Durabilispora carpatica: a new genus raised from Tatra soil in a pot

Nine authors, two pots, one new genus

MycoKeys published on 24 June 2026 the description of Durabilispora, a new genus of arbuscular mycorrhizal fungi, alongside its type species Durabilispora carpatica and a second new species, Dominikia tatrensis. The paper occupies pages 313–340 of volume 134; it was received on 2 February 2026 and accepted on 13 April. Janusz Błaszkowski of the West Pomeranian University of Technology in Szczecin leads a team of nine drawn from that university, the Jagiellonian University and the W. Szafer Institute of Botany of the Polish Academy of Sciences in Kraków, and the Universidade Federal do Rio Grande do Norte in Natal, Brazil. The description is open access.

Neither fungus was picked in a mountain meadow. Thirty-two trap cultures were examined, each holding roughly 500 mL of autoclaved quartz sand inoculated with 50 g of a fresh field mixture of rhizosphere soil and root fragments from Tatra meadow plants, with about twenty seeds of ribwort plantain, Plantago lanceolata, sown on the surface as the culture host. Sealed in bags against contamination, the pots spent six months in a growth chamber at 22 degrees Celsius on a twelve-hour day, watered with 50 mL of distilled water once a week. Two of the isolates recovered from them, numbered 527 and 530, proved to be undescribed. The authors say plainly why they worked from trap cultures at all: repeated attempts to grow either isolate in single-species culture failed.

The inoculum behind Isolate 527, now Durabilispora carpatica, was collected on 18 July 2023 by Anna Ronikier and Paulina Janik in the lower Miętusia Valley at 938 metres, in the lower montane belt, on open ground left by windthrow beside a forest and a grazed meadow; the field soil there had a pH of 5.17. Isolate 530, Dominikia tatrensis, came from soil the same two researchers collected on 11 May 2023 on the Upłaz meadow at 1,298 metres, at the edge of a spruce stand in the upper montane belt. Both typifications credit both collectors.

A spore wall that will not come apart

Durabilispora carpatica — abbreviated Du. carpatica in the paper, because D. is spoken for by Dominikia — forms compact underground glomerocarps without a peridium: 394–446 µm across when globose to subglobose, 384–395 × 596–646 µm when ovoid, holding anything from tens to about a hundred spores scattered at random inside. That randomness is part of the generic diagnosis. The only other species in the Glomeromycota described with fused glomerocarps, Sclerocystis coremioides, arranges its spores radially around a central plexus of hyphae and wraps the whole thing in a peridium.

The spores themselves are orange to yellowish brown, globose to subglobose at (35–)45(–56) µm in diameter, and frequently ovoid to oblong, in which case they measure 19–57 × 24–71 µm. The spore wall has four permanent, smooth layers. Layer three is laminate, built of sublayers thinner than 0.5 µm which, the description records, do not separate from one another even in vigorously crushed spores — whereas in most Rhizoglomus, the nearest genus in the tree, the same laminae come apart readily under the same treatment. That stubbornness is what the genus is named for: Durabilispora refers to the durability of the components of the spore wall and of the subtending hyphal wall. The epithet carpatica points to the Carpathians, tatrensis to the Tatras themselves. Germination is recorded as unknown.

Both novelties rest on sequences of the 45S nuclear rDNA region, that is 18S-ITS-28S, concatenated with the rpb1 gene. In the first tree, Isolate 527 occupied an autonomous, fully supported clade basal to the Rhizoglomus clade, set against seventeen Rhizoglomus species, with Sclerocystis sinuosa and Silvaspora neocaledonica as the outgroup. Its 45S sequences differ from those of its two nearest neighbours, Rhizoglomus invermaium and R. melanum, by 12.4 to 13.2 per cent — above the ten per cent threshold that separates generic clades elsewhere in the Glomerales. Dominikia tatrensis was analysed in a separate tree against thirteen other Dominikia species, where it came out sister to the clade holding Dominikia aurea. The two species share a paper, not a family: Durabilispora sits in the Rhizoglomeraceae and Dominikia in the Dominikiaceae, and they are not sister taxa.

The clade already had a working number

The genus did not appear from nowhere. The environmental sequences that clustered with Du. carpatica here had already been assigned in the EUKARYOME database to an undescribed generic clade labelled gen21, and in the 2024 tree by Tedersoo and colleagues that clade sat next to Rhizoglomus, exactly where the Tatra material puts it. The 2026 paper therefore names and typifies a lineage that sequence surveys had already isolated in the data.

The holotype of Du. carpatica is a slide of spores, ZT Myc 0067490, deposited at ETH Zurich; the isotypes stayed in Szczecin. MycoBank carries the genus as 863195, the type species as 863196 and Dominikia tatrensis as 863197. The full text and figures are also mirrored in PubMed Central.

What the description stops short of establishing

The easiest mistake to make is to file this under Tatra speciality. The paper says the opposite. Under ecology and distribution it concludes that Du. carpatica is widespread through the world although it occurs rarely, and it says the same of D. tatrensis. The evidence is entirely environmental DNA: four sequences from forest soils in Estonia, Mexico and Qatar at 96.6 to 99.5 per cent identity for the type species; five sequences for D. tatrensis, from Estonian shrubland and broadleaf forest soil, from roots of Brachypodium pinnatum in a North Bohemian meadow and from a soil sample taken on the Qinghai-Tibet Plateau in China. The verifying analyses are marked as data not shown, and no specimen exists from any of those countries. A sequence recovered from soil or from a root is not a find.

The paper also carries an older Polish record. Isolate 299, morphologically identical to Isolate 527, came from rhizosphere soil of Rosa rugosa in Baltic dunes near Jastarnia on 5 September 2015, collected by Błaszkowski himself, and was never described because attempts to obtain genetic data from it were unsuccessful. That is a morphological match without sequences: a reason to suspect, not a record to count. For much the same reason no plant can be called a host here. The Tatra cultures began as a mixture of soil and roots from many meadow species, the ribwort plantain was sown in the pot afterwards, and where the authors mention R. rugosa they state outright that no molecular analysis confirmed the symbiosis.

Even the coordinates deserve care, which is why none appear above. The paragraph listing other examined isolates repeats the Tatra type locality coordinates for the Baltic sample, although the methods section gives Jastarnia its own, entirely different pair — an evident slip in the setting. The Tatra pair itself falls conspicuously south of the Miętusia Valley and of the second Tatra site in the same paper, though that is our reading of a map rather than anything the authors assert. In this article the valley name, the altitude and the habitat description are firmer ground than the figures.

Finally, neither the genus nor either species has any extinction-risk assessment, and the paper proposes none. Nor is there a fruiting body in any everyday sense. These fungi make no cap, and nothing here bears on foraging.

What is left for someone standing in the valley

Almost nothing to look at, and that is the interesting part. Glomerocarps are a fraction of a millimetre wide and live in the soil; the whole story played out between a sieve, a microscope and a sequencer, six months after the soil left the mountain. Earlier work by Błaszkowski and by Zubek and colleagues had recorded 26 species of arbuscular mycorrhizal fungi in the Tatra region; this paper adds two species and a genus, from material traced to two named sites. Walking up the Miętusia Valley, you can look at the windthrow and the grazed grass knowing that 50 g of that soil and root litter, given half a year in a pot, was enough to bring out a lineage nobody had named — and one that, on the evidence of the sequence databases, is also sitting in forest soils on the other side of the world.