Ribosomal DNA variation and population differentiation in Salicornia L

Analysis of nuclear rDNA variation within and between populations of Salicornia from the salt marsh at Stiffkey, Norfolk has detected new patterns of genetic differentiation. Individual plants may have alternative 10.5 and 12 kb repeat units. A strong association between the frequency of particular genetic variants and the environmental gradient of tidal inundation was evident. Salicornia dolichostachya Moss from sandy flats on the lowest parts of the marsh, near to the seaward edge, and S. fragilis Ball & Tutin from deep creeks were both monomorphic for the 12 kb variant; in contrast, S. pusilla J. Woods, characteristic of the highest parts of the tidal range, was monomorphic for the 10-5 kb variant. The more phenotypically variable S. europaea L. and S. ramosissima J. Woods populations that are found in large areas of lower and upper marsh, respectively, were heterogeneous for rDNA variant type. Nevertheless, the frequency of the 12 kb variant was significantly higher in plants from the lower marsh than in those from upper marsh, and in plants from low-lying pans than in those from raised interfluves; the 10.5 kb variant had the converse distribution. Variation in rDNA was not obviously associated with variation in morphology, or with variations in isozyme frequency established previously. Comparison of these results with those for populations from an extensive study, ranging from Anglesey (N. Wales) to S.W. Spain and the Gulf coast of Saudi Arabia, suggested that the rDNA variation and its association with environmental variation are complex and site-dependent. Plants from Anglesey had 10.5 and 11.5 kb variants, whereas only a 10-75 kb variant was detected in material from Spain and Saudi Arabia. rDNA variants were used as genetic markers in order to test the hypothesis that Salicornia is predominantly an inbreeder; conventional breeding experiments have been hampered by its highly specialized morphology and this represents the first direct, sensitive test of an idea that has been suggested mainly on morphological and phenological grounds. Analysis of rDNA in 38 maternal plants from Stiffkey and 2112 of their progeny provided no evidence for out-crossing.