Hybridization with invasiveRosa rugosathreatens the genetic integrity of nativeRosa mollis

Abstract

Hybridization between invasive and native species often has severe consequences on fitness and survival of the native relative. We investigated the extent of hybridization between the neophyte Rosa rugosa and native R. mollis, an endangered species in Germany. Rosa mollis is found in only one large population in Germany close to the Baltic coast, which has been heavily invaded by R. rugosa for at least 60 years. We analysed all individuals of R. mollis from this mixed population using microsatellite markers and morphological characters and compared these data with those from allopatric populations of R. mollis and R. rugosa. In the mixed population we identified nine plants (45% of the population) as hybrids between R. mollis (seed parent) and R. rugosa (pollen parent) by the presence of microsatellite alleles private for R. rugosa. These individuals were also morphologically intermediate between the parental species. Gene flow from R. mollis into R. rugosa was negligible. We detected a very low genetic diversity and a low number of seeds per hip in the mixed population of R. mollis, pointing to genetic depletion and low fitness. In the light of these results and the difficulties in removing invasive R. rugosa from European coastlines, we discuss possible conservation strategies for this endangered population. © 2012 The Linnean Society of London, Botanical Journal of the Linnean Society, 2012, 170, 472–484.

Low genetic and morphological differentiation in the European species complex ofRosa sherardii,R. mollisandR. villosa(RosasectionCaninaesubsectionVestitae)

“…High levels of within‐population diversity may point to high outcrossing rates in dog roses, as detected by Kellner et al . (), who found a high percentage of spontaneous hybrids (45%) between the tetraploid R. mollis and the diploid R. rugosa in the field. Jürgens, Seitz & Kowarik () observed high RAPD‐based variation within populations, but low variation among populations in the pentaploid R. canina .…”

“…Measurements, a principal component analysis and a canonical discriminant analysis of the data were conducted according to methods described in Kellner et al . (). Additionally, measurements of two taxonomically important characters, length and width of the terminal leaflet, were visualized in box–whisker plots and compared with an analysis of variance (ANOVA) using SPSS v. 20.…”

“…We analysed ten microsatellite DNA loci in 192 individuals of the three species. Primer sequences were taken from Esselink, Smulders & Vosman (2003) and the protocol followed the description given in Kellner, Ritz & Wissemann (2012).…”

“…Selection and terminology of morphological characters followed mainly Leaf Architecture Working Group (1999) and De Cock et al (2008) (Table 2, see also Supporting Information, Table S2). Measurements, a principal component analysis and a canonical discriminant analysis of the data were conducted according to methods described in Kellner et al (2012). Additionally, measurements of two taxonomically important characters, length and width of the terminal leaflet, were visualized in boxwhisker plots and compared with an analysis of variance (ANOVA) using SPSS v. 20.…”

Rosids

“…Rosaceae, the source of the names Rosales and rosids, contain several genera in which taxonomy and genetic patterns are complicated due to apomixis, hybridization and/or polyploidy [e.g. Sorbus L. (Ennos et al ., ; Pellicer et al ., ), Crataegus L. (Campbell & Dickson, ) and Rosa L. (Kellner, Ritz & Wissemann, )]. In this issue, Wrońska‐Pilarek, Bocianowski & Jagodziński () evaluate the use of pollen morphology for assessing hybridization in Crataegus , and Gareca et al .…”