G. Fekete szerk.: Studia Botanica Hungarica 9. 1974 (Budapest, 1974)
Vida, László: Diatoms from a brooklet affluent to Arax river (Armenian Republic, USSR)
Summarizing these data one conclusion seems to be inevitable, namely that the differing diagnoses of Thalassiosira fluviatilis and Thalassiosira hustedtii represent the same species in different phases of the frustule development. The formation of jelly that might lead to formation of colonies (a rare phenomenon, here) is not an absolutely obligatory taxonomic mark, as it can be observed in case of related species as well. When the structure of a cell wall tends to be at the border of microscopic distinction, then structure is not in the least an exact taxonomic mark. The radially punctated lines of Thalassiosira fluviatilis valves can be observed clearly only when frustules are large enough, the embedding medium is of a high refraction index, and observation is made with an objective of excellent aperture. If all these conditions are not present simultaneously, valves of Thalassiosira fluviatilis show exactly the same structure as described in the diagnosis and presented in the drawings of Thalassiosira hustedtii. It is necessary to remark here that neither the number of central pores nor the density of striae on the margin can be regarded as specific marks when separating different taxa of centric diatoms. The number of pores in the centre of valves might vary from 0(?), 1 to 8, according to my observations (cf. the number of central pores at Melosira roeseana! ) Meanwhile marginal striae varied between 12/10 p and 18/10 p . I think highly probable that Thalassiosira hustedtii belongs to the form range, or rather to the synonims, of Thalassiosira fluviatilis . It is also necessary to remark here another feature connected with this diatom species. Namely, Thalassiosira fluviatilis apperently belongs, according to the available data, to the rarest diatoms. However, one should take into consideration the difficulties connected with the observation of valve structure, and the fact that by a superficial perusal of mixed diatom material Thalassiosira fluviatilis valves might appear at first sight as strongly corroded Melosira variáns frustules. So it looks easily possible that the species is much widerspread than supposed to be. In any case, I collected Thalassiosira fluviatilis several times from different localities in Hungary, too. I found it in the river Danube, in its affluent rivers Rába and Rábca, in saline brooks and canals near Gyor, and also in lake Balaton; and found its frequency to be often abundant. The microscopic examination and comparison proved that specimens of Armenian and Hungarian Thalassiosira fluviatilis are taxonomicaUy identical. ad 7. Diatoma vulgare var, ovális (FRICKE) HUST. Fig. 2. HUSTEDT' s diagnosis gives 8-14ju as the length of cells for the taxon. All specimens of the sample are much smaller than that; their length varies between 4-7 p . It may perhaps represent an independent saline form of the species. ad 8. Fragilaria capucina var, lanceolata GRUN. Fig. 3. In contrast with the original diagnosis, specimens of the sample never form colonies, and are always solitary ones. The transapical ribs are strong, the central area has a distinct contour. The habitus of frustules has a Synedra character in all respects. These contradictions underline the difficulties in the separation, acceptable only from practical points of view, of the genera Fragilaria and Synedra, emphasized clearly by HUSTEDT (1930).
