Monoraphid and Naviculoid Diatoms from the Coastal Laurentian Great Lakes

By Euan D. Reavie and Norman A. Andresen

This volume contains two monographs. 1) Monoraphid Diatoms from the Coastal Laurentian Great Lakes, by Euan D. Reavie (18 plates. 109 p.) This monograph contains descriptions of taxa from the diatom genera Achnanthes, Achnanthidium, Platessa, Psammothidium, Rossithidium, Planothidium, Karayevia, Eucocconeis, Cocconeis and Rhoicosphenia from periphytic and surface sediment samples in the coastal ecosystems of the Laurentian Great Lakes. We provide light micrographs of diatom taxa recorded in 207 samples from 106 wetlands, embayments, high-energy and deep, nearshore locales. We characterize 79 taxa and name two previously undescribed taxa. For the 50 more common taxa we characterized lake and habitat specificity, modeled optima for phosphorus and chloride and tolerance to coastal anthropogenic stressors. 2) Navicula from the Coastal Laurentian Great Lakes, by Euan D. Reavie and Norman A. Andresen (15 plates, 221 p.) This monograph contains descriptions of taxa from the diatom genus Navicula and some closely related taxa from the coastal ecosystems of the Laurentian Great Lakes. We provide light micrographs of diatom taxa recorded in 207 samples from 106 wetlands, embayments, high-energy and deep, nearshore locales of the five Great Lakes. Diatoms were identified from periphytic and surface sediment samples. The taxonomic and iconographic sections of this book include descriptions and illustrations of the taxa encountered, as well as autecological information when available. 26 new species are described.

• Language: English
• Publisher: Koeltz Botanical Books
• Release date: Mar 1, 2021
• ISBN: 9783946583325

Book preview

CHAPTER 1

Monoraphid Diatoms

from the Coastal Laurentian Great Lakes

Euan D. REAVIE

ABSTRACT

This monograph contains descriptions of taxa from the diatom genera Achnanthes, Achnanthidium, Platessa, Psammothidium, Rossithidium, Planothidium, Karayevia, Eucocconeis, Cocconeis and Rhoicosphenia from periphytic and surface sediment samples in the coastal ecosystems of the Laurentian Great Lakes. We provide light micrographs of diatom taxa recorded in 207 samples from 106 wetlands, embayments, high-energy and deep, nearshore locales of the five Great Lakes. We characterize 79 taxa and name two previously undescribed taxa. For the 50 more common taxa we characterized lake and habitat specificity, modeled optima for phosphorus and chloride and tolerance to coastal anthropogenic stressors.

ACKNOWLEDGEMENTS

Diatom taxonomic support was provided by E. Stoermer†, J. Kingston†, A. Kireta, N. Andresen†, G. Sgro, J. Johansen and M. Ferguson. A. Bellamy, K. Kennedy and K. Rewinkel helped with graphic design and document production. This research was partially supported by a grant to G. Niemi from the U.S. Environmental Protection Agency’s Science to Achieve Results (STAR) Estuarine and Great Lakes (EaGLe) program through funding to the Great Lakes Environmental Indicators (GLEI) project, USEPA Agreement EPA/R–8286750. This document has not been subjected to the Agency’s required peer and policy review and therefore does not necessarily reflect the view of the Agency, and no official endorsement should be inferred.

1.1. INTRODUCTION

Since 2000, the USEPA-funded Great Lakes Environmental Indicators (GLEI) project (Niemi et al. 2006) has developed and applied indicators of condition for the coasts of the United States. The project focussed on development of indicators of coastal condition in the Laurentian Great Lakes. In addition to water quality, organic contaminants, vegetation, invertebrates, amphibians, fish and birds, hundreds of diatom collections enabled development of diatom-based indicators for water quality and anthropogenic stress (Reavie et al. 2006). Every type of coastal habitat was sampled: wetlands (coastal, protected and riverine), embayments, high-energy (unprotected) shorelines and nearshore locales within 3 km of shore (Danz et al. 2005). Investigations of diatom assemblages followed standard methods of light microscopy and identification based on siliceous cell wall ornamentation. Many works have resulted from these diatom investigations, including phosphorus (Reavie et al. 2006, Sgro et al. 2007) and coastal stressor (Reavie 2007) transfer functions, multimetric indicators (Reavie et al. 2008, Brazner et al. 2007), diatom habitat characterization (Kireta et al. 2007) and monographs (Reavie & Kireta 2015).

