Multiple origins of Southern Hemisphere Anemone (Ranunculaceae) based on plastid and nuclear sequence data

 Using two molecular data sets, the plastid atpB-rbcL intergenic spacer region and the nuclear ribosomal internal transcribed spacer regions (ITS), the taxonomic affinities of two newly available Anemone species from the Southern Hemisphere were tested. From previous work based on morphology and geographic distribution, it was assumed that A. tenuicaulis from New Zealand was most closely related to the Tasmanian A. crassifolia, whereas the affinity of A. antucensis from Chile and Argentina was regarded as uncertain. Analyses of molecular sequence data from these and 18 other species of Anemone s.lat. (with Clematis as outgroup) result in trees largely congruent with past analyses based on morphology and plastid restriction site data. They strongly support A. richardsonii and A. canadensis (with boreal distributions in the Northern Hemisphere) as paraphyletic to a well supported Southern Hemisphere clade consisting of A. antucensis and A. tenuicaulis. This group of four species is part of an otherwise predominantly Northern Hemisphere assemblage (subgenus Anemonidium s.lat., chromosome base number x=7), including A. narcissiflora, A. obtusiloba, A. keiskeana and A. (=Hepatica) americana. All other austral species included in the present sampling, A. crassifolia (Tasmania), A. knowltonia (=Knowltonia capensis), and A. caffra (both South African), form a separate clade, sister to A. (=Pulsatilla) occidentalis and other Northern Hemisphere anemones (subgenus Anemone s.lat., x=8). Possible phytogeographical links of the Southern Hemisphere species are discussed. Received April 23, 2001 Accepted October 4, 2001     

Circum-Antarctic continental distribution patterns in pteridophyte species

Four major austral continental distribution patterns are evident in pteridophytes. Twenty-two species are completely circum-Antarctic. Another 39 species are partially circum-Antarctic, occurring in Australasia (Australia and New Zealand) and Africa (including Madagascar) but not South America, while 29 are in Africa and South America but not Australasia, and 13 are in South America and Australasia but not Africa. Two hypotheses are considered as explanations for the patterns: continental drift following the breakup of Gondwana and long-distance dispersal. Fossil evidence indicates that the majority of pteridophyte families involved appeared after the southern continents had drifted apart, so long-distance dispersal is likely to explain the distribution of species in these families on now widely separated continents. For those families extant before the break-up, there is no indication in the fossil record that the species involved were present in Gondwana. Aspects of the ecology of the species that are partly or completely circum-Antarctic indicate that long-distance dispersal, rather than continental drift, is a likely explanation for the patterns.     

Imprints of glacial history and current environment on correlations between endemic plant and invertebrate species richness

Aim We evaluate how closely diversity patterns of endemic species of vascular plants, beetles, butterflies, molluscs and spiders are correlated with each other, and to what extent similar environmental requirements or survival in common glacial refugia and comparable dispersal limitations account for their existing congruence. Location Austria. Methods We calculated pairwise correlations among species numbers of the five taxonomic groups in 1405 cells of a 3′ × 5′ raster (c. 35 km²) using the raw data as well as the residuals of regression models that accounted for: (1) environmental variables, (2) environmental variables and the occurrence of potential refugia during the Last Glacial Maximum, or (3) environmental variables, refugia and spatial filters. Results Pairwise cross‐taxonomic group Spearman’s rank correlations in the raw data were significantly positive in most cases, but only moderate (0.3 < ρ < 0.5) to weak (ρ < 0.3) throughout. Correlations were closest between plants and beetles, plants and butterflies, and plants and snails, respectively, whereas the distribution of endemic spiders was largely uncorrelated with those of the other groups. Environmental variables explained only a moderate proportion of the variance in endemic richness patterns, and the response of individual groups to environmental gradients was only partly consistent. The inclusion of refugium locations and the spatial filters increased the goodness of model fit for all five taxonomic groups. Moreover, removing the effects of environmental conditions reduced congruence in endemic richness patterns to a lesser extent than did filtering the influence of refugium locations and spatial autocorrelation, except for spiders, which are probably the least dispersal‐limited of the five groups. Main conclusions The moderate to weak congruence of endemic richness patterns clearly limits the usefulness of a surrogacy approach for designating areas for the protection of regional endemics. On the other hand, our results suggest that dispersal limitations still shape the distributions of many endemic plant, snail, beetle and butterfly species, even at the regional scale; that is, survival in shared refugia and subsequent restricted spread retain a detectable signal in existing correlations. Concentrating conservation efforts on well‐known Pleistocene refugia hence appears to be a reasonable first step towards a strategy for protecting regional endemics of at least the less mobile invertebrate groups.