Solitary or mat-forming herb growing in loose tussocks or dense mats with a short, often extensively branched rhizome, with abundant, brown, narrowly lanceolate, acuminate to long-pointed scales at branch ends and at leaf bases. Leaves or bunches of leaves from ends of rhizome branches, in loose clusters.

Leaves 8–20(25) × 1–3 cm, with petiole 3–8(12) cm, 1/2–1/3 of entire leaf length, petiole without joint (see Woodsia). Blades with oblong outline, pinnate to bipinnate with 8–12(14) pairs of pinnae. Pinnae 8–20 × 6–10 mm, longer than broad when well developed (but often stunted), dissected to the mid axis when well developed. Pinnulae dissected 1/4–1/2 to mid axis, with acute lobes but often with lobe ends curved inwards in dry sites (a characteristic of arctic plants of the species vs. temperate ones). Petiole with a few scales in lower parts, otherwise glabrous.

Sori on lower surface of pinnae or pinnulae, (4)5–7(9) per pinnula, ca. 1/2–2/3 along the distance from mid axis to margin (a difference from Woodsia), initially covered by a small, flimsy, glabrous indusium that soon is shed (caducous) when the sporangia begin to develop. Each sorus with 4–10 sporangia. Spores brown, rugose (see Comments for the significance of this feature).

Sexual reproduction by spores; no vegetative reproduction.

Spores are wind dispersed, potentially over very long distances.

The two 'ordinary' ferns of Svalbard are easily distinguished. Cystopteris fragilis has petioles without a joint, twice dissected blades, sori well inside the margin of pinnulae, and indusium glabrous and deciduous; Woodsia glabella has petioles jointed, blades once dissected, sori submarginal, and indusia persistent and hairy.

Cystopteris fragilis is mainly confined to cliffs and coarse screes, in crevices among stones, on both circumneutral and basic substrates but perhaps not on the more acidic ones. It occurs in Svalbard mostly in dry and sunny sites, with a characteristic appearance with incurved leaf apices and a yellowish green colour; however, in shade it retains its flat leaves and green colour but rarely produces spores.

Slightly thermophilous. In the middle and northern arctic tundra zones and the transitional to clearly continental sections. In Svalbard, Cystopteris fragilis is only reported from Spitsbergen but there as rather frequent in the fjord parts, more rare on the coast.

The global distribution is very wide, almost cosmopolitan, and so are the taxonomic complications.

Two very different spore types are found in this species: echinate (with long spines) and rugose (with broad warts), see illustrations in Elven (1984; reproduced in Berg 2000). In plants with intermediate spore shapes the spores are obviously aborting (Elven 1984, see illustration there), suggesting a reproductive barrier. Nearly all arctic plants observed with spores, and certainly all observed in Svalbard, have rugose spores. The plants with rugose spores were described by R. Sim in 1848 from Scotland as a new species, C. dickieana, and this has been an accepted taxon from then on until the late 1990s. Elven (1984) found that the plants of Scandinavia with rugose spores did not belong to a single macromorphological type different from those with echinate spores, but rather belonged to several regional macromorphological types. Berg (1992) did not find support for a separate taxon but rather considered the spore polymorphy to possibly be a single gene effect, a conclusion reflected in his account for Flora Nordica (Berg 2000). The results of the molecular investigations of Dyer et al. (2000) and Parks et al. (2000) made these authors reject C. dickieana as a taxon. Vida's (1974) hypothesis, that partly different diploid genomes partake in "fragilis" and "dickieana", has not found support and is countered by Parks et al. (2000). However, the suggested spore abortion in plants with intermediate spore morphology remains to be explained as does the distributional pattern, i.e., that nearly all arctic plants and a majority of the high alpine ones have rugose spores. The “dickieana” plants are therefore retained, as a var. dickieana, by Elven et al. (2022).

Berg, R.Y. 1992. Sporevariasjon i skandinaviske populasjoner av Cystopteris fragilis. – Blyttia 50: 143–157.

Berg, R.Y. 2000. Cystopteris Bernh. – In: Jonsell, B. (ed.), Flora Nordica. 1. Lycopodiaceae – Polygonaceae: 62–67.

Dyer, A.F., Parks, J.C. & Lindsay, S. 2000. Historical review of the uncertain taxonomic status of Cystopteris dickieana R.Sim (Dickie's bladder fern). – Edinburgh Journal of Botany 57: 71–81.

Elven, R. 1984. Skjørlokene i Nord-Norge. – Blyttia 42: 57–67.

Elven, R., Bjorå, C.S., Fremstad, E., Hegre, H. & Solstad, H. 2022. Norsk flora. 8. Ed. – Det Norske Samlaget, Oslo.

Parks, J.C., Dyer, A.F. & Lindsay, S. 2000. Allozyme, spore and frond variation in some Scottish populations of the ferns Cystopteris dickieana and Cystopteris fragilis. – Edinburgh Journal of Botany 57: 83–105.

Vida, G. 1974. Genome analysis of the European Cystopteris fragilis complex. – Acta Botanica Academica Scientiarum Hungarica 20: 181–192.