Typification of the puzzling diatom species Neidium iridis (Neidiaceae)

Background and aims – Confusion over the taxonomic identity of Neidium iridis (Ehrenb.) Cleve (= Navicula iridis Ehrenb.) has persisted for more than 130 years with identifications of valve shapes varying from elliptical to linear and sizes from 100 to 300 μm. This confusion can be linked to the rarity of the species and poor attention to the original line drawings of Ehrenberg. This study examines the type material of Navicula iridis and further uses additional material from North America to define the species Neidium iridis. Methods – Mica and material from Christian Gottfried Ehrenberg’s Collection were studied to examine Navicula iridis. Light and electron microscope techniques were used to document the variability of valve morphology of original material in comparison with other fossil and recent materials. Key results – Ehrenberg designated no nomenclatural type for Navicula iridis at the place of its first description, and until recently the original material was not studied. One freshwater extant locality was identified (New York, West Point) in the original 1843 publication, and in his notes and drawings. Examination of the micas and material from the type locality, showed that Navicula iridis was very rare. Two specimens were found that matched the line drawing presented by Ehrenberg. In addition, we examined Ehrenberg’s original material with SEM, as well as two other localities in North America and present LM and SEM documentation of additional valve forms. A further evaluation of the status of Neidium columnaris, Neidium maximum and an unknown taxon in relation to Neidium iridis is presented. Conclusions – The findings of this study elucidate the taxon Navicula iridis (= Neidium iridis) as described by Ehrenberg in 1843. This will assist in the identification of other large Neidium taxa which to date have been associated with Neidium iridis.


INTRODUCTION
The genus Neidium Pfitzer (1871: 39) represents a group of benthic freshwater diatoms with > 300 taxa. Some of the original species placed in the genus were identified by Ehrenberg (1843) from North America (Hamilton & Jahn 2005). More recently a number of the species described by Ehrenberg and Grunow have been genetically defined (Lefebvre & Hamilton 2015, Neidium amphigomphus (Ehrenb.) Pfitzer, Neidium dilatatum (Ehrenb.) Cleve, Neidium tume-scens Grunow, Neidium hitchcockii (Ehrenb.) Cleve), but Neidium iridis (Eherenb.) Cleve (= Navicula iridis Ehrenb.) still remained an unknown entity. Ehrenberg first described Navicula iridis in his paper "das mikroskopische Leben in Süd-und Nord-Amerika" (Ehrenberg 1843: 382, 418) with a Latin description, location (New York, West Point) and two line drawings (pl. IV, fig. I.2a, b) of specimens. The drawings clearly show the shape, large size, small central area and number of canals (longitudinal lines) along the margin of the valve. Neidium iridis s. lat. currently contains 53 in-fraspecific taxa, 34 varieties and 19 formae. Although little is known about Ehrenberg's Neidium iridis s. str. the erroneous concept of the species presented today is based on large elliptic valves with multiple marginal canals (e.g. Grunow in Van Heurck 1880-1885, Patrick & Reimer 1966, Krammer & Lange-Bertalot 1986. Indeed, subsequent to the original publication 10 authors up to the year 1966 presented line drawings of their specimens representing what they considered to be Neidium iridis; none of these matched the original concept of Ehrenberg (1843: 418, pl. IV, 2a, b). The wide diversity in valve morphology amongst the infraspecific taxa of Neidium iridis illustrates the identification problem of the nominate form and the broad concept that present authors maintain.
Researchers have looked through the New York mica's (West Point) for Navicula iridis (e.g. Lange-Bertalot, Senkkenberg Institute, Germany, pers. comm.) but were not able to find a specimen of Navicula iridis. Katarzyna Stachura-Suchoples and Wolf-Henning Kusber were able to find two specimens from the New York (West Point) micas matching the line drawing of Ehrenberg. In addition, original material was found. The objective of this study is to circumscribe and emend the species description of Neidium iridis using original and current samples from North America. Comparisons are also made with an unknown taxon, and two other closely related taxa Neidium columnaris (Ehrenb.) Mills and Neidium maximum (Cleve) F.Meister which have not been typified.

