Discussion

Psammina filamentosa is characterized by the long conidial arms and can therefore not be confused with any other Psammina species when mature conidia are examined (Figs 2 & 4). Mature conidia of P. filamentosa differ distinctly from the mature conidia of the more common P. stipitata in the longer and often also somewhat tapered conidial arms. The width of the conidial arms in P. filamentosa is variable and overlaps with the width of the arms of P. stipitata (Fig. 2). No collections were found, however, in which the length of mature conidial arms were intermediate between the two species (Fig. 2). Young conidia of P. filamentosa are very variable in the shape and size of the conidial arms but in all specimens the typical full-grown conidia with long conidial arms were present. Also, compared to mature conidia of P. stipitata, the arms of similar-sized young conidia of P. filamentosa lack clear septa and are more slender. Mature conidia of P. inflata cannot be confused with young conidia of P. filamentosa or P. stipitata since they are inflated and on average 4.5 μm wide (Figs 2 & 4). Psammina filamentosa and P. stipitata can, however, co-occur: at two locations they were collected from the same tree and stone. This is not surprising given that both species have a similar ecology, growing mostly lichenicolous and algicolous in the same habitat, and given that P. stipitata seems to be rather common. In the field, with a strong hand lens (≥ ×15 magnification), the surface structure of the sporodochia can provide a hint as to the identity of the species. In P. stipitata, the conidia are not intertwined with each other and they therefore mostly have a granular appearance (Fig. 3C). The conidia of P. filamentosa, in contrast, generally do intertwine with each other, resulting in the surface appearing more or less smooth rather than granular (Fig. 1C).

Fig. 2. Conidial arm width and length of 36 Psammina collections of four species growing on algae and lichens on the dry side of bark or stone. Points and error bars depict the means and standard deviations of 20 conidial arm measurements. Dots connected with an interrupted line indicate measurements from different sporodochia of the same collection. In colour online.

Fig. 3. Psammina stipitata. A–C, sporodochia. A, on Psilolechia lucida (H.K. 1857). B, on Lepraria incana (H.K. 1874). C, close-up on Lepraria incana (H.K. Kolk 1855). D, conidiophore (arrow 1) and young conidium (arrow 2) (H.K. 1754). E, mature squashed conidia (H.K. 1873). Scales: C = 100 μm; D & E = 20 μm.

Fig. 4. Comparison of mature conidia of algicolous and lichenicolous Psammina species. Psammina inflata (H.K. 1745), P. stipitata (H.K. 419), Psammina sp. (H.K. 1632) and P. filamentosa (H.K. 1856). Scale: applies to all images = 20 μm.

The arms of Psammina stipitata were originally described as 25–50 × 3–3.5 μm on the basis of the single collection then known (Hawksworth Reference Hawksworth1979). Here, we found that the arms are generally somewhat shorter and more variable in width, 21–40 × 3.0–4.0 μm (Fig. 2). The dimensions of the conidial arms of P. stipitata thus overlap with the dimensions of those originally reported for P. simplex, (7–)15–25(–29) × (1.5–)2–4 μm (Earland-Bennett & Hawksworth Reference Earland-Bennett and Hawksworth1999). The number of septa in the conidial arms was originally described as the main difference between P. stipitata and P. simplex, but later Earland-Bennett & Hawksworth (Reference Earland-Bennett and Hawksworth2005) considered that feature less reliable. Also, the number of arms per conidium is variable in P. stipitata and cannot be used to distinguish it from P. simplex. Given the overlapping measurements between the two species and the lack of other diagnostic features separating them, we suggest that P. simplex might prove to be a synonym of P. stipitata. A molecular study, ideally including the holotypes of both species, could conclusively resolve this issue but it was not possible to attempt this during the current study.

An unidentified Psammina was collected three times and the dimensions of the conidial arms resemble those of P. stipitata but differ in being more slender and slightly attenuated towards the tip (Figs 2 & 4). The conidial arms of these unidentified collections measure 28–37 × 2–3 μm and so match those reported for P. mariae-theresiae Dias & Teixeira, a species described from dead branches of Smilax nigra (Dias & Teixeira Reference Dias and Teixeira1963), material of which we have not seen.

Psammina inflata, characterized by the short and inflated conidial arms, was originally described from the UK (Earland-Bennett & Hawksworth Reference Earland-Bennett and Hawksworth1999), and has also been reported from France (Roux et al. Reference Roux, Monnat, Gonnet, Gonnet, Poumarat, Esnault, Gardiennet, Bertrand, Bauvet and Houmeau2018), Germany (Eichler et al. Reference Eichler, Cezanne and Teuber2010a) and Luxembourg (Eichler et al. Reference Eichler, Cezanne, Diederich, Ertz, van den Broeck, van den Boom and Sérusiaux2010b). We now report it also from three localities in the Netherlands (Figs 2 & 4).

In conclusion, the new species P. filamentosa is distinguished from all other described Psammina species by the long conidial arms. The genus Psammina now includes ten species which are saprophytic, lichenized, lichenicolous, or algicolous. The genus has no known sexual morph and is currently classified in the subphylum Pezizomycotina, but not referred to any order or family (Wijayawardene et al. Reference Wijayawardene, Hyde, Al-Ani, Tedersoo, Haelewaters, Rajeshkumar, Zhao, Aptroot, Leontyev and Saxena2020). There is, however, a need for a molecular study of the genus to determine whether the three species described from plants belong in the same genus. Hawksworth (Reference Hawksworth1979) described the first lichenicolous Psammina species and pointed out that the type, P. bommeriae Sacc. & M. Rousseau, had been considered an acervular coelomycete rather than a hyphomycete by Sutton (Reference Sutton1980; incorrectly as ‘bommerae’). The conidiophores, arising from a basal tissue (textura intricata) structure, were more appropriately viewed by Hawksworth (Reference Hawksworth1979) as a rudimentary stroma. This feature alone, however, does not provide a basis for separating the plant-based species from the others now recognized in the genus. In the case of P. elegiae S. J. Lee & Crous described from dead culms of Elegia juncea, no definite hymenium-like layer was formed (Lee & Crous, Reference Lee and Crous2003). Also, the conidiophores of P. lobariae (Diederich & Etayo) Earl. Benn. & D. Hawksw., the only lichenicolous species growing on foliaceous lichens, do arise from a structure similar to an acervulus (Etayo & Diederich Reference Etayo, Diederich, Daniels, Schulz and Peine1995; Earland-Bennett & Hawksworth Reference Earland-Bennett and Hawksworth2005). Wijayawardene et al. (Reference Wijayawardene, Hyde, Wanasinghe, Papizadeh, Goonasekara, Camporesi, Bhat, McKenzie, Phillips and Diederich2016) excluded Psammina from their treatment of coelomycete fungi with brown or brownish conidia but noted a similarity in the conidia to those of Coelodictyosporium Thambug. & Hyde, but that genus has definite rounded pycnidia enclosing the conidiogenous structures and walls of angular pseudoparenchymatous cells (textura angularis).