Much of the previous work on Great Lakes diatoms was performed by Eugene Stoermer and several of his colleagues (e.g., Stoermer et al. 1999). Early in the development of our assessments we collaborated closely with Dr. Stoermer in an effort to better characterize the Great Lakes diatom flora. He long recognized that diatom taxonomy in the lakes lacked refinement, which has been a hindrance to accurate studies of primary producers in the Great Lakes (Stoermer 1978). In this work we aimed to clearly illustrate the monoraphid taxa observed during the GLEI assessment and, when possible, to summarize species autecology. We especially hope this work facilitates applying the correct diatom species names for ecologists studying the Great Lakes and other freshwater aquatic environments.

1.2. MATERIALS AND METHODS

The following preparation and analytical methods are closely paraphrased from Reavie & Kireta (2015), and we re-present them here for clarity. We acknowledge this repetition in case there are plagiarism concerns.

Figure 1. Coastal sample locations for diatom samples from the Great Lakes. Station numbers match those for each specimen photograph in the plates. Modified from Reavie & Kireta (2015)

Sampling: Field sites (Fig. 1) were sampled from June to September 2002 and May to August 2003. In addition to diatom samples, a suite of environmental measurements was collected at each sample location, and a detailed account of these parameters is provided by Reavie et al. (2006). Benthic and sedimented diatoms were sampled from natural substrates from 0.5 to 3 m depth. Additional surface sediment samples were collected from nearshore locations at a 30-m depth from the USEPA’s research vessel Lake Explorer. Surface sediments were sampled using a 6.5 cm diameter push corer and core tube. Sediments were extruded in the boat or on shore, and the top 1 cm of sediment was carefully removed using a spoon and/or spatula. In areas where coring was not feasible, a petite ponar sampler was used to collect unconsolidated bottom substrates, or rocks were carefully collected by hand. Approximately 1 cm of surface sediments from ponar samples was removed using a spoon and/or spatula. The surfaces of rocks and pebbles were scrubbed clean with a small brush or plastic knife and collected in vials as epilithic samples. All samples were iced at 4–6°C until processing. Approximately 75% of sites were cored, 13% required ponar grab samples, and 12% relied on epilithic samples collected by hand.

Sample Preparation and Analysis: In the lab, subsamples were taken from homogenized sediment samples and the diatom remains were cleaned using concentrated nitric or hydrochloric acid, or 30% hydrogen peroxide. Samples were digested in a water bath (85°C) for one hour. Samples were allowed to cool and settle at room temperature for 24 hours and then were centrifuged at 1800 RPM for 10 minutes. The tubes were aspirated, refilled with deionized water and shaken to break up the pellet. This centrifugation process was repeated five times. Four microscope slides were prepared for each sample using the Battarbee (1986) method. Diatom assessments for the GLEI project relied on light microscopy for timely data collection. For each sample, 400 diatom valves were counted along random transects at 1000 × magnification using oil immersion microscopy. Counts were made continuously along transects as wide as the field of view until sufficient valves were counted. Specimen photography did not follow a regimented protocol; digital photographs of diatom valves generally occurred as new taxa, or variations thereof, were encountered. Species were identified to the lowest taxonomic level possible using numerous diatom checklists, journal articles and iconographs. The most recently published nomenclature was used as much as possible for identifications.