MATERIAL AND METHODS
The sample which contains original material of Navicula iridis is deposited in the Ehrenberg Collection, Institute of Paleontology, Museum für Naturkunde, Berlin (BHUPM). The following material was studied: 1. Reference: Ehrenberg (1843: 382, pl Canada) were examined. The specific location for the type of Navicula iridis (e.g. New York, West-Point) is not known, therefore materials with Neidium iridis from the collections of ANSP and CANA were used to further valid morphology, distribution and in the future genetic signature of this taxon in North America. The Canadian material was acid cleaned and washed with distilled water to remove the acid. The cleaned liquid samples were subsampled and dried on coverglass and mounted onto microscope slides with Naphrax. Additional subsamples were dried onto aluminium foil and mounted onto SEM stubs following the protocol of Siver (1987). SEM were examined uncoated or with a thin gold palladium coat; the coating was completed using a Denton Vacuum DESK II sputter coater. LM studies were conducted using a Leica DMR HC microscope with Phase, DIC and Brightfield optics: 40x: Plan Apo HCX 40/1.25 and 100x: Plan Apo HCX 100/1.35 objectives. Samples for SEM study were observed with an FEI XL30 (ESEM), or a FEI APREO (FEG) using 2-10 KV with a working Distance < 10 mm under high vacuum. Terminology used to describe morphological structures follows von Stosch (1975), Ross et al. (1979), Krammer & Lange-Bertalot (1986) and Siver et al. (2003). For old Neidium names and concepts VanLandingham (1978) was consulted.

Observations on Ehrenberg's notes, drawings and mica
In 1843 Ehrenberg wrote: "Ausser diesen fossilen, dicht unter der Oberfläche eines Torfmoors vorkommenden, mithin vielleicht sämmtlich auch der Jetztwelt angehörigen Formen habe ich nur noch Gelegenheit gehabt, von West-Point auch noch eine grosse Reihe von entschieden jetzt lebenden und sogar im lebenden Zustande in Berlin zu beobachten. Herr Prof. Bailey sandte mir in Jahre 1842 einige Gläschen voll Torf-Wasser aus West-Point mit vielen lebenden Bacillarien. Diese waren am 2. April 1842 dort mit dem Wasser gefüllt   worden, und am 16. Juni darauf konnte ich sie der Akademie in Berlin noch zahlreich lebend vorzeigen. (1) Ich habe mich bemüht, diese lebendigen oder doch sämmtlich als Jetztwelt mit Sicherheit angehörigen Formen Nord-Amerikas mit den europäischen möglichst genau zu vergleichen und habe deshalb alle gezeichnet." Translation, "Apart from these fossil forms, which occur just below the surface of a peat bog, and perhaps all of which belong to the world of the present, I have had occasion to observe from West Point only a great number of resolutely living and in still living conditions in Berlin. In the year 1842, Prof. Bailey sent me from West Point some jars full of peat water with many living Bacillaria. These had been filled with water on April 2, 1842, and on June 16, I was able to show them to the Academy in Berlin still numerously alive. I have endeavoured to compare as closely as possible all these forms of North America which are alive or certainly belong to the world of the present with the European ones, and have therefore drawn all of them." In total 60 micas identified as from West Point (New York) are present in the Ehrenberg Collection. Seven of these micas have been displaced from the original strips and their identification position on the strips is therefore uncertain. Four of the 60 micas did not have (or have lost) circles marking identified specimens. Twenty-one micas had one circle identifying specimens, 17 had two, 16 had three and three had four circles marking on the micas. For this study, mica strips 260109 and 260110 ( fig. 1), as noted above the specimens on the drawing sheet ( fig. 2A & B), were examined.
The Latin description for Navicula iridis reads "testula magna elongate bacillaris, lateribus planis, apicibus leviter attenuates obtusis, superficie subtilissime transverse et longitude. Lineolate iridis coloerm emittente. Icon! An sui generis forma?" In this description Ehrenberg (1843: 382) notes the large elongate frustule with obtuse tapering apices and that the valve surface is composed of fine transverse lines (striae) which show a rainbow of colour under the microscope. Finally, he notes, "Icon! Or is the form of its own kind(?)"; this indicates that Ehrenberg recognized that this was likely not a Navicula but a species which belonged to another genus. The genus Neidium (as typified by Hamilton & Jahn 2005) was not erected until 1871 (Pfitzer 1871) and Neidium iridis was later transferred to Neidium by Cleve (1894: 69).
The original description presents only a few characters to ascribe to the species. However, Ehrenberg had an attention for detail in his drawings, which present additional character clues for the species. Ehrenberg (1843: pl. IV, fig. 2a, b, also on drawing sheet no 2074) presents a large linear diatom with cuneate apices ( fig. 2). Two black lines document the narrow axial area with a central area extending approximately 1/3 across the valve. Two small half circle lines within the central area document a central nodule. The striae are presented as linear (not radiate) with individual dots suggesting that are-olae could be observed within the striae. The original drawing of the valve (specimen on mica 261109) has two lines (widely spaced) running along each margin indicating one or two longitudinal canals. A circular solid spot is present at both apices interrupting the parallel lines along the margin. In girdle the frustule (specimen on mica 261110 [a, blue circle]) is rectangular with clearly punctate striae. Three-four lines extend along the margins and recessed indentations are present on both valves in the mid-region. Line drawn (indentations/spots) are also present at each apex. In both line drawings residual pigments are presented indicating the species was extant as indicated in Ehrenberg's original text.

DISCUSSION
Specimens of Navicula iridis were not found (at least by P.B. Hamilton) on the West Point mica preparations labelled with Navicula iridis in the associated text ( fig. 1). K. Stachura-Suchoples and W.-H. Kusber, after an extensive search through the micas, found two unobstructed specimens of Navicula iridis (strip 260109, mica d); they were found on those mica strips which Ehrenberg had mentioned but somewhat outside the colored ring. There are no handwritten notes underneath the mica (see fig. 1), which explains the extended time to find this taxon. It is now evident that Navicula iridis (Neidium iridis) was accurately and effectively presented in the line drawings of Ehrenberg. This is a rare species which to date has only been observed in a few samples both fossil and extant from North America. At the type locality, Neidium iridis is associated with centric diatoms, and the diatoms Pinnularia spp. and Tabellaria sp., as well as with typical moorland desmids such as Cosmarium spp. and Euastrum spp. It is easily identified by the large linear valves with one large longitudinal canal along each margin and extended rostrate to sagittate apices. This taxon also has renilimbi (figs 4D & 7F), small oppositely deflected central raphe ends and a lacinia (bifurcate terminal raphe ends, fig. 3B) which places this taxon morphologically well within the genus. Neidium iridis is currently the longest Neidium species, although there are some large unidentified taxa with a similar size and shape. These unidentified taxa from the Americas have multiple longitudinal canals, do not have well developed pseudosepta, and do not have sagittate apices (e.g. Metzeltin & Lange-Bertalot 2007: pl. 170, figs 3-5;pers. obs. from North America). In the future, with better documentation, we anticipate that a few more large Neidium species will be distinguished.
lat. Their specimen is smaller with sagittate apices similar to larger Neidium iridis valves.
The original line drawing and description of Navicula firma Kütz. (Kützing 844: 82,pl and Neidium indicum f. undulatum "undulata" Gonzalves & H.P.Gandhi. These smaller taxa are distributed around the world suggesting that the sagittate shape-form is not biogeographically restricted. The relationship among these taxa (genetic and morphological if any) is yet to be determined.

CONCLUSION
Neidium iridis is an extant species which to date has primarily been observed s. str. from North America. Reports from Japan suggest that this taxon extends beyond North America; however, to date no positive identifications have been recorded for South America (Metzeltin & Lange-Bertalot 2007) and Europe (e.g. Krammer & Lange-Bertalot 1986). The valves are large linear, to weakly linear-elliptic with rostrate to sagittate apices. One large longitudinal canal is present along the valve face-mantle margin. The termination of the longitudinal canals do not extent to the apex, thereby forming a circular-type cavity at the apex which is easily seen in LM. Externally, the central raphe endings are small deflected hooks and the terminal ends form a lacinia. Areolae can be covered by a dense radiating cribra. Internally, renilimbi form around selected areolae, especially along the longitudinal canals and the axial area. The internal central helictoglossae fuse to form a single line projection and the terminal helictoglossae are bent away from the internal valve face. The extensive documentation of species (line drawings and marked specimens) by Ehrenberg and his daughter Clara has given us the opportunity to find, document and further emend the species Neidium iridis using finer examination tools.