Continental concordance

Although a degree of phylogeographic concordance can sometimes be found at regional scales, the picture becomes more complicated when we extend the geographical area of comparison. In many parts of the world, broad-scale phylogeographic patterns have been influenced by the Pleistocene glaciations (Hewitt, 2004). Over the past 700 000 years, major climatic oscillations caused large areas of land to be covered intermittently by vast sheets of ice that receded as the temperatures increased, and then spread out once more when temperatures began to drop. During the late Pleistocene, climatic variation occurred particularly rapidly, with temperatures sometimes changing by 10--12 per cent within a 10-year period. In North America, the most recent glaciation period reached its maximum ice coverage between 18 000 and 23 000 years ago, with the ice sheets reaching their maximum retreat between 8 000 and 15 000 years ago. Similar time scales of glaciation and deglaciation occurred in Europe, where ice sheets were less extensive but nevertheless covered a substantial area (Figure 5.15).

Preserved pollen, beetle fragments and fossil records have illustrated how dramatically species' distributions fluctuated throughout glacial-interglacial cycles. When large expanses of land were covered by ice sheets, many species could survive only in glacial refugia, which were located in areas free of ice. In Europe these tended to be located south of the ice front, although there is some

Figure 5.15 Maps of (A) western Europe and (B) North America, showing the approximate southernmost extent of ice (dotted lines) and tundra (dashed lines) during the last Ice Age

evidence for several more northerly refugia in which temperate or cold-tolerant species, including trees (Willis, Rudner and Sumegi 2000) and forest mammals (Deffontaine et al., 2005), may have persisted. The survival of species in isolated refugia, alternating with their more widespread distribution during the interglacial periods when the ice retreated, has had a profound effect on their current genetic distributions. During glacial periods, populations were kept at relatively small sizes within refugia, and therefore a combination of genetic drift and new mutations promoted substantial levels of genetic divergence between these isolated populations. Dispersal away from refugia during interglacial periods often brought these previously separated populations into contact with one another, at which point hybridization between lineages would often occur. Numerous phylogeographic studies have provided us with considerable information on postglacial recoloniza-tion routes and the current distribution of genetic diversity in formerly-glaciated versus refugial areas.

European postglacial recolonization routes

During interglacial periods, as the ice began to retreat, species dispersed out of refugia and recolonized the newly available habitats further north. In Europe, the three main refugia were in Iberia (Portugal and Spain), Italy, and the Balkans (see also Box 5.2). Fossil and molecular data have been used to reconstruct the dispersal routes of species spreading northwards out of these refugia as the ice sheets receded. By comparing the genetic similarity of populations in formerly glaciated regions to those in or near glacial refugial sites, we can sometimes get a reasonably clear picture of which refugial populations the current non-refugial populations are descended from. This in turn allows us to retrace the colonization routes that were followed by individuals emanating from each refugium. In Europe, dispersal was often impeded by major mountain ranges, including the Cantabrians, Pyrenees, Alps and Transylvanians, all of which tend to run in an east--west orientation. The Pyrenees and Alps, for example, presented a barrier to the meadow grasshopper (Chorthippus parallelus) travelling north from Iberia, although other species such as the hedgehog (Erinaceus europaeus) apparently were able to cross the mountains (Hewitt, 1999).

Differences in ecology, particularly with respect to dispersal abilities, inevitably will mean that species followed a variety of colonization routes. Nevertheless, there is some degree of concordance in the routes that were followed, even between some markedly different taxa. For example, the route taken from a refuge in Iberia towards southern Scandinavian seems to be remarkably similar for the brown bear (Ursus arctos) and white oaks (Quercus spp.) (Taberlet and Bouvet, 1994; Dumolin-Lapegue et al., 1997); similarly, the common beech (Fagus sylvatica) and the meadow grasshopper (Chorthippus parallelus) apparently followed a comparable route from their refugia in the Balkans to the southeast of France

Figure 5.16 The main routes of postglacial colonization (arrows) that were followed by a wide range of taxa as individuals moved away from their refugia in western Europe. The thick lines represent the zones where populations emanating from different refugia tend to come into contact with one another. Hybridization between conspecific populations has often occurred at these so-called contact zones. Redrawn from Taberlet et al. (1998)

Figure 5.16 The main routes of postglacial colonization (arrows) that were followed by a wide range of taxa as individuals moved away from their refugia in western Europe. The thick lines represent the zones where populations emanating from different refugia tend to come into contact with one another. Hybridization between conspecific populations has often occurred at these so-called contact zones. Redrawn from Taberlet et al. (1998)

(Cooper, Ibrahim and Hewitt, 1995; Demesure, Comps and Petit, 1996). Figure 5.16 shows what are believed to be the main routes of postglacial dispersal for species moving away from European refugia.