Photographic sessions involving scanning and photographing specific genera also occurred in preparation for the many axonomic workshops held throughout the project period. Photographs were always collected at the highest magnification possible (1000 × and higher) using standard brightfield or employing differential interference contrast.

Preserved material and prepared microscope slides are currently stored in the Natural Resources Research Institute’s diatom collection at the University of Minnesota Duluth.

1.3. RESULTS, DISCUSSION AND TAXONOMY

Of the 93,368 diatom specimens observed during the GLEI project, approximately 2,200 diatom taxa encountered in our samples. From this, 273 monoraphid taxa were encountered and 79 of these taxa are described here. (We recognize that Rhoicosphenia is not considered a monoraphid genus, but we include it here due to its rudimentary raphe on one valve and possibility that it is a member of the Achnanthidiaceae; Thomas et al. 2016.) The majority of the observed taxa were rare, often single observations, and they are not included because suitably photogenic specimens could not be found. The 79 taxa we describe comprise over 98% of the monoraphid specimens observed.

1.4. SYSTEMATIC ACCOUNT

As for the methods (above), the following is closely paraphrased from Reavie & Kireta (2015), so we felt it was unnecessary to make major revisions to the text as presented.

Photographs are linked to taxonomic and autecological descriptions. In cases where specimens could not be reconciled with existing literature, a few approaches were followed. For species resembling known taxa, but differing slightly (e.g., slightly lower or higher striae density, slightly smaller or larger valve), we used a cf. qualifier. Sometimes we associated a question mark (?) with ambiguous photographs with uncertain affinity but seemed to have enough similarity to the described taxon for tentative consideration. When a species adequately matched morphological parameters with previously published accounts, the ranges of those parameters from the existing account are provided and cited.

Environmental Information for Taxa: Environmental characteristics were quantified for common taxa. Taxa were considered common if: they occurred in at least five samples with greater than 1% relative abundance in at least one of those samples; or they represented more than 5% relative abundance in at least one sample. Autecology for each of these common taxa was presented in a summary diagram. A histogram presents habitat affinity according to the relative frequency the taxon was encountered in each geomorphic habitat (high-energy [HE], embayment [EM], riverine wetland [RW], protected wetland [PW], coastal wetland [CW], open water nearshore [NS]; Kireta et al. 2007). This diagram is intended to depict whether taxa have habitat specificity in our samples, or if one should expect to encounter a taxon across a wide range of physical conditions. A second histogram illustrates the relative occurrence of each taxon in the five lakes (Superior [SU], Michigan [MI], Huron [HU], Erie [ER], Ontario [ON]; Kireta et al. 2007) incorporating standardized weighting by the number of samples collected in each lake.

Water quality optima for total phosphorus (TP) and chloride (Cl) were presented along vertical bars representing the measured water quality gradient for all GLEI samples. These autecological data for common taxa were based on statistical evaluations of species-environmental relationships (covered in greater detail by Reavie et al. [2006] and Kireta et al. [2007]). Species optima for these variables were estimated using weighted averaging regression and calibration, implemented by the rioja package (Juggins 2015) using the R statistical program. Diatom assemblages were related to water chemistry assuming unimodal species responses.

