Rorippa sylvestris × R. amphibia (R.
× anceps (Wahlenb.) Rchb.), Hybrid Yellow-cress

Native, rare, but very probably under-recorded.

1945; MCM & D; Corraslough Point, Upper Lough Erne.

This was declared the most common hybrid Rorippa in B & I by Stace et al. (2015). It has petals much longer than the sepals (so differs from R. palustris (Marsh Yellow-cress)), pinnate leaves (so not R. amphibia (Great Yellow-cress)) and non-clasping leaf bases and a relatively long terminal leaf lobe (so not R. sylvestris (Creeping Yellow-cress)). The slender fruits stand out from the inflorescence stalk at almost 90^o (patent) and are shorter (4mm) than the pedicels that bear them (7 mm) (Crawley 2005).

Just seven records of this hybrid from separate tetrads exist in the Fermanagh Flora Database, but as both parent species are common in the VC, very probably it is simply overlooked and under-recorded. The New Atlas map displays records from five Fermanagh hectads in three date classes, but it should actually display one more. The plant is found on damp to wet lakeshores and so far it has only been found scattered around Lough Erne and the nearby Derryhowlaght Lough.

When, in the 1960s, Jonsell looked at Rorippa forms in B & I during the research for his 1968 taxonomic revision of the genus (from which, it is important to remember, he omitted section Cardaminum, ie the white-flowered water-cresses), he found that records in these islands of the R. sylvestris × amphibia hybrid were sparse, both in herbaria and in the literature (Jonsell 1968). However, his own field study and examination of herbaria specimens in both B & I showed that this hybrid, intermediate in form between its parent species, and like both its parents, variably fertile, was not rare in the areas he studied. It had merely been previously overlooked (see also C.A. Stace, in: Hybridization, p. 152).

Jonsell found both tetraploid and pentaploid forms of this hybrid in English rivers, clearly indicating that both tetraploid and hexaploid cytotypes of R. sylvestris take part in hybridisation with tetraploid R. amphibia. Both of these cytotypes are partially fertile, producing some good seed, but are not fully fertile (Jonsell 1968). Backcrossing to both parents, particularly with R. amphibia, has been detected in several places (Jonsell 1975). In the Thames valley, for instance, Jonsell found that, "from a number of these rivers repeated collections of intermediate hybrid specimens exist". Elsewhere in S England, eg in the River Severn from Shrewsbury down to Tewkesbury, he found evidence of extensive hybridisation, leading to distinct signs of introgression in R. sylvestris, "so that pure specimens of the latter [ie R. sylvestris] were in some cases not found at all" (Jonsell 1968, pp. 131-2).

In Ireland, R. × anceps was first reported from Co Leitrim (H29) by Praeger in 1909. Subsequently, it was discovered in Cos Waterford (H6), Clare (H9), Roscommon (H25) and Fermanagh (H33) (Cen Cat Fl Ir 2). Three voucher specimens of this hybrid discovered by R.D. Meikle exist at K (Nos 116 & 2123 plus one other), collected in 1945 and determined by Jonsell (Jonsell 1968, pp. 132-3). On the basis of his examination of Irish herbarium material, Jonsell questioned whether the curious serrated leaf-shape typical of Irish R. amphibia plants is not an introgressed character from R. sylvestris as a result of their frequent hybridisation and consequent genetic mingling (Jonsell 1968, p. 133, and see his Fig. 15, H & I).

The details of the other six records are: Gubbaroe Point, Lower Lough Erne, 5 July 1945, MCM & D, K; marshy shore, River Erne near Enniskillen, 13 July 1945, MCM & D, K; Duross Point, Lower Lough Erne, 16 August 1946, R. Mackechnie, det. T.C.G. Rich, RNG; jetty at Tully Bay, Lower Lough Erne, 22 July 1984, M.J.P. Scannell, det. T.C.G. Rich, DBN; Derryhowlaght Lough, Upper Lough Erne, 1988, NI Lakes Survey, det. T.C.G. Rich, BEL; Geaglum Jetty, Upper Lough Erne, 28 August 1988, C.D. Preston.

In Britain, R. × anceps is most frequently recorded on the banks of the larger rivers and by drains in wetlands that flood in winter. There are also records from the sides of reservoirs and ponds, water meadows and occasionally from rubbish dumps (Stace et al. 2015). In Europe, R. × anceps is widespread in C & E Europe and local in the W and S. It is, however, absent from many islands and from most of the Mediterranean region (Jalas & Suominen 1994, not mapped).