The unravelling of postglacial dispersal routes has helped researchers to identify contact zones where populations from separate refugia have met. Hybridization between species or genetic lineages frequently occurs at these zones, which means that they provide another aspect of phylogeographic concordance. We therefore find a degree of concordance at the continental level in Europe with respect to both recolonization routes and zones of hybridization. Nevertheless, despite some similarities, overall phylogeographic patterns across Europe are broadly dissimilar. A review of cpDNA from 22 European tree and shrub species found significant differences in the distribution of genetic variation within and among populations that reflected the different history of each species (Lascoux et al., 2004). Another study surveyed ten taxa, including mammals, amphibians, arthropods and plants, and this too revealed markedly different phylogeographies for each species. Differences included variable numbers of genetic lineages, discordant distributions of lineages, and varying levels of genetic divergence (Taberlet et al., 1998). The author of that study did note that expectations of substantial concordance among such an ecologically diverse group of taxa may be unrealistic (Taberlet, 1998). Further evidence of continental concordance may emerge as more data become available, but so far it seems that relatively large spatial scales lead to highly variable phylogeographic patterns. There is, however, one other aspect of phylogeography that has been remarkably consistent across a wide range of taxa in several continents, and that is the distribution of genetic diversity.

Box 5.2 Marine glacial refugia

Most of the research into the effects that glacial--interglacial cycles have had on the current distributions of species and their genetic lineages has been conducted on terrestrial plants and animals, although some recent studies have obtained evidence of marine glacial refugia. Both on land and in the sea, populations at the sites of former glacial refugia are expected to show higher levels of genetic diversity than populations in formerly glaciated areas, largely because the former often contain numerous genotypes that did not disperse northwards following glacial retreat. In the red seaweed Palmaria palmata, haplotype and nucleotide diversity estimates were higher in the English Channel compared with a number of other sites along the European and North American coasts, a finding that identifies the English Channel as a likely marine glacial refugium (Provan, Wattier and Maggs, 2005). Similar patterns of genetic diversity in the brown seaweeds Fucus serratus and Ascophyllum nodosum add further support to the identity of this refugium (Stam, Olsen and Coyer, 2001; Coyer et al., 2003). These initially may seem paradoxical because the English Channel was mostly dry land during the last glacial maximum, but a trench known as the Hurd Deep was maintained at this site throughout the glacial-interglacial cycles. A marine lake was formed in this trench, and it is here that an unknown number of marine species could have persisted until the ice sheets retreated (Lericolais, Auffret and Bourillet, 2003).

Distribution of genetic diversity

We know from previous chapters that the genetic diversity of populations is influenced by many factors, including population size, gene flow, mating system and natural selection, and therefore we may expect little agreement between species in the way that genetic diversity is distributed across broad spatial scales. Alternatively, it is possible that widespread environmental phenomena such as glaciation had similar impacts on the distribution of genetic diversity in multiple species, for example the postglacial colonization of more northerly sites was typically associated with founder effects and therefore we may expect to find relatively high levels of genetic diversity in or near glacial refugia and relatively low levels in formerly glaciated regions. This would depend to some extent on dispersal patterns. Founder effects should be particularly pronounced in species that often disperse over short distances but also periodically achieve long-distance dispersal. In these species there will be a time lag between the arrival of occasional longdistance dispersers and most other representatives of the gene pool, and this delay will allow the early arrivals to establish themselves and colonize neighbouring sites so that when the slower dispersers arrive they will struggle to compete with established populations (Ibrahim, Nichols and Hewitt, 1996). In contrast, species that show more constant dispersal rates should maintain a greater proportion of genetic diversity throughout their range. This should be true of both slow and rapid dispersers, as long as multiple genotypes disperse and colonize new sites en masse.

We therefore have the potential for postglacial dispersal with and without associated founder effects, and a review of the literature provides examples of both scenarios. A number of highly dispersive freshwater invertebrate species have relatively low levels of genetic diversity at northerly latitudes, which may be attributed to generally low levels of gene flow combined with occasional longdistance dispersal. Furthermore, because these species can reproduce asexually, populations can undergo rