASIONYM: Coscinodiscus anguste-lineata A. Schmidt SYNONYM: Coscinodiscus polychordus Gran T. polychorda (Gran) Jo¨rgensen Coscinosira polychorda (Gran) Gran
LIVE HABIT: Occurs in chains with a visible connecting thread. Characteristic resting spores (Fig. 12). DISTRIBUTION: Occurs in cold to cold-temperate waters.
Thalassiosira species of the German North Sea
275
Figs 2–15. LM and SEM of Thalassiosira species. Figs 2, 3. T. aestivalis (SEM). Fig. 2. Valve outside view. Fig. 3. Valve inside view. Figs 4, 5. T. angulata (SEM). Fig. 4. Valve outside view. Fig. 5. Complete cell in oblique valve view also showing the girdle. Figs 6, 7. T. anguste-lineata (SEM). Fig. 6. Valve outside view. Fig. 7. Valve inside view. Figs 8–10. T. concaviuscula (SEM). Fig. 8. Complete cell in valve view, note the threads extruded from the strutted processes. Fig. 9. Valve outside view. Fig. 10. Valve inside view. Figs 11, 12. T. constricta (LM). Fig. 11. Chain of vegetative cells in girdle view. Fig. 12. Two recently formed, still connected resting spores in girdle view. Figs 13–15. T. curviseriata (SEM). Figs 13, 14. Valve outside view, note the different number of central strutted processes. Fig. 15. Valve inside view. Scale bars: 5 mm (Figs 2–10, 13–15) 10 mm (Figs 11, 12). Arrows indicate the position of the labiate processes and arrowheads the central strutted process, except in Figs 6, 7 where there is a subcentral ring of strutted processes.
M. Hoppenrath et al. SEASONALITY: Identified at Sylt and recorded in early spring (February to April). T. curviseriata Takano (Figs 13–15) Cells 9.5–9.9 (5.0–14.0) mm in diameter. Valve with radial rows of areolae (26 to 30 areolae in 10 mm) and siliceous granules all over the surface. One or two central strutted processes next to an annulus. One marginal ring of 4 to 7 conspicuous winged strutted processes, two wings per process diverging into two or three branches. One labiate process adjacent to a marginal strutted process, just inside the marginal ring of strutted processes. Cells discoid to octagonal in girdle view. LIVE HABIT: Occurs in chains. DISTRIBUTION: Worldwide occurrence, excluding polar regions. SEASONALITY: Isolated from Helgoland and Sylt; nearly all year round at Sylt, with highest numbers from November to April; and at Helgoland from January to May with highest numbers from February to April. Not distinguished from T. tealata during qualitative LM monitoring. T. decipiens (Grunow) Jo¨rgensen (Figs 16–18) BASIONYM: Coscinodiscus eccentricus var.? decipiens Grunow Cells 8.0–32.0 (40.0) mm in diameter. Cells heterovalvate with eccentric, relatively coarse hexagonal areolae (10 to 15 areolae in 10 mm). Valve face covered by minute siliceous granules. One tiny central strutted process on one valve only. One marginal ring of strutted processes (external tubes); four to six in 10 mm. One prominent labiate process between two marginal strutted processes. Cell almost lens-shaped to cylindrical in girdle view. LIVE HABIT: Occurs as single cells, conspicuously coated with detritus and sediment. Areolation visible in LM. DISTRIBUTION: Tychopelagic. Widely distributed, see Hasle (1979). SEASONALITY: Isolated from Helgoland and Sylt in winter and recorded all year round at Sylt, with highest numbers from November to April (Fig. 56). From October to May at Helgoland with a clear winter maximum. T. delicatula Ostenfeld (Figs 19–21) non T. delicatula Hustedt SYNONYM: T. coronata Gaarder Cells 19.4–27.1 (9.0–30.0) mm in diameter. Valve with fine areolae in radial rows, sometimes
276 arranged in sectors; 22 to 26 areolae in 10 mm. Valve face around the areolae covered by minute siliceous spinulae. One central strutted process adjacent to a larger areola. Three marginal rings of strutted processes (four to five in 10 mm); scattered strutted processes on valve face. One marginal ring of occluded processes, longer than the strutted processes. One labiate process between strutted processes of the inner marginal ring. Valve flat with central cavity in girdle view. Pervalvar axis as long as cell diameter. LIVE HABIT: Occurs as single cells, but also forming chains. Areolation not visible in LM. DISTRIBUTION: Occurs worldwide. SEASONALITY: Isolated from Helgoland and Sylt during spring, but not recorded during routine monitoring. Specimens resembling T. fallax as depicted by Drebes (1974, p. 29, Fig. 16C) isolated in March and April at Helgoland belong to T. delicatula. T. diporocyclus Hasle (Figs 22–24) Cells 13.4–14.8 (12.0–24.0) mm in diameter. Valve with fine fasciculate areolae; 24 to 31 areolae in 10 mm. Valve face around the areolae covered by minute siliceous spinulae. One central strutted process adjacent to an annulus. Two marginal rings of strutted processes (3–4 mm apart) without external tubes. One labiate process in the outer marginal ring. Valve flat to convex with curved mantle in girdle view. Pervalvar axis as long as cell diameter. LIVE HABIT: Occurs in mucilaginous colonies. DISTRIBUTION: Occurs in temperate to warm waters. SEASONALITY: Isolated from Helgoland in December 2002 as colony, but not recorded during routine monitoring. T. eccentrica (Ehrenberg) Cleve (Figs 25, 26) BASIONYM: Coscinodiscus eccentricus Ehrenberg Cells 25.0–108.0 (12.0–110.0) mm in diameter. Valve with strongly eccentric and relatively coarse hexagonal areolae; 17 to 19 areolae in 10 mm. One tiny central strutted process. Central areola surrounded by seven areolae. Two marginal rings of small strutted processes (two to five in 10 mm) and additional strutted processes scattered over the valve. One row of marginal spines. One prominent labiate process. Cells rectangular in girdle view. LIVE HABIT: Occurs as single cells, but could also form short chains. Areolation and labiate process easily visible in LM.
Thalassiosira species of the German North Sea
277
Figs 16–28. SEM of Thalassiosira species. Arrows indicate labiate processes, arrowheads strutted processes. Figs 16–18. T. decipiens. Fig. 16. Girdle view of complete cell. Fig. 17. Valve outside view. Fig. 18. Valve inside view. Figs 19–21. T. delicatula Figs 19, 20. Valve outside view. Fig. 21. Valve inside view. Figs 22–24. T. diporocyclus. Fig. 22. Girdle view of complete cell. Fig. 23. Valve outside view with girdle band. Fig. 24. Valve inside view, Figs 25, 26. T. eccentrica. Fig. 25. Valve outside view. Fig. 26. Girdle view. Figs 27, 28. T. guillardii. Fig. 27. Girdle view, note the three subcentral strutted processes. Fig. 28. Valve outside view, note the two subcentral strutted processes (arrowheads). Scale bars: 5 mm (Figs 16–24, 27, 28), 10 mm (Figs 25, 26).
M. Hoppenrath et al. DISTRIBUTION: Worldwide occurrence. SEASONALITY: Isolated from Helgoland and Sylt; recorded all year around at Sylt, with a clear winter maximum and lowest frequency from May to July (Fig. 57). T. guillardii Hasle (Figs 27, 28) SYNONYM: Cyclotella nana Guillard in Guillard & Ryther (clone 7–15) Cells 8.5–10.0 (4.0–14.0 (35.0)) mm in diameter. Valve with very delicate ‘areolation’ at the margin (70 to 80 areolae in 10 mm) and fine siliceous ribs radiating from the centre. Two (0–3) subcentral strutted processes; one marginal ring of regularly spaced strutted processes (seven to ten in 10 mm). One labiate process taking the place of a strutted one. Cells quadrangular to rectangular in girdle view. LIVE HABIT: Occurs as single cells. DISTRIBUTION: Unknown. SEASONALITY: Isolated from Sylt in February, but not recorded during routine monitoring. T. hendeyi Hasle et Fryxell (Fig. 29) SYNONYM: Coscinodiscus hustedtii Mu¨ller-Melchers Cells 60.0–82.0 (38.0–120.0) mm in diameter. Valve with straight/linear areolation with relatively coarse hexagonal areolae; five to six areolae in 10 mm. Valve face covered by minute siliceous granules. One tiny central strutted process. Three marginal rings of small strutted processes; five to six in 10 mm. Two prominent labiate processes on opposite sides of the valve. Cells rectangular in girdle view. LIVE HABIT: Occurs as single cells, but could also form short chains. Areolation and labiate processes easily visible in LM. DISTRIBUTION: Occurs in temperate to warm water regions. SEASONALITY: Isolated from Helgoland and Sylt; recorded from October until April at Sylt and from August until May at Helgoland.
278 scattered on the valve face. One labiate process taking the place of a strutted one. Cells disc-shaped in girdle view. Connecting threads in three to six groups at some distance from the centre. LIVE HABIT: Occurs as single cells. DISTRIBUTION: Occurs in temperate waters. SEASONALITY: Seen under SEM from Helgoland in March; but not recorded during routine monitoring. T. lundiana Fryxell (Figs 31, 32) Cells 20.8–27.6 (7.0–43.0) mm in diameter. Valve with fine areolae arranged in sectors (24 to 30 areolae in 10 mm), fasciculated and marginal striae. One central strutted process; one marginal ring of strutted processes; several strutted processes scattered over the valve face. One ring of up to 16 large occluded marginal processes. One labiate process in the marginal ring. Cells convex in girdle view. LIVE HABIT: Only as single cells, but probably also occurs in chains. Areolation not visible in LM. DISTRIBUTION: Unknown. SEASONALITY: Isolated from Sylt in December, but not recorded during routine monitoring. T. mediterranea (Schro¨der) Hasle (Figs 33, 34) BASIONYM: Coscinosira mediterranea Schro¨der SYNONYM: T. stellaris Hasle et Guillard in Fryxell & Hasle Cells 8.9–16.2 (6.0–20.0) mm in diameter. Valve with relatively fine areolation (about 30 areolae in 10 mm) with two to eight radial rays extending from the centre. Siliceous granules at the corners between areolae. Single strutted processes (two to seven) in a ring half way between centre and margin. One marginal ring of strutted processes; three to six in 10 mm. One labiate process located between two marginal strutted processes. DISTRIBUTION: Occurs in warm to temperate waters. SEASONALITY: Isolated from Sylt in December and April, but not recorded during routine monitoring. T. minima Gaarder (Figs 35–37)
T. kuschirensis Takano (Fig. 30) SYNONYM: T. solitaria Gayoso
SYNONYM: Coscinosira floridana Cooper T. floridana (Cooper) Hasle
Cells 12.9–13.5 (6.0–35.0) mm in diameter. Valve with fasciculate areolation; 15 to 24 (or 40 to 45 at the mantle) areolae in 10 mm. Central strutted processes in clusters (1 to 3) in a central ring. One marginal ring of short tube-like strutted processes with two wings parallel to the valve margin; seven to eight in 10 mm. Several strutted processes
Cells 6.2–10.0 (5.0–15.0) mm in diameter. Valve with radial rows of areolae; 30 to 40 in 10 mm. Two or sometimes one central strutted processes with short external tube. One marginal ring of 7 to 12 tube-like strutted processes, with special granule (small process) in front of them (Fig. 35 small arrows). One large funnel-shaped labiate process
Thalassiosira species of the German North Sea
279
Figs 29–40. SEM of Thalassiosira species. Arrows indicate labiate processes, arrowheads strutted processes. Fig. 29. T. hendeyi, valve outside view, Fig. 30. T. kuschirensis, valve outside view. Figs 31, 32. T. lundiana. Fig. 31. Valve outside view. Fig. 32. Valve inside (upper left) and outside views. Figs 33, 34. T. mediterranea, note the single strutted process half way between centre and margin (arrowheads). Fig. 33. Valve outside view. Fig. 34. Valve inside view. Figs 35–37. T. minima. Valve outside views, note the different number of central strutted processes (arrowheads), the labiate process (large arrows), the granule associated with the marginal strutted processes (small arrows) and the girdle bands (Fig. 37). Figs 38–40. T. minuscula. Fig. 38. Valve outside view. Fig. 39. Valve in oblique side view. Fig. 40. Valve inside view. Scale bars: 10 mm (Figs 29, 31, 32), 5 mm (Figs 30, 33–40).
M. Hoppenrath et al. next to a marginal strutted process, just inside the marginal ring of strutted processes. Cells rectangular in girdle view. LIVE HABIT: Occurs in chains, often with attached fine detritus. DISTRIBUTION: Worldwide occurrence. SEASONALITY: Isolated from Helgoland and Sylt, recorded all year round at Sylt, with highest numbers in winter and early spring (Fig. 58). Observed from November until May at Helgoland with highest frequencies in net samples in March and April.
T. minuscula Krasske (Figs 38–40) SYNONYM: T. monoporocyclus Hasle Cells 9.3–15.7 (10.0–24.0) mm in diameter. Valve with fine areolae (32 to 48 areolae in 10 mm) forming sectors. One central strutted process; one marginal ring of strutted processes without external tubes; 3–4 mm apart. One conspicuously large labiate process close to one or two strutted processes just inside the marginal ring. Valve with curved mantle. Pervalvar axis shorter than cell diameter. LIVE HABIT: Occurs as single cells or in mucilage colonies. DISTRIBUTION: Occurs in temperate water regions. SEASONALITY: Isolated from Helgoland in October and December, but not recorded during routine monitoring.
T. nordenskioeldii Cleve (Figs 41–43) Cells 8.6–43.0 (9.0–50.0) mm in diameter. Valve with hexagonal areolae in sectors; 14 to 18 areolae in 10 mm. One central strutted process next to an annulus; one marginal ring of prominent strutted processes with long external tubes; three in 10 mm. Marginal strutted processes with collar at the top. One large, tube-like labiate process in the marginal ring of strutted processes. Cells octagonal in girdle view, connected with a central thread. LIVE HABIT: Occurs in (long) chains with relatively short connecting thread. Central connecting thread and radiating marginal threads visible in LM. DISTRIBUTION: Occurs in cold to temperate waters. SEASONALITY: Isolated from Helgoland and Sylt; being recorded from October until May at Sylt, with highest numbers in April (Fig. 59); and from January until May at Helgoland, also with highest numbers in April. This species may bloom in spring.
280 T. oceanica Hasle (Fig. 44) SYNONYM: Cyclotella nana Guillard in Guillard & Ryther (clone 13-1) Cells 5.5–10.8 (3.0–12.0) mm in diameter. Valve with particularly fine ornamentation of radial ribs and/or poroid areolation (40 to 60 areolae in 10 mm) covered by a finely perforated layer; undulating marginal ridge. One central strutted process; one marginal ring of (six to eight) relative widely spaced strutted processes without external tubes. One labiate process close to one marginal strutted process. Cells rectangular in girdle view. Pervalvar axis shorter as cell diameter. LIVE HABIT: Occurs as single cells (in samples fixed for SEM). DISTRIBUTION: Occurs mainly in warm waters, cosmopolitan. SEASONALITY: Seen with SEM from Helgoland in February and March, but not recorded during routine monitoring.
T. pacifica Gran et Angst (Figs 45, 46) Cells 5.0–20.0 (7.0–55.0) mm in diameter. Valve with areolae in straight or curved rows; 20 to 28 areolae in 10 mm. Valve face covered by tiny siliceous granules. One central strutted process adjacent to an annulus; one marginal ring of strutted processes (relatively long coarse external tubes); four to seven in 10 mm. One labiate process (slightly larger than the marginal strutted processes) nearly takes the place of a marginal strutted process or lies between two strutted processes, a little inside of the marginal ring. Cells rectangular in girdle view. LIVE HABIT: Usually occurs as single cells, but also forms short chains. Areolation hardly visible in LM. DISTRIBUTION: Occurs in polar regions. SEASONALITY: Isolated from Helgoland and Sylt. Identified very rarely in spring, but not recorded during routine monitoring.
T. partheneia Schrader (Fig. 47) Cells 6.9–7.6 (6.0–14.0) mm in diameter. Valve with relatively coarse fasciculate areolation; 40 to 60 areolae in 10 mm. One central strutted process; one marginal ring of short strutted processes; three to five in 10 mm. One marginal labiate process in between tow strutted processes. Cells convex in girdle view. LIVE HABIT: Normally occurring in colonies with cells entangled in threads.
Thalassiosira species of the German North Sea
281
Figs 41–54. LM (Fig. 51) and SEM of Thalassiosira species. Arrows indicate labiate processes, arrowheads strutted processes. Figs 41–43. T. nordenskioeldii. Figs 41, 42. Valve outside view. Fig. 43. Valve inside view. Fig. 44. T. oceanica. Figs 45, 46. T. pacifica. Valve outside views. Fig. 47. T. partheneia. Valve outside view. Fig. 48. T. proschkinae. Valve outside view, note the central (!) labiate process (arrow) next to the central strutted process (arrowhead). Fig. 49. T. punctigera. Valve outside view of a complete cell. Fig. 50. T. rotula. Oblique girdle view showing also the valve face with central threads (arrowhead) and labiate process (arrow). Fig. 51. T. subtilis. detail of a larger colony. Fig. 52. T. tealata. Oblique valve view. Figs 53, 54. T. tenera. Fig. 53. Complete cells showing the size range of the species. Fig. 54. Oblique valve view. Scale bars: 30 mm (Fig. 51), 10 mm (Figs 41–43, 45, 46, 49, 50, 54), 5 mm (Figs 44, 47, 52), 1 mm (Fig. 48).
DISTRIBUTION: Occurs mainly in warm (to temperate) waters. SEASONALITY: Isolated at Sylt in December, but not recorded during routine monitoring. T. proschkinae Makarova (Fig. 48) Cells 4.7–5.0 (2.5–11.5) mm in diameter. Valve with eccentric areolation (25 to 30 areolae in 10 mm)
covered by a special ornamentation (irregular coarse granulation forming radial ‘ribs’). One central strutted process; one marginal ring of relative widely spaced strutted processes (about 1.5 mm apart) without external tubes. One central labiate process close to the strutted one. Cells rectangular, short in girdle view. LIVE
HABIT:
Occurs as single cells (or in colonies).
M. Hoppenrath et al.
282
DISTRIBUTION: Worldwide occurrence. SEASONALITY: Seen only with SEM from Helgoland in February; not recorded during routine monitoring. T. punctigera (Castracane) Hasle (Fig. 49) BASIONYM: Ethmodiscus punctiger Castracane SYNONYM: Ethmodiscus japonicus Castracane T. angstii (Gran) Makarova T. japonica Kiselev further synonyms in Hasle 1983 Cells 24.3–80.0 (40.0–186.0) mm in diameter. Valve with fine fasciculated areolation (10 to 23 areolae in 10 mm) and ribbed margin. One central strutted process; one marginal ring of many close tubular strutted processes (four to five in 10 mm), which have an external tulip-shaped part. One (inner) ring of widely spaced, large occluded marginal processes (variable in number) could be present (very variable feature). One very long, tubular labiate process just inside the marginal ring. Cells convex, disc-shaped in girdle view. LIVE HABIT: Usually occurs as single cells, but could also form chains. Areolation not visible in LM, but the strikingly long labiate process visible, especially in girdle view. DISTRIBUTION: Occurs in warm to temperate waters. SEASONALITY: Isolated from Helgoland and Sylt, present all year round (Fig. 60); slightly lower frequencies during June and July.
Figs 55–62. Seasonal frequency (¼ relative occurrence) during the monitoring of living taxa from net samples at Helgoland (striated bars) and at Sylt/North Frisian Wadden Sea (dark bars).
T. subtilis (Ostenfeld) Gran (Fig. 51) BASIONYM: Podosira (?) subtilis Ostenfeld
T. rotula Meunier (Fig. 50) Cells 24.9.0–60.0 (8.0–61.0) mm in diameter. Valve with fine radial ribs and ‘areolation’ only at the margin; 18 to 24 areolae in 10 mm. Cluster of central strutted processes; marginal rings of strutted processes (probably 12 to 15 in 10 mm); several strutted processes scattered on the valve face; all tube-shaped. One relatively small marginal labiate process. Cells flattened/rectangular in girdle view, discoid. Unevenly thick intercalary bands. LIVE HABIT: Occurs in (long) chains with relatively short connecting threads. Central connecting threads (seen as one thick thread) easily visible in LM. The uneven thickness of the intercalary bands visible as white lines, when focussing on girdle view. DISTRIBUTION: Occurs in cold-temperate to warm waters. Seasonality: Isolated from Helgoland and Sylt, observed all year round. Bi-modal frequencies, with highest values in March/April and in October (Fig. 61).
Cells 15.0–32.0 mm in diameter. Valve with fine fasciculate areolae; about 30 areolae in 10 mm. One subcentral strutted process; one marginal ring of and scattered intermediate strutted processes without external tubes. One labiate process in the second marginal ring. Cells convex with curved mantle in girdle view. Pervalvar axis as long as cell diameter. LIVE HABIT: Occurs in irregular mucilage colonies. DISTRIBUTION: Occurs in temperate to tropical (?) waters. SEASONALITY: Recorded at Helgoland in July and December 1969 by Drebes. Because only LM observations (Drebes, 1974) are available, it is possible that it was T. diporocyclus (see above), a very similar taxon (Hasle, 1972). Not recorded during routine monitoring. T. tealata Takano (Fig. 52) Cells 8.6–11.2 (6.0–10.6) mm in diameter. Valve with radial rows of areolae (30 to 40 areolae in 10 mm) and siliceous granules all over
Thalassiosira species of the German North Sea the surface. One central strutted process next to an annulus. One marginal ring of four to seven conspicuous winged strutted processes. Two long wings per process with a single slender tip. One labiate process next to a marginal strutted process, just inside the marginal ring of strutted processes. Cells discoid to octangular in girdle view. LIVE HABIT: Occurs in chains. DISTRIBUTION: Occurs in cold to warm-temperate waters. SEASONALITY: Isolated from Helgoland and Sylt, but not distinguished from T. curviseriata (see above) during monitoring. T. tenera Proschkina-Lavrenko (Figs 53, 54) Cells 10.0–45.5 (10.0–29.0) mm in diameter. Valve with straight/linear areolation (most of the time) with relatively coarse hexagonal areolae; 10 to 16 areolae in 10 mm. One central strutted process, surrounded by a central areola slightly larger than the other areolae. One marginal ring of closely standing (canine) tooth-shaped strutted processes; three to five in 10 mm. One labiate processes in the marginal ring directly next to a strutted one. Cells rectangular in girdle view. LIVE HABIT: Occurs as single cells. Areolation and strutted processes easily visible in LM. DISTRIBUTION: Cosmopolitan species. SEASONALITY: Isolated from Helgoland and Sylt. Observed from July until April at Helgoland, with highest frequencies from October until February;
283 and all year round at Sylt, with highest frequencies from November until February (Fig. 62).
Relative importance of Thalassiosira in the Helgoland Reede 1962–2003 Throughout the study period Thalassiosira species rarely comprise more than 10% of the total diatom cell counts (>10% Thalassiosira on only 915 out of 8378 days) and thus cannot be considered as numerically important (Figs 63, 64). The monthly relative contributions of Thalassiosira species to the total diatom counts for 1962 to 2000 are shown in Fig. 63, and the means relative to all diatoms are shown in Fig. 64. This reveals the relative seasonality patterns of Thalassiosira populations. However, there is enormous variability in cell numbers over the period. Cell counts vary between 0 and 1,400,0001 and most cell counts are under 200,000 cells l1. Is the seasonality insignificant? Probably, but Fig. 64 also suggests that the genus is most important in April. The above variability is also reflected in the detailed examination of data from 2001–2003. The Helgoland Reede mean monthly relative percentages, (Thalassiosira relative to all diatoms) are shown in Fig. 65. It is clear that Thalassiosira was relatively unimportant in terms of cell numbers, especially in 2001 and 2002, but also that the years are extremely different. There is no clear seasonality over the entire period, mainly due to the 2003 data. There were distinct spring and summer populations in 2001 and 2002.
Fig. 63. Helgoland Reede, percentage of Thalassiosira for all months from 1962 to 2000.
M. Hoppenrath et al.
284
Fig. 64. Helgoland Reede mean monthly % of Thalassiosira cell numbers relative to total diatom counts. Mean distribution for all months 1962–2003.
Fig. 65. Helgoland Reede mean monthly % of Thalassiosira cell numbers relative to the total diatom counts (2001–2003).
Molecular data The 18S rDNA sequences provide more complete characterization of some species for future monitoring studies; the sequences were deposited in GenBank under the accession numbers AJ810854AJ810859, AM050629. Phylogenetic analyses were performed to integrate these data with published 18S rDNA sequences for other Thalassiosirales species. Fig. 66 shows the results of these phylogenetic analyses. Our results consistently supported the monophyly of the Thalassiosiraceae and resolved a few strongly supported clades within this group, but failed to provide good resolution within some of these clades. Cyclotella and Skeletonema were monophyletic in all analyses with strong support. This was not the case with Thalassiosira, which did not form a single clade in any of the analyses. Instead, it separated into three distinct, well-supported clades. The clade comprising T. weissflogii, T. oceanica and T. guillardii (clade C in Fig. 66) formed the first divergence at the base of the Thalassiosiraceae. The second well-supported divergence separated a clade consisting of
Fig. 66. Phylogenetic tree (neighbour joining) for the 18S rDNA sequences using maximum likelihood distances with the best fitting nucleotide substitution model (base frequencies: A – 0.2678, C – 0.1974, G – 0.2542, substitution rates: rAC, rAT, rCG – 1.0000, rAG – 2.5421, rCT –4.3784), a gamma distribution with a shape parameter of 0.5474 and proportion of invariable sites ¼ 0.4474 accounting for among-site rate variation). Scale bar represents 0.01 unit corrected evolutionary distance. Bootstrap values above 60% are shown above the nodes, with the values from neighbour joining (with maximum likelihood distances based on 1000 replicates) followed by the maximum parsimony (500 replicates) bootstrap values. Species names corresponding to sequences obtained in this study are printed in bold. The groups are labelled as in Kaczmarska et al. (2006), except that, considering the weak support for its monophyly, their clade F is shown as Group F. The tree was rooted using Porosira pseudodelicatula (AY485469), P. pseudodenticulata (X85398) and Ditylum brightwellii (AY188181). The groups containing sequences from Cyclotella and Skeletonema spp. were collapsed because of the strong support for their monophyly.
Cyclotella spp., Detonula and two sequences designated T. pseudonana from the clade including Skeletonema, Minidiscus and all remaining Thalassiosira sequences. All the sequences from North Sea cultures grouped within the last clade, as a moderately supported sister group to Skeletonema. The relationships within this group (designated group F in Fig. 66) remained almost completely unresolved. The only strongly supported, informative clade grouped T. rotula with T. eccentrica and T. punctigera. In the maximum likelihood tree, T. delicatula grouped at the base of this clade, but this relationship was not supported
Thalassiosira species of the German North Sea by bootstrap values. The rest of group F collapsed into a polytomy in bootstrap consensus trees (in both maximum parsimony and neighbour joining analysis). The paraphyly of Thalassiosira is strongly supported: the clade including all Thalassiosira sequences also includes the freshwater genus Cyclotella and the marine genera, Minidiscus, Detonula, and Skeletonema. The phylogenetic position of Skeletonema is ambiguous. Depending on the analyses used, it either constituted the sister group to clade F, or was within this clade. Discussion Species records and seasonality From the 1966 to 1971 species survey (Drebes, 1974) until the present study there was no continuous record of species composition from net-samples at Helgoland. Drebes (1974) provided descriptions of selected species, including eight Thalassiosira species, namely T. nordenskioeldii, T. rotula, T. fallax Meunier, T. decipiens, T. levanderi Goor, T. anguste-lineata as T. polychorda, T. subtilis, and T. eccentrica as Coscinodiscus eccentricus Ehrenberg for Helgoland, but he (Drebes, 1974) was unsure about the occurrence of T. gravida Cleve. A later checklist (Drebes & Elbra¨chter, 1976) listed nine Thalassiosira species for the German Bight (the above eight plus T. mala Takano), seven being recorded from Sylt, while Du¨rselen & Rick (1999) investigated the spatial and temporal distribution of T. punctigera (Castracane) Hasle in the same area. Phytoplankton monitoring has been conducted at the Wadden Sea Station, Sylt, since 1985, and 14 Thalassiosira species have been routinely identified: T. angulata, T. angustelineata, T. decipiens, T. eccentrica, T. hendeyi, T. minima, T. nordenskioeldii, T. punctigera, T. rotula, T. tenera, while T. aestivalis/ concaviuscula and T. curviseriata/tealata were recorded as species groups. The present study revealed the presence of at least 27 species, 21 at Helgoland and 21 at Sylt, plus three unidentified taxa. Fifteen species had not previously been recorded for Helgoland, and 16 for Sylt, and thus represent new records for these areas. Most Thalassiosira species showed a similar seasonal cycle with only minor exceptions, e.g. T. angulata and T. tenera had shorter summer minima. T. minima occurred more frequently near Sylt, which might indicate the importance of benthic stages in the life cycle. However, differences in frequency could also be the result of the longer time series for Sylt, or indicate inter-annual differences in the presence of
285 Thalassiosira species. The Thalasssiosira counts at Helgoland showed that the genus comprised >10% total diatom numbers in about 10% of the samplings. This corresponds to about one month per year (in spring), when Thalassiosira species could be relatively important. Species diversity Comparison of our results with previously published observations at other geographic regions revealed that species diversity is relatively high. Elsewhere 10 Thalassiosira species were recorded from the Elbe estuary and 13 from the Schelde estuary (Muylaert & Sabbe, 1996), while Harris et al. (1995) found 18 taxa in a Scottish sea-loch. Of the species recorded from the Elbe, nine also occurred at Helgoland, and eight at Sylt. The only taxon exclusive to the Elbe was T. pseudonana, which was restricted to tidal freshwater (Muylaert & Sabbe, 1996). Lange et al. (1992) listed 28 Thalassiosira species, plus two unidentified ones, for the Skagerrak, close to our investigation region. Further afield, Mahood et al. (1986) found 20 taxa in the San Francisco Bay system, Herzig & Fryxell (1986) found 20 taxa in Gulf Stream warm core rings, Sar et al. (2002) found 18 taxa in the northern San Matı´ as Gulf, Argentina and Fukuyo et al. (1990) recorded 28 species from Japan. Similar species diversity (25 taxa) has been reported in Australian tropical and subtropical waters (Hallegraeff, 1984). The variation in species richness may reflect differences in sampling effort; species diversity tends to increase with sampling intensity, and the sampling effort of the current study was very high. Phylogenetic studies Recent phylogenetic analyses of 18S rDNA sequences of the Thalassiosirales (Kaczmarska et al., 2006) have indicated the existence of several well-supported clades in this group, but proved less successful in resolving the within-clade relationships. Our phylogenetic analyses (Fig. 66) are largely in accordance with those of Kaczmarska et al. (2006), and we used the same clade designations for ease of comparisons. The paraphyly of Thalassiosira is strongly supported. Thalassiosira species occur in three large groups in the phylogenetic trees. Detonula confervacea, and T. pseudonana form a well-supported sister clade to Cyclotella; the monophyly of the group including these species and Cyclotella (clade F) is also strongly supported by bootstrap values. T. weissflogii and T. oceanica form another well supported clade (clade C), which previously (Kaczmarska et al. 2006) only contained
M. Hoppenrath et al. T. weissflogii and the sequence AJ535170, which they designated T. fluviatilis. Because the latter is now considered a synonym of the former, we labelled this sequence T. weissflogii. Thalassiosira guillardii (not included by Kaczmarska et al., 2006) grouped with these species in our analyses (although with significantly weaker support: 82% neighbour joining, 48% maximum parsimony). All new Thalassiosira sequences belong to the clade F, with unresolved within-clade relationships, except T. punctigera, which showed strongly supported affinities to T. rotula/T. eccentrica. In contrast to Kaczmarska et al.’s (2006) results (no significant bootstrap support for the monophyly of group F, excluding Skeletonema), this group had moderate bootstrap support (76% neighbour joining; 64% maximum parsimony). We conclude that neither the monophyly of this group nor the phylogenetic relationships within it are unambiguously resolved by this 18S rDNA dataset. The polytomy at the base of the group probably indicates that it originates from a strong radiation event of marine Thalassiosira species. Better resolution of the within-group relationships may be possible with the use of less conserved molecular markers. Methodological issues The two methods used for the monitoring (qualitative net-plankton and quantitative bottle sample investigations) resolved the contribution of Thalassiosira to the phytoplankton differently. Species composition was recorded in a time-series framework at Helgoland (Wiltshire & Du¨rselen, 2004) and until 2001, six Thalassiosira species were distinguished and enumerated, namely T. angustelineata, T. eccentrica, T. minima, T. nordenskioeldii, T. punctigera and T. rotula. All other species were counted in size classes (see Wiltshire & Du¨rselen, 2004). Because the numbers of Thalassiosira relative to total diatom numbers are relatively small, it is difficult to improve the resolution of Utermo¨hl counts and to identify the rarer species in routine analyses. We recommend the use of regular net hauls to obtain quasi-total species lists, particularly where rare species are involved. It is also important to distinguish between routine daily investigations and those aimed at maximizing taxonomic resolution. The identification of Thalassiosira species, especially small ones, is problematic. For many taxa, the morphological features that are essential for their identification, such as number and distribution of processes, are only visible using electron microscopy. Thus the lack of records for many of these species is probably due to their small size and the scope of previous studies, rather
286 than their absence. Thalassiosira diporocyclus, T. kuschirensis, T. minuscula, T. oceanica, T. proschkinae and T. subtilis (only by Drebes in 1969) were exclusively found at the Helgoland Reede station, whereas T. aestivalis, T. constricta, T. guillardii, T. lundiana, T. mediterranea and T. partheneia were only identified for the Sylt station. Thalassiosira angulata, T. anguste-lineata, T. concaviuscula, T. curviseriata, T. decipiens, T. delicatula, T. eccentrica, T. hendeyi, T. minima, T. nordenskioeldii, T. pacifica, T. punctigera, T. rotula, T. tealata and T. tenera were recorded in the open North Sea (at Helgoland) as well as in the Wadden Sea (at Sylt) phytoplankton community. The only difference between T. aestivalis and T. concaviuscula is the position of the labiate process, which replaces a strutted process in T. aestivalis (Gran & Angst, 1931; Harris et al., 1995) and lies between two strutted processes in T. concaviuscula (Marakova, 1978). The strutted processes of T. curviseriata have two wings diverging into two or three branches (Takano, 1981), whereas in T. tealata the long wings each have a single slender tip (Takano, 1980). Whether these two species groups represent four species or simply the morphological variability of two species requires reinvestigation with morphometric and genetic analyses of clonal cultures. T. constricta was reliably identified only by its resting spores (Heimdal, 1971), while T. delicatula and other tiny species were listed as Thalassiosira sp. and spp. (small and generally solitary) respectively. Eleven tiny species were identified by isolating and culturing (clonal or mixed cultures) and using SEM. This is a time-consuming and highly selective method that is useful for taxonomic accounts, but not for regular monitoring work. Clonal cultures have long been an important tool for resolving phytoplankton systematics, but with the increasing use of molecular markers for diversity studies of eukaryotic microorganisms, they are also critical to the development of molecular markers for phylogenetic and diversity assessment purposes. Small subunit rDNA is the most widely used molecular marker for unicellular eukaryotes and its predictive value is dependent on the availability of sequences from a wide range of organisms, linked to reliable taxonomic identifications. rDNA sequences are becoming part of the routine taxonomic tools for characterizing phytoplankton species, while the availability of a large public small subunit rDNA dataset also provides the basis for diversity monitoring and ecological investigations using oligonucleotide probes (Simon et al., 2000; Biegala et al., 2003; John et al., 2003). While we obtained 18S rDNA data with the aim of contributing to the available Thalassiosira 18S
Thalassiosira species of the German North Sea rDNA database, we appreciate their potential use for monitoring purposes, particularly through the use of fluorescently labelled oligonucleotide probes (John et al., 2003) and microarrays (TaroncherOldenburg et al., 2003, Metfies & Medlin, 2004). These are particularly promising for taxa that are difficult or impossible to distinguish morphologically, such as T. aestivalis/T. concaviuscula and T. curviseriata/T. tealata. Future work With respect to future monitoring work, we stress the importance of establishing a collection of DNA sequences from well-characterized strains. SEM preparations and unprepared cleaned frustules of the Thalassiosira cultures used for sequencing, as well as samples of their DNA, have been archived at the Alfred Wegener Institute (AWI). These strains can thus form the basis of species identifications using either environmental clone libraries or molecular probes. We recommend the following strategy for a future species-level monitoring. In addition to routine Utermo¨hl counts, a detailed species check of living specimens from net-samples should be carried out once a week. This will also improve species identifications from Lugol-fixed samples. During winter and spring when Thalassiosira species diversity is high, SEM preparations should be made and investigated at least once a month (twice a month during the Thalassiosira spring bloom). This may not be possible in all monitoring laboratories, and we would support the idea of developing molecular probes for the problematic Thalassiosira species, using them every other week during the appropriate season. The above recommendations are equally applicable to the species-rich genus Chaetoceros but we also wish to reiterate a statement made by Reid et al. (1990, p. 296): ‘‘Complicated inter-relationships and successional patterns between individual species which are limited by varying physiological requirements and adaptation to differing hydrographic regimes re-emphasizes the importance of species identification in phytoplankton studies. Many future problems in phytoplankton research will not be resolved without accurate identification of algal species. Taxonomic expertise takes many years to acquire; there is at present a shortage of skills in this area and more resources should be turned towards training and long-term support.’’ Acknowledgements Dr L. Medlin, AWI Bremerhaven, kindly helped identify some Thalassiosira species and provided her laboratory facilities for the molecular genetic work.
287 We want to thank Dr E. Hagmeier for comments on an early version of this manuscript and Dr B. S. Leander, University of British Columbia, for improving the English.
References ASMUS, R., JENSEN, M.H., MURPHY, D. & DOERFFER, R. (1998). Prima¨rproduktion von Mikrophytobenthos, Phytoplankton und ja¨hrlicher Biomasseertrag des Makrophtobenthos im Sylt-Rømø Wattenmeer. In: O¨kosystem Wattenmeer: Austausch, Transport und Stoffwandlungsprozesse (Ga¨tje, C. & Reise, K., editors), 367–392. Springer Verlag, Heidelberg, Berlin, Germany. BAUERFEIND, E., HICKEL, W., NIERMANN, U. & WESTERNHAGEN, H. Von. (1990). Phytoplankton biomass and potential nutrient limitation of phytoplankton development in the southeastern North Sea in spring 1985 and 1986. Neth. J. Sea Res., 25: 131–142. BIEGALA, I.C., NOT, F., VAULOT, D. & SIMON, N. (2003). Quantitative assessment of picoeukaryotes in the natural environment by using taxon-specific oligonucleotide probes in association with tyramide signal amplification-fluorescence in situ hybridization and flow cytometry. Appl. Environ. Microbiol., 69: 5519–5529. BRAARUD, T., GAARDER, K.R. & GRØNTVED, J. (1953). The phytoplankton of the North Sea and adjacent waters in May 1948. Rapp. P.-v. Re´un. Cons. int. Explor. Mer., 133: 1–87. DIEZ, B., PEDROS-ALIO, C. & MASSANA, R. (2001). Study of genetic diversity of eukaryotic picoplankton in different oceanic regions by small-subunit rRNA gene cloning and sequencing. Appl. Environ. Microbiol., 67: 2932–2941. DREBES, G. (1974). Marines Phytoplankton, Eine Auswahl der Helgola¨nder Planktonalgen (Diatomeen, Peridineen). Georg Thieme Verlag, Stuttgart, Germany. DREBES, G. & ELBRA¨CHTER, M. (1976). A checklist of planktonic diatoms and dinoflagellates from Helgoland and List (Sylt), German Bight. Bot. Mar., 19: 75–83. DU¨RSELEN, C.-D. & RICK, H.-J. (1999). Spatial and temporal distribution of two new phytoplankton diatom species in the German Bight in the period 1988 and 1996. Sarsia, 84: 367–377. EHRENHAUSS, S., WITTE, U., BU¨HRING, S.I. & HUETTEL, M. (2004). Effect of advective pore water transport on distribution and degradation of diatoms in permeable North Sea sediments. Mar. Ecol. Prog. Ser., 271: 99–111. ELWOOD, H.J., OLSON, G.J. & SOGIN, M.L. (1985). The smallsubunit ribosomal RNA gene sequences from the hypotrichous ciliates Oxytricha nova and Stylonichia putalata. Mol. Biol. Evol., 2: 399–410. FRYXELL, G.A. (1975). Three new species of Thalassiosira, with observations on the occluded process, a newly observed structure of diatom valves. Nov. Hedwig. Beih., 53: 57–75. FRYXELL, G.A., GOULD, R.W. & WATKINS, T.P. (1984). Gelatinous colonies of the diatom Thalassiosira in Gulf Stream warm core rings including T. fragilis, sp. nov. Br. Phycol. J., 19: 141–156. FRYXELL, G.A. & HASLE, G.R. (1972). Thalassiosira eccentrica (Ehrenb.) Cleve, T. symmetrica sp. nov., and some related centric diatoms. J. Phycol., 8: 297–317. FRYXELL, G.A. & HASLE, G.R. (1977). The genus Thalassiosira: some species with a modified ring of central strutted processes. Nov. Hedwig. Beih., 54: 67–98. FUKUYO, Y., TAKANO, H., CHIHARA, M. & MATSUOKA, K. (1990). Red Tide Organisms in Japan – An Illustrated Taxonomic Guide. Uchida Rokakuho, Tokyo, Japan. GRAN, H.H. & ANGST, E.C. (1931). Plankton diatoms of Puget Sound. Publ. Puget Sound Biol. St., 7: 417–519. GRØNTVED, J. (1952). Investigations on the phytoplankton in the southern North Sea in May 1947. Meddrs. Komm. Danm. Fisk. Havunders., Ser. Plankton, 5: 1–49. HAGMEIER, E. (1971). Coastal influence on microplankton in Heligoland Bay (North Sea). Thalassia Jugoslavica, 7: 87–100.
M. Hoppenrath et al. HAIGH, R., TAYLOR, F.J.R. & SUTHERLAND, T.F. (1992). Phytoplankton ecology of Sechelt Inlet, a fjord system on the British Columbia coast. I. General features of the nano- and microplankton. Mar. Ecol. Prog. Ser., 89: 117–134. HALLEGRAEFF, G.M. (1984). Species of the diatom genus Thalassiosira in Australian waters. Bot. Mar., 27: 495–513. HARRIS, A.S.D., MEDLIN, L.K., LEWIS, J. & JONES, K.J. (1995). Thalassiosira species (Bacillariophyceae) from a Scottish sea-loch. Eur. J. Phycol., 30: 117–131. HASLE, G.R. (1972). Thalassiosira subtilis (Bacillariophyceae) and two allied species. Norw. J. Bot., 19: 111–137. HASLE, G.R. (1973a). Thalassiosiraceae, a new diatom family. Norw. J. Bot., 20: 67–69. HASLE, G.R. (1973b). Some marine plankton genera of the diatom family Thalassiosiraceae. Nov. Hedwig. Beih., 45: 1–49. HASLE, G.R. (1978a). Some Thalassiosira species with one central process (Bacillariophyceae). Norw. J. Bot., 25: 77–110. HASLE, G.R. (1978b). Some freshwater and brackish water species of the diatom genus Thalassiosira Cleve. Phycologia, 17: 263–292. HASLE, G.R. (1979). Thalassiosira decipiens (Grun.) Jørg. (Bacillariophyceae). Bacillaria, 2: 85–108. HASLE, G.R. (1983). Thalassiosira punctigera (Castr.) comb. nov., a widely distributed marine planktonic diatom. Nordic J. Bot., 3: 593–608. HASLE, G.R. & FRYXELL, G.A. (1977). The genus Thalassiosira: some species with a linear areola array. Nov. Hedwig. Beih., 54: 15–66. HASLE, G.R. & HEIMDAL, B.R. (1970). Some species of the centric diatom genus Thalassiosira studied in the light and electron microscopes. Nova. Hedwig. Beih., 31: 559–581. HASLE, G.R. & SMAYDA, T.J. (1960). The annual phytoplankton cycle at Drobak, Oslofjord. Nytt. Mag. Bot., 8: 53–75. HASLE, G.R. & SYVERTSEN, E. (1996). Marine diatoms. In: Identifying Marine Diatoms and Dinoflagellates (Tomas, C.R., editor), 5–385. Academic Press, Inc., San Diego, USA. HEIMDAL, B.R. (1971). Vegetative cells and resting spores of Thalassiosira constricta Gaarder (Bacillariophyceae). Norw. J. Bot., 18: 153–159. HERZIG, W.N. & FRYXELL, G.A. (1986). The diatom genus Thalassiosira Cleve in the Gulf Stream core rings: taxonomy, with T. intrannula and T. lineoides, spp. nov. Bot. Mar., 29: 11–25. HICKEL, W., BERG, J. & TREUTNER, K. (1992). Variability in phytoplankton biomass in the German Bight near Helgoland, 1980–1990. ICES Mar. Sci. Symp., 195: 249–259. HILLEBRAND, H., DU¨RSELEN, C.-D., KIRSCHTEL, D., POLLINGHER, U. & ZOHARY, T. (1999). Biovolume calculation for pelagic and benthic microalgae. J. Phycol., 35: 403–424. HOPPENRATH, M. (2004). A revised checklist of planktonic diatoms and dinoflagellates from Helgoland (North Sea, German Bight). Helgol. Mar. Res., 58: 243–251. IANORA, A., MIRALTO, A., POULET, S.A., CAROTENUTO, Y., BUTTINO, I., ROMANO, G., CASOTTI, R., POHNERT, G., WICHARD, T., COLUCCI-D’AMATO, L., TERRAZZANO, G. & SMETACEK, V. (2004). Aldehyde suppression of copepod recruitment in blooms of a ubiquitous planktonic diatom. Nature, 429: 403–407. JOHN, U., CEMBELLA, A., HUMMERT, C., ELBRA¨CHTER, M., GROBEN, R. & MEDLIN, L.K. (2003). Discrimination of the toxigenic dinoflagellate species Alexandrium tamarense and Alexandrium ostenfeldii in co-occurring natural populations from Scottish coastal waters. Eur. J. Phycol., 38: 25–40. JOINT, I. & POMROY, A. (1993). Phytoplankton biomass and production in the southern North Sea. Mar. Ecol. Prog. Ser., 99: 179–182. KACZMARSKA, I., BEATON, M., BENOIT, A.C. & MEDLIN, L.K. (2006). Molecular phylogeny of selected members of the order Thalassiosirales (Bacillariophyta) and evolution of the fultoportula. J. Phycol., 42: 121–138.
288 KARENTZ, D. & SMAYDA, T.J. (1984). Temperature and seasonal occurrence patterns of 30 dominant phytoplankton species in Narragansett Bay over a 22-year period (1959–1980). Mar. Ecol. Prog. Ser., 18: 277–293. LANGE, C.B., HASLE, G.R. & SYVERTSEN, E.E. (1992). Seasonal cycle of diatoms in the Skagerrak, North Atlantic, with emphasis on the period 1980-1990. Sarsia, 77: 173–187. LUND, J.W.G., KIPLING, C. & LE CREN, E.D. (1958). The inverted microscope method of estimating algal numbers and the statistical basis of estimations by counting. Hydrobiologia, 11: 143–170. MAHOOD, A.D., FRYXELL, G.A. & MCMILLAN, M. (1986). The diatom genus Thalassiosira: species from the San Francisco Bay system. Proc. Calif. Acad. Sci., 44: 127–156. MARAKOVA, I.V. (1978). New data to the study of Thalassiosira Cl. (Bacillariophyta) from Atlantic Ocean. Bot. Zh., 63: 222–224. MEDLIN, L.K., ELWOOD, H.J., STICKEL, S. & SOGIN, M.L. (1988). The characterization of enzymatically amplified eukaryotic 16S-like rRNA-coding regions. Gene, 71: 491–499. METFIES, K. & MEDLIN, L. (2004). DNA microchips for phytoplankton: the fluorescent wave of the future. Nov. Hedwig. Beih., 79: 321–327. MOON-VAN DER STAAY, S.Y., DE WACHTER, R. & VAULOT, D. (2001). Oceanic 18S rDNA sequences from picoplankton reveal unsuspected eukaryotic diversity. Nature, 409: 607–610. MUYLAERT, K. & SABBE, K. (1996). The diatom genus Thalassiosira (Bacillariophyta) in the Estuaries of the Schelde (Belgium/ The Netherlands) and the Elbe (Germany). Bot. Mar., 39: 103–115. POSADA, D. & CRANDALL, K.A. (1998). MODELTEST: testing the model of DNA substitution. Bioinformatics, 14: 817–818. REID, P.C., LANCELOT, C., GIESKES, W.W.C., HAGMEIER, E. & WEICHART, G. (1990). Phytoplankton of the North Sea and its dynamics: a review. Neth. J. Sea Res., 26: 295–331. ROUND, F.E., CRAWFORD, R.M. & MANN, D.G. (1990). The Diatoms. Cambridge University Press, Cambridge, UK. SAR, E.A., SUNESEN, J. & LAVIGNE, A.S. (2002). The diatom genus Thalassiosira: species from the northern San Matias Gulf (Rio Negro, Argentina). Nov. Hedwig. Beih., 74: 373–386. SIMON, N., CAMPBELL, L., ORNOLFSDOTTIR, E., GROBEN, R., GUILLOU, L., LANGE, M. & MEDLIN, L.K. (2000). Oligonucleotide probes for the identification of three algal groups by dot blot and fluorescent whole-cell hybridization. J. Eukaryot. Microbiol., 47: 76–84. SOMERS, D. (1972). Scanning electron microscope studies on some species of the centric diatom genera Thalassiosira and Coscinodiscus. Biol. Jb. Dodonaea, 40: 304–315. SWOFFORD, D.L. (1998). PAUP, Phylogenetic analysis using Parsimony. Sinauer Associates, Sunderland, USA. TAKANO, H. (1980). New and rare diatoms from Japanese marine waters – V. Thalassiosira tealata sp. nov. Bull. Tokai Reg. Fish. Res. Lab., 103: 55. TAKANO, H. (1981). New and rare diatoms from Japanese marine waters – VI. Three new species in Thalassiosiraceae. Bull. Tokai Reg. Fish. Res. Lab., 105: 34. TARONCHER-OLDENBURG, G., GRINER, E.M., FRANCIS, C.A. & WARD, B.B. (2003). Oligonucleotide Microarray for the Study of Functional Gene Diversity in the Nitrogen Cycle in the Environment. Appl. Environ. Microbiol., 69: 1159–1171. THOMPSON, J.D., GIBSON, T.J., PLEWNIAK, F., JEANMOUGIN, F. & HIGGINS, D.G. (1997). The ClustalX windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucl. Acids Res., 24: 4876–4882. WILTSHIRE, K.H. & DU¨RSELEN, C.-D. (2004). Revision and quality analyses of the Helgoland Reede long-term phytoplankton data archive. Helgol. Mar. Res., 58: 252–268. WILTSHIRE, K.H. & MANLY, B.F.J. (2004). The warming trend at Helgoland Reede, North Sea: phytoplankton response. Helgol. Mar. Res., 58: 269–273.
