Discussion

Gomphonema juettnerii resembles Gomphonema pumilum var. elegans Reichardt et Lange-Bertalot in general shape (Reichardt 1997). It differs, however by less radiate striae, ultrastructure of interstriae trench and external areola structure; areolae of G. pumilum v. elegans show well developed “C” shape across the valve (Figs 2, 3, Tafel I in Reichardt 1997), whereas G. juettnerii
shows “C”-shaped structure within the central area, and areolae become reduced to comma shape towards the poles. The central area of G. pumilum v. elegans is formed by shortened striae (Figs. 1-11, Tafel II in Reichardt 1997), which are absent in G. juettnerii. In LM, G. juettnerii.

Figs 23 – 25. Scanning electron micrographs of Gomphonema doonensis. 23. External view of whole valve showing double rows of areolae across the valve and undulate raphe. 24. Internal view of the whole valve showing striae pattern and central area. 25. External girdle view showing the areolae opening, apical pore fields and girdle bands. Scale bar represents 1 μm.

resembles G. minutum Agardh in general valve outline. However, these species differ from G. juettnerii by absence of the central area created by shortened striae. Furthermore, G. minutum has striae composed of two rows of areolae (Fig. 91 in Kociolek & Kingston 1999). Striae pattern and absence of stigmata in Gomphonema juettnerii differentiate it from G. sinestigma Reichardt, Jüttner & Cox (Jüttner et al. 2004). Gomphonema doonensis resembles some of the tropical species of Gomphonema, including G. eurycephalus Spaulding & Kociolek (Spaulding & Kociolek 1998) described from Madagascar and G. latistigmata Kociolek, Passy & Lowe (Passy et al. 1997) described from South Africa. Gomphonema doonensis has narrower valves and the presence of shortened striae in the central area and these features distinguish it from G. eurycephalus. Gomphonema doonensis dif

Figs 26 – 31. Scanning electron micrographs of Gomphonema doonensis. 26. External view of the headpole with the external distal raphe end curving onto the valve mantle. 27. External view of the central area with undulate raphe has dilated proximalends and round external stigmal opening. 28. External view of the footpole showing distal raphe end and bilobed apical pore field. 29. Internal view of the headpole showing striae found in deep troughs, separated by thick interstriae. 30. Internal view of the raised central area showing ecurved proximal raphe ends and a slit like internal stigmal opening. 31. Internal view of the footpole showing the radiate striae and helictoglossae and pseudoseptum. Scale bar represents 1 μm.

fers from G. latistigmata in valve outline and length: the former is generally shorter and clavate, with broadly rounded headpole. There is considerable confusion as to whether presence of doubly punctate striae alone is both necessary and sufficient to assign gomphonemoid diatoms to the genus Gomphoneis Cleve. Doubly punctate striae were first identified by Hustedt (1942, 1945) in several new species of Gomphonema. Hustedt noted (1942, p.121) that there were two features that diagnose Gomphoneis, namely doubly-punctate striae and the presence of longitudinal lines. The latter character was not mentioned by Dawson (1982), who documented doubly-punctate striae in the common diatom Gomphonema olivaceum Hornemann using transmission and scanning electron microscopy. For this reason, Dawson (1982) subsequently transferred G. olivaceum to Gomphoneis. Patrick & Reimer(1975, p.147) further complicated the situation by stating “This genus [Gomphoneis] is distinguished by the double rows of decussating puncta alternating with the costae, and the narrow septum near the margins of the valve.” Confusion over the taxonomy of this group was expressed by Lange-Bertalot (1980). In his identification of a large, doubly-punctate Gomphonema species from Israel (as “Gomphoneis herculeana (Ehrenberg) Cleve”, a species known only from North America!) and his argument was that doubly-punctate striae could not diagnose Gomphoneis.

Kociolek & Stoermer (1989) showed in the context of a formal phylogenetic analysis that Gomphoneis is diagnosed by the presence of doubly-punctate striae and the presence of a marginal lamina that gives the impression of the longitudinal lines in primitive members of the genus. Furthermore, Kociolek & Stoermer (1989) documented that two groups may be recognized: the Elegans group (that is the name bearer for the genus, and whose species also either have no stigmata or 4 stigmoids, as well as apical pore fields that are undifferentiated from the areolae) and the Herculeana subgroup (more commonly referenced when discussions of the genus are made). Krammer & Lange-Bertalot (1991) echoed Lange-Bertalot’s (1980) mistake when they suggested that some species included the Elegans group of Gomphoneis found in Lake Baikal were Gomphonema taxa, and made nomenclatural changes reflecting that opinion. See Kociolek et al. (2014) about the diversity and high levels of endemism of Gomphoneis species in Lake Baikal. More recently Reichardt (2007) argued on similar lines, while recognizing some gomphonemoid species with doubly-punctate striae and assigning them all to Gomphonema, without considering other features. To be clear, as Kociolek (1998) noted, we should refrain from thinking typologically about how species are included into groups (such as genera), and we have to recognize that there may be the loss or gain of evolutionary novelties. Formal phylogenetic analyses will guide these determinations, not a priori determinations made without such analyses. To date, a morphology-based phylogeny of the freshwater gomphonemoid diatoms (Kociolek & Stoermer 1993a) suggests that a suite of features diagnose Gomphonema. These features include: differentiated apical pore fields, a single, “typical” stigma (or in a few cases no stigma), no marginal laminae and no axial plates, and with doubly-punctate striae with flaps occluding the areolae. Diatoms described here from the Himalayas are placed firmly in the genus Gomphonema. We are far past the time to be constrained by typological, non-evolutionary, approaches to taxonomy and systematics.