Two additional bars, stress power and stress rank, depict the relative ability of a taxon to track stress and whether that taxon reflects low or high stress. These results were achieved by evaluating relationships between each individual taxon and agricultural, industrial and urban development stressors quantified for each sample location. Briefly, the U.S. coastline of the Great Lakes was divided into 762 segments, each consisting of a shoreline reach and associated watershed (i.e., a segment-shed). Each segment-shed was summarized using 207 geographic information system (GIS)-based environmental variables that included anthropogenic activities (e.g., agricultural activities, urban density, industrial polluters) (Danz et al. 2005). For example, 26 agricultural variables (including pesticide runoff and leaching, cropland area, nitrogen and phosphorus exports, percent of county treated for various pests, and livestock inventories) comprised an agricultural category. Principal components analysis (PCA) within each category of environmental variation was used to reduce dimensionality and derive comprehensive gradients for agriculture, atmospheric deposition, point source pollution and urbanization. For each common taxon, species relative abundance data were regressed against the set of comprehensive watershed-level predictors using multiple linear regression and evaluated using the coefficient of determination (R2). The stress power value for a taxon represents its R2 position on the gradient of the R2 values for all common taxa. Taxa with higher values along this gradient had more acute relationships with watershed stressors, whether they reflected high or low stress, and so those taxa are assumed to be better indicators of condition. To derive stress rank, a canonical correspondence analysis was performed including the species assemblages and the comprehensive stressor variables. A primary gradient of stress, largely driven by agricultural activities, was derived by the distillation of these data, similar to that achieved by Reavie (2007). A taxon’s stress rank was taken as its score relative to that stressor gradient, standardized by the range of scores for all common taxa.

ACHNANTHES Bory (1822: 79)

Achnanthes cf. ingratiformis Lange-Bertalot (Pl. 1: 19)

Comments: This uncommon taxon fits the shape of Achnanthes ingratiformis Lange-Bertalot in Lange-Bertalot & Krammer (1989: 70), but specimens presented by other authors (e.g. Krammer and Lange-Bertalot 1991; Pl. 25) appear to have a lower striae density around the central area and somewhat less capitate ends. Striae in our specimens appeared slightly punctate using light microscopy. The presence of a central punctum (stigma?) in the raphid valve (Pl. 1: 19a) is not typical for monoraphid diatoms, so the taxonomy of this specimen remains uncertain.

Taxonomic notes (for A. ingratiformis): Length = 14–20 µm, width = 5–6 µm, striae = 23–25/10 µm (Krammer and Lange-Bertalot 1991).

Our specimen: Length = 16 µm, width = 5.5 µm, striae = 25/10 µm.

Achnanthes minuscula Hustedt (1945: 907) (Pl. 1: 65–74)

Comments: Our observations of this taxon match those by Hofmann et al. (2011) and Krammer and Lange-Bertalot (1991), although Great Lakes specimens sometimes had a lower striae density, making them look similar to Planothidium granum (Hohn & Hellerman) Lange-Bertalot (1999: 276) which typically does not have convergent striae. Some wider specimens (e.g. Pl. 1, Fig. 74a, b, c, d) had a valve shape like that of Achnanthidium exiguum (Grunow) Czarnecki (1994: 157), but otherwise valve features matched Achnanthes minuscula.

Taxonomic notes: Length = 7–9(10) µm, width = 3.5–4.5 µm, striae = (18)23–26/10 µm (Hofmann et al. 2011).

Figure 2. Geomorphic habitat distribution, lake specificity and environmental characteristics for A. minuscula Hustedt in the United States Great Lakes coastlines

Achnanthes sp. 100 GLEI (Pl. 1: 42)

Comments: This temporary name was used for rare, broadly-lanceolate, rapheless valves with some similarities to Achnanthes, Platessa and Psammothidium taxa, as well as Achnanthes duthii (Sreenivasa 1971: 81) as described by Kreis & Stoermer (1979 as Achnanthes) from the Great Lakes. Without observations of raphe valves, it was not possible to confirm taxonomy.

Taxonomic notes: Length = 11–12 µm, width = 5–6 µm, striae = 20/10 µm.

ACHNANTHIDIUM Kützing (1844: 75)

Achnanthidium affine (Grunow) Czarnecki (1994: 156) (Pl. 3: 77–83)

Comments: Members of the Achnanthidium minutissimum complex with coarsely-striate specimens with a bowtie-shaped central area on the raphid valve generally fit this taxon. This taxon likely incorporates valves of Achnanthidium kranzii (Lange-Bertalot) Round & Bukhtiyarova (1996: 350) (Hofmann et al. 2011), which is supposed to have a higher striae count but intergraded with affine.

