Navicula lothargeitleri Poulí
publication ID |
https://doi.org/ 10.11646/phytotaxa.273.1.2 |
persistent identifier |
https://treatment.plazi.org/id/B372C228-FFEF-315C-BCE1-F833FC8DFB3B |
treatment provided by |
Felipe |
scientific name |
Navicula lothargeitleri Poulí |
status |
|
Navicula lothargeitleri Poulí č ková & E.J. Cox, sp. nov. ( Figs 1 View FIGURE 1 , 2 View FIGURE 2 )
Valves lanceolate, 23.1–46.0 μm long, 5.2–8.5 μm wide, with slightly rostrate apices, without a clear point of inflexion between the main part of the valve and the apices. Striae are fine, 16–18 in 10 μm, radiate in the central part, but becoming transverse at the apices. The central area is broad, more or less rounded, wider than long, extending over half the valve width ( Figs 1L–P View FIGURE 1 ). Live cells have two plate-like chloroplasts that lie along each side of the girdle during interphase ( Fig. 1J View FIGURE 1 ). Long bar-like pyrenoids (one per chloroplast) are present ( Fig. 1K View FIGURE 1 ), becoming visible when the chloroplasts move from the girdle to the valves prior to mitosis. The interphase nucleus lies in a central cytoplasmic bridge between two vacuoles. DAPI staining reveals a central mass of heterochromatin that occupies well over half the diameter of the nucleus, and one or two peripheral nucleoli (Fig lI). SEM morphology: Valve structure is typical of Navicula sensu stricto, with striae comprised of lineate areolae, occluded internally by hymenes. The external raphe fissures are hooked at the apices, slightly expanded and deflected at the centre. The axial area is defined externally by slight thickening of the valve, about 0.5 μm wide (compared to about 0.7 μm wide in N. cryptocephala sensu stricto). Internally the virgae are narrower than the intervening striae, the central nodule is markedly wider than the rest of the raphe rib and the internal raphe fissures are deflected to the side of the raphe rib at the edge of the central area, in contrast to N. cryptocephala sensu stricto in which virgae are as wide as the striae, the central nodule is hardly widened and the raphe fissures open more centrally in the raphe rib ( Figs 2 G–K View FIGURE 2 ).
Type: ― SCOTLAND, U.K. Loch Lubnaig , 56° 16’ N, 4° 17’ W, 123 m a.s.l., D. G. Mann & A. Poulíčková, 29 September 2005 (holotype: OL! 100005, Nav 27L, isollated/subcultured on 23. November 2005, here illustrated as Fig. 4O View FIGURE 4 . Paratypes: voucher slides OL! 100004 and OL! 100006-7 from Nav 647K, Kew Billabong, Australia, 17 April 2008; Nav29L from Loch Lubnaig, GoogleMaps UK, 11 April 2008; and NavO/26 from Obectov, Czech Republic, 23 January 2008) GoogleMaps .
Habitat:―lake/pond epipelon.
Etymology: ― Navicula lothargeitleri is named in honour of Prof. Lothar Geitler from Vienna University, being the first who recognized the different “Rassen” in N. cryptocephala .
Ecology and distribution:―Because the species was previously reported as N. cryptocephala , apart from our strains ( Poulíčková & Mann 2006, Veselá et al. 2009, Poulíčková et al. 2010), the ecology and distribution of N. lothargeitleri sp. nov. is not known. All our strains were isolated from the epipelon of lakes or ponds. Strains belonging to N. cryptocephala sensu stricto were encountered more often than N. lothargeitleri , and over a wide spectrum of ecological conditions, except eutrophic, productive fish ponds ( Hašler et al. 2008). Sixteen strains have been isolated from eight sites: Royal Botanic Garden Pond, Scotland, UK; Lunz Untersee, Austria (Lothar Geitler’s locality); Horní Ves, Záhlinice, Hradčanský pond, Líšnice and Obectov, the Czech Republic; and Kew Billabong, Australia. Navicula lothargeitleri seems to occur less frequently. We isolated only seven strains: five from Loch Lubnaig, Scotland, UK; one from Obectov pond, the Czech Republic; and one from Kew Billabong, Australia. Loch Lubnaig is a deep glacial lake of low pH, with abundant N. cryptocephala -like cells within the epipelic sample (40% abundance). Obectov pond is shallow with neutral pH and low abundance of N. cryptocephala -like cells within the epipelic sample (0.2%). The taxa coexist sympatrically at Obectov and Kew Billabong.
Comparisons with similar taxa: ― Navicula cryptocephala has previously been found to be genetically (LSU rDNA, ITS of the rRNA operon), cytologically (interphase nucleus structure) and morphologically heterogeneous ( Poulíčková et al. 2010). In this study, we compared four strains of each clade (clade I “RBG” and clade II “Lubnaig”) with the type specimen of N. cryptocephala . Both clades differ slightly in their outline and their central areas. Clade II cells ( Figs 1M–P View FIGURE 1 ) taper more sharply to slightly broader, slightly rostrate apices than clade I cells ( Figs 1B–E View FIGURE 1 ), which taper gently to bluntly rounded apices. Clade II cells have transversely broader, rather than more or less circular central areas (clade I) and the central striae are more parallel in the former than the latter, in which shorter striae are sometimes inserted opposite the central area. However there is considerable overlap in the morphological features of both clades, as illustrated by the relative warps analysis of geometric morphometric data ( Figure 3 View FIGURE 3 ). On the ordination plot summing 89.9% of the total shape variation (RW1: 61.1%, RW2: 28.8%) we can see that cells belonging to the different clades overlapped in their morphological characteristics. It can also be seen that the lectotype of traditional N. cryptocephala falls closer to the centroid of clade I, the point that indicates the average morphology of that group. Discriminant analysis of the relative warps scores corroborated the pattern shown by RWA. The groups overlapped and their optimal discriminant axis correctly classified 85.3% of the cells ( Figure 4 View FIGURE 4 ). The Hotelling’s T 2 test indicated that such a pattern was unlikely to be a result of random differences between groups belonging to a single, morphologically homogenous complex (T 2 = 320.1, F = 2.21, p <10 -5). The same result was obtained from the two-group permutation test based on Euclidean distances between group centroids (d = 0.158, p <10 -3). The leave-one-out cross-validation procedure showed that the morphological characteristics of the lectotype place it unambiguously in clade I (DA score of the lectotype valve: 56.7).
The multivariate regression of shape data on the discriminant axis explained 8.88% of the morphological variation. The regression model was highly significant, with Wilk’s λ = 0.043, F s = 196.3, p <10 -5, permutation p = <10 -3. The regression model illustrated that valves with distinctly pronounced apices were typical for members of clade I, including the lectotype of N. cryptocephala ( Fig. 5A View FIGURE 5 ), while clade II was characterised by more broadly rostrate, less subcapitate, apices and broader central areas ( Fig. 5B View FIGURE 5 ).
Other valves on the type slide showed broad variation in morphology (not illustrated), thus the possibility that both morphs are present on the type slide could not be rejected. That is why we based our decision (which clade corresponds to the type) on a comparison with the designated lectotype ( Cox 1995). The same specimen from permanent slide BM 18785 is shown in Fig. 4F View FIGURE 4 and corresponds to clade I “RBG”. This clade therefore represents N. cryptocephala sensu stricto.
Ultrastructurally the two clades are predictably similar, although there are some subtle differences. Externally the central raphe endings are finer and more crozier-like in N. cryptocephala ( Fig. 2C View FIGURE 2 ) than in N. lothargeitleri sp. nov. ( Fig. 2H View FIGURE 2 ), in which they are wider, more or less hooked, while the external slightly thickened axial area is wider in the former ( Fig. 2C View FIGURE 2 ) than the latter ( Fig. 2H View FIGURE 2 ). Internally, the central nodule of N. cryptocephala is hardly wider than the raphe rib ( Fig. 2D View FIGURE 2 ) and the raphe fissures open closer to the centre of the raphe rib than in N. lothargeitleri , which has a markedly widened central nodule ( Fig. 2I View FIGURE 2 ), the raphe fissures becoming lateral just beyond the central area. Although the relative widths of the striae and the intervening solid virgae are more or less the same externally, internal views reveal that the virgae of N. cryptocephala are as wide as the areolae are long, whereas in N. lothargeitleri the virgae are narrower than the areolae are long. Areola spacing along the striae is similar in both taxa, but there are more pores along the most apical striae in N. lothargeitleri because its apices are slightly broader than those of N. cryptocephala ( Figs 5E, J View FIGURE 5 ). Separation of the two taxa with LM is more difficult, although careful focussing should reveal the wider central nodule of N. lothargeitleri and the difference in axial area width between the two. Typically, the central area of N. lothargeitleri valves comprises at least 50% of the maximum valve width. Conversely, N. cryptocephala valves typically have smaller central areas that are usually less than 50% of the maximum valve width ( Figs 1 View FIGURE 1 , 5 View FIGURE 5 ).
Although the clades are difficult to separate on the basis of valve morphology and ultrastructure, they can be readily separated on the basis of their interphase nucleus structure. Cells of N. cryptocephala sensu stricto (Clade I “RBG”) have two densely staining peripheral plaques of heterochromatin, lying opposite each other, one adjacent to each chloroplast. Nucleoli are not obvious ( Figure 1A View FIGURE 1 ). On the other hand, clade II “Lubnaig” cells have a central mass of heterochromatin that occupies more than half the diameter of the nucleus plus one or two peripheral nucleoli ( Fig. 1I View FIGURE 1 ). As this difference is stable and the genetic difference is also sufficient we have decided to recognise clade II “Lubnaig” as a new species, Navicula lothargeitleri sp. nov.
No known copyright restrictions apply. See Agosti, D., Egloff, W., 2009. Taxonomic information exchange and copyright: the Plazi approach. BMC Research Notes 2009, 2:53 for further explanation.
Kingdom |
|
Phylum |
|
Class |
|
Order |
|
Family |
|
Genus |
Kingdom |
|
Phylum |
|
Class |
|
Order |
|
Family |