The North-Atlantic has warmed faster than all other ocean basins and climate change scenarios predict sea surface temperature isotherms to shift up to 600 km northwards by the end of the 21st century. The pole-ward shift has already begun for many temperate seaweed species that are important intertidal foundation species. We asked the question: Where will climate change have the greatest impact on three foundational, macroalgal species that occur along North-Atlantic shores: Fucus serratus, Fucus vesiculosus, and Ascophyllum nodosum? To predict distributional changes of these key species under three IPCC (Intergovernmental Panel on Climate Change) climate change scenarios (A2, A1B, and B1) over the coming two centuries, we generated Ecological Niche Models with the program MAXENT. Model predictions suggest that these three species will shift northwards as an assemblage or “unit” and that phytogeographic changes will be most pronounced in the southern Arctic and the southern temperate provinces. Our models predict that Arctic shores in Canada, Greenland, and Spitsbergen will become suitable for all three species by 2100. Shores south of 45° North will become unsuitable for at least two of the three focal species on both the Northwest- and Northeast-Atlantic coasts by 2200. If these foundational species are unable to adapt to the rising temperatures, they will lose their centers of genetic diversity and their loss will trigger an unpredictable shift in the North-Atlantic intertidal ecosystem.

No abstract

To better assess the current state of phaeophycean phylogeny, we compiled all currently available rbc L, 18S, and 26S rDNA sequences from the EMBL/GenBank database and added 21 new rbc L sequences of our own. We then developed three new alignments designed to maximize taxon sampling while minimizing information loss due to partial sequences. Phylogenetic analyses were performed on separate and combined data sets (with and without taxa from the sister classes Tribophyceae and Phaeothamniophyceae as outgroups) using a variety of assumption sets, tree-drawing algorithms (parsimony, neighbor joining, and likelihood), and resampling methods (bootstrap, decay, jackknife). Partition homogeneity testing (PHT) by codon position within rbc L showed that all positions could be used despite mild third position saturation. PHT by gene and domain within rDNA showed that the 26S D1 and D2 regions do not enhance phylogenetic signal even when combined with the 18S. The rbc L and rDNA (excluding the 26S D1 and D2) could be combined under PHT. The topology of the combined tree was the same as that of the rbc L tree alone, but bootstrap support was consistently higher in the combined analysis, applied to more branches, and enabled the establishment of sister group relationships among six orders. Although the taxon sampling for the combination tree was lower ( n ϭ 22) than for individual gene analyses ( n ϭ 58 for rbc L and n ϭ 59 for rDNA), results show that the Laminariales (previously reported) and Sphacelariales (new) are both paraphyletic. Choristocarpus tenellus (Kützing) Zanardini is the most basal phaeophyte and the Dictyotales the most basal order. In contrast, the Laminariales sensu stricto ( s.s. ) and Ectocarpales sensu lato ( s.l. ) are the most derived. For phylogenetic studies in the Phaeophyceae, rbc L has more resolving power than rDNA, though the reason for this is unclear based on the fact that both genes are highly conserved.

No abstract

In plant populations alleles often deviate from a random distribution and reveal positive autocorrelation at short distances. In species with both clonal and sexual reproduction, such clustering may be because ramets of the same genet were sampled at nearby locations. Alternatively, clustering may be the result of limited gene¯ow through pollen or seeds (isolation-by-distance). Here, we modify a conventional spatial autocorrelation analysis using the join-count statistic in order to dierentiate between these two causes of genetic structure. We examined the distribution of seven microsatellite loci representing 37 alleles in a 20´80 m plot of a perennial population of eelgrass Zostera marina L. In analysing join-counts between all like genotypes we found signi®cant genetic autocorrelation among ramets at distances between 1 and 7 m (P < 0.001). We then excluded joins between clonemates which were identi®ed from the expected likelihood of their seven-locus genotypes. Without joins within genets, no autocorrelation was evident, indicating that most of the signi®cant genetic clustering was caused by clonal spread. At distances up to 27 m, alleles were distributed at random, indicating a panmictic population at this spatial scale. These results illustrate the need for an a priori estimation of genet±ramet structure in clonally reproducing plants in order to avoid erroneous inferences about putative gene¯ow at various spatial scales.Keywords: clonal reproduction, genetic structure, join-count, microsatellites, spatial autocorrelation, Zostera marina. IntroductionSpatial autocorrelation techniques are a powerful tool for detecting the nature and scale of genetic dierentiation over a range of spatial scales (e.g. Sokal & Oden, 1978a,b; Cli & Ord, 1981). Simulations have demonstrated that these methods are able to distinguish between several causes of genetic structure, i.e. directional migration, selection or restricted gene¯ow (isolation-by-distance, Sokal et al., 1997). In contrast to statistics based on allele frequencies such as F ST , autocorrelation techniques make few assumptions regarding the underlying population genetic model (Heywood, 1991). No information is lost by pooling of individuals into arbitrary sampling areas for subsequent comparison of gene frequencies among subpopulations. Instead, in autocorrelation analyses, the explicit information of each spatial location of individuals relative to one another can be utilized (reviewed in Heywood, 1991; Epperson, 1993). Therefore, autocorrelation techniques may have a superior power over methods based on nested samples of gene frequencies for detecting spatial patterns (Epperson & Li, 1996; Epperson, 1997).The numerical value of an autocorrelation statistic is often examined as a function of the Euclidean distance among pairs of plants, a correlogram. The shape of the correlogram allows inferences about the direction and magnitude of the evolutionary processes at work (Sokal et al., 1997). Isolation-by-distance typically results in a clustered distribution...