Taxonomic notes: Length = 8–30 µm, width = (2.5)3.5–5 µm, striae = 22–30/10 µm (Hofmann et al. 2011).

Achnanthidium atomus (Hustedt) Monnier, Lange-Bertalot & Ector in Monnier et al. (2004: 130) (Pl. 5: 98, 99)

Comments: This taxon matches that described by Potapova (2010a). Although we captured only a few images of this taxon, it looked much like Achnanthidium cf. pyrenaicum (Hustedt) Kobayasi and was separated based on the interruption in the striae in the central area of the raphe valve.

Taxonomic notes: Length = 9.9–21 µm, width = 3–3.5 µm, striae = 28–30/10 µm (raphe valve) and 22–25/10 µm (rapheless valve) (Potapova 2010a).

Achnanthidium exiguum (Grunow) Czarnecki (1994: 157) (Pl. 1: 1–13)

Comments: This taxon was similar to images of Achnanthes in Krammer and Lange-Bertalot (1991; Pl. 23, Fig. 1–27) and Hofmann et al. (2011), although some Great Lakes taxa had finer striae measurements on the rapheless valve. This taxon is complicated by identifications of this form as Achnanthes exigua var. heterovalva Krasske (1923: 193) (Patrick & Reimer 1966, Camburn & Charles 2000) as well as the varieties heterovalvata and heterovalvum (see Potapova 2010b). Main distinguishing features are the distinct rectangular shape, irregular central area on both valves, visibly radiating striae on the raphe valve and strongly rostrate to slightly capitate ends.

Taxonomic notes: Length = 5–17 (20) µm, width = 4–8 (10) µm, raphe valve striae = 24–>30/10 µm, rapheless valve striae = 20–24/10 µm (Krammer & Lange-Bertalot 1991).

Figure 3. Geomorphic habitat distribution, lake specificity and environmental characteristics for A. exiguum (Grunow) Czarnecki in the United States Great Lakes coastlines

Achnanthidium exiguum fo. semiaperta Guermeur (1954: unknown) (Pl. 1: 14–18)

Comments: This taxon has been identified previously from the Great Lakes (Andresen et al. 2000; Kreis & Stoermer 1979). Specimens had a more inflated, oval outline, and narrower, more rostrate ends than the nominate form. These specimens were originally enumerated as A. exigua var. 1 or A. cf. exigua Grunow in Cleve & Grunow (1880: 21) during assessments.

Taxonomic notes: Length = (12)15–15 µm, width = (5)7–7 µm, raphe valve striae = (24)30– >30/10 µm, rapheless valve striae = (20)24–28/10 µm (Kreis & Stoermer 1979).

Achnanthidium minutissimum (Kützing) Czarnecki (1994: 157) (Pl. 3: 1–68)

Comments: A highly variable complex of this taxon and similar taxa occurred in our Great Lakes samples. Valve shape could be linear-lanceolate, rhombic and slightly rostrate, and it is likely several additional taxa are included in our depictions of A. minutissimum, minutissimum var. jackii, neomicrocephalum and affine, despite best efforts to split according to contemporary literature. This nominate taxon likely incorporates valves of Achnanthidium kranzii (Lange-Bertalot) Round & Bukhtiyarova (due to difficulty confirming curved terminal raphe fissures) and Achnanthidium eutrophilum (Lange-Bertalot) Lange-Bertalot (1999: 271) (slightly more rhombic valves, a shape that intergrades with the more typical linear-lanceolate form of A. minutissimum). Taxonomic confirmation can be difficult using light microscopy due to substantial intergrading within the complex.

Taxonomic notes: Length = 5–25 µm, width = (2)2.5–4 µm, striae = 26–30(35)/10 µm (Hofmann et al. 2011).

Figure 4. Geomorphic habitat distribution, lake specificity and environmental characteristics for A. minutissimum Kützing Czarnecki in the United States Great Lakes coastlines

Achnanthidium minutissimum var. jackii (Rabenhorst) Lange-Bertalot in Lange-Bertalot and Krammer (1989: 105) (Pl. 2: 63–96)

Comments: This variety tended to have rostrate to slightly capitate ends, but otherwise was difficult to discern from the nominate form and Achnanthidium neomicrocephalum Lange-Bertalot & Staab in Krammer & Lange-Bertalot (2004: 431), as illustrated by several transitional forms (Pl. 2: 58–61). Further, specimens with a coarser striae count may be Achnanthidium gracillimum (F.Meister) Lange-Bertalot in Krammer & Lange-Bertalot (2004: 430).

Taxonomic notes: Length = 5–25 µm, width = (2)2.5–4 µm, striae = 26–30(35)/10 µm (Hofmann et al. 2011).

Achnanthidium minutissimum var. 1 GLEI (Pl. 3: 69–75, 76?)

Comments: This variety was used to characterize finely-striate valves with a rectangular central area. This taxon has some similarity to Achnanthes silvahercynia Lange-Bertalot in Lange-Bertalot & Krammer (1989: 139) as shown by Krammer & Lange-Bertalot (1991), but our specimens tended to have a more bowtie-shaped central area. For environmental characterization this taxon was included with the nominate form.

Taxonomic notes: Length = 6.5–12 µm, width = 2–2.5 µm, striae = 30–35/10 µm (range of Great Lakes specimens).

Achnanthidium minutissimum var. 2 GLEI (Pl. 2: 97–109)

Comments: This variety was used to characterize especially narrow valves that were otherwise similar to the nominate form. The central areas of raphid specimens were rectangular to irregular. Distinction of this variety was sometimes arbitrary during sample assessments, so for environmental characterization this taxon was included with the nominate form.

Taxonomic notes: Length = 8.5–16 µm, width = 1.5–2.5 µm, striae = 30–40/10 µm (range of Great Lakes specimens).

Achnanthidium cruciformis Reavie sp. nov. (Pl. 5: 46–52)

Comments: This somewhat cruciform diatom was originally characterized as Achnanthes thermalis (Rabenhorst) Schönfeldt (1907: 122) (now Crenotia thermalis (Rabenhorst) Wojtal (2013: 81)), but striae densities were much higher than that described by Krammer & Lange-Bertalot (1991) and our specimens have a more inflated central region. Because of its similarity to C. thermalis, A. cruciformis may belong to the genus Crenotia, but additional observations of internal valve structures at higher resolution are needed. Temporarily we named this taxon A. minutissimum var. 3 GLEI, and based on its unique characteristics we now apply a new species name. The thickened central area is similarly widened, forming an open, cruciform shape around the raphe. The proximal raphe ends are slightly enlarged. The raphe is straight and under LM fades to being non-discernable near the valve ends. Striae are radial and more widely spaced (~25/10 µm) at the central region and become much finer (>30/10 µm), parallel and difficult to discern near the ends. One specimen (Pl. 5: 49) indicates that the valve is distinctly curved in girdle view. The valve ends are rounded and may be slightly capitate.

Taxonomic notes: Length = 14.5–17.5 µm, width = 3–4.5 µm, striae = 24–31(35)/10 µm (range of Great Lakes specimens).

Figure 5. Geomorphic habitat distribution, lake specificity and environmental characteristics for A. cruciformis Reavie in the United States Great Lakes coastlines

Achnanthidium cf. neomicrocephalum Lange-Bertalot & Staab (Pl. 2: 46–56, 110–112)

Comments: Although this taxon is characterized by capitate ends, intergrading with the more rostrate Achnanthidium minutissimum var. jackii makes assured identification difficult. Valve shape matches well with Achnanthidium neomicrocephalum Lange-Bertalot & Staab in Krammer & Lange-Bertalot (2004) as described by Hofmann et al. (2011), but valves in the Great Lakes tended to be shorter. A single specimen (Pl. 2: 57) has superficial resemblance to A. cf. neomicrocephalum, but the inflated center and bowtie-shaped central area surrounded by shortened striae suggest a different taxon.

Taxonomic notes (for A. neomicrocephalum): Length = (11)22–38 µm, width = (1.5)1.8–2.8 µm, striae = (25)27–33/10 µm (Hofmann et al. 2011).

Our specimens: Length = 11.5–12.5 µm, width = 2.5 µm, striae = 35/10 µm.

Figure 6. Geomorphic habitat distribution, lake specificity and environmental characteristics for A. cf. neomicrocephalum in the United States Great Lakes coastlines

Achnanthidium cf. pyrenaicum (Hustedt) Kobayasi (Pl. 5: 100–114)

Comments: This taxon largely follows Achnanthidium pyrenaicum (Hustedt) Kobayasi (1997: 148) per Hofmann et al. (2011). While having finer striae counts, it was often difficult to distinguish from Rossithidium petersenii due to ambiguity in the structure of the central area. As for several other small Achnanthidium, Rossithidium and Psammothidium taxa, this species occurred among a complex of similar-looking taxa, so we recommend further work to confirm whether these are environmental forms and/or varieties as opposed to unique species.

Taxonomic notes (for A. pyrenaicum): Length = 6–35 µm, width = (2.5)3–6 µm, striae = 20–40/10 µm (raphe valve) and 20–38/10 µm (rapheless valve) (Hofmann et al. 2011).

Our specimens: Length = 5.5–13.5 µm, width = 2.5–3.5 µm, striae = 30–35/10 µm (raphe valve) and 30–40/10 µm (rapheless valve).

Achnanthidium rivulare Potapova & Ponader (2004: 36) (Pl. 4: 1–33)

Comments: Despite significant splitting of these small achnanthoid taxa, several diagnostic features were not easily distinguishable under light microscopy. It is highly likely this taxon includes specimens of Achnanthidium linearioides (Lange-Bertalot) Lange-Bertalot in Lange-Bertalot & Moser (1994: 95) (Hofmann et al. 2011; formerly Achnanthes linearis (W.Smith) Grunow in Cleve & Grunow (1880: 23), but we observed many specimens shorter than the apparent lower limit of 10 µm) and A. pyrenaicum (Hustedt) Kobayasi (1997: 148) (Hofmann et al. 2011; incorporating Achnanthes biasolettiana Grunow in Cleve and Grunow (1880: 22); drawn-out ends and curved terminal raphe fissures characteristic of the species were not easily confirmed). Smaller, narrow specimens often intergraded with A. minutissimum (Kützing) Czarnecki (1994: 157) as features became more difficult to discern. Shorter valves lacked the characteristic tapering ends, so rapheless valves looked like Rossithidium petersenii (Hustedt) Round and Bukhtiyarova (1996: 178), which was generally distinguishable by having a higher striae density around the central area.

Taxonomic notes: Length = 5.4–23.8 µm, width = 2.6–4.4 µm, striae = 19–28/10 µm, and as high as 55/10 µm at the ends (Potapova 2009a).

Figure 7. Geomorphic habitat distribution, lake specificity and environmental characteristics for A. rivulare Potapova & Ponader in the United States Great Lakes coastlines

Achnanthidium rosenstockii (Lange-Bertalot) Lange-Bertalot in Krammer & Lange-Bertalot (2004: 433) (Pl. 5: 1–43)

Comments: We follow Hofmann et al.’s (2011) depiction of this abundant taxon, although it also appears to occur under the genus Psammothidium (Bukhtiyarova & Round 1996). Based on our comparison, this taxon was previously observed by Kreis & Stoermer (1979) in the Great Lakes, but they identified it as Achnanthes biasolettiana (GLRD) (Kützing) Grunow, prior to naming of this trinodal form by Lange-Bertalot & Krammer (1989; as Achnanthes rosenstockii Lange-Bertalot).

Taxonomic notes: Length = 6–14 µm, width = 3–4.5 µm, striae = (25)27–32/10 µm (Hofmann et al. 2011).

Figure 8. Geomorphic habitat distribution, lake specificity and environmental characteristics for A. rosenstockii (Lange-Bertalot) Lange-Bertalot in Krammer & Lange-Bertalot in the United States Great Lakes coastlines

PLATESSA Lange-Bertalot (2004: 442)

Platessa conspicua (Mayer) Lange-Bertalot in Krammer & Lange-Bertalot (2004: 445) (Pl. 2: 1–35)

Comments: The recent depiction of this taxon by Hofmann et al. (2011) generally matches Great Lakes specimens, although variations in valve dimensions and striae counts were greater in the Great Lakes. We observed valves as small as 4 µm in length and as narrow as 3 µm, which is smaller than Great Lakes specimens previously observed (Kreis & Stoermer 1979, as Achnanthes). Some specimens also had higher striae densities. This taxon is usually identifiable by the two middle striae that are more distant from each other. While others claim a lanceolate central area is characteristic of the rapheless valve of this species (Potapova 2010c) this was not always observed, especially in smaller specimens.

Taxonomic notes: Length = (4)7–20 µm, width = (3)4–7.5 µm, striae = 11–16(20)/10 µm (Hofmann et al. 2011).

Figure 9. Geomorphic habitat distribution, lake specificity and environmental characteristics for P. conspicua (Mayer) Lange-Bertalot in Krammer & Lange-Bertalot in the United States Great Lakes coastlines

Platessa conspicua var. 104 GLEI (Pl. 2: 36–45)

Comments: With no gap in the striae at the central area it was difficult to confirm taxonomy of some of these valves, so we used this tentative identification to capture uncertain specimens that were otherwise similar to P. conspicua. Without observations of the rapheless valve, some of these raphid specimens may be Planothidium frequentissimum (Lange-Bertalot) Lange-Bertalot (1999: 282), Platessa bahlsii Potapova (2012: 1) or small, ovoid valves of Planothidium granum. Confirmation from other specimens in a given sample may be needed to determine taxonomy of similar specimens.

Taxonomic notes: Length = 4.5–9 µm, width = 3–4 µm, striae = 14–20/10 µm (range of Great Lakes specimens).

Platessa ziegleri (Lange-Bertalot) Lange-Bertalot in Krammer & Lange-Bertalot (2004: 445) (Pl. 1: 20–41)

Comments: Our specimens compare well with those presented by Krammer and Lange-Bertalot (1991, as Achnanthes) and Hofmann et al. (2011). Distinguishing characteristics appear to be a gap in the striae in the central area (often asymmetrical) and central expansion of the axial area on the rapheless valve. Measured breadths in Great Lakes specimens could be smaller than that listed by Krammer and Lange-Bertalot (1991), and striae counts were sometimes finer. Rostrate ends sometimes gave way to a more lanceolate valve shape.

Taxonomic notes: Length = (7)8–14 µm, width = (5)6–8 µm, striae = 20–22(28)/10 µm (Krammer and Lange-Bertalot 1991).

Figure 10. Geomorphic habitat distribution, lake specificity and environmental characteristics for P. ziegleri (Lange-Bertalot) Lange-Bertalot in Krammer & Lange-Bertalot in the United States Great Lakes coastlines

PSAMMOTHIDIUM Bukhtiyarova & Round (1996: 3)

Psammothidium bioretii (Germain) Bukhtiyarova & Round (1996: 9) (Pl. 11: 1–23)

Comments: Great Lakes specimens largely followed Hofmann et al. (2011) and Kreis & Stoermer (1979, as Achnanthes), although valve lengths were always between 10 and 15 µm. The valve twist that is characteristic of Eucocconeis alpestris became less apparent in smaller specimens, so valves of these two taxa sometimes looked similar. Although an oblique axial area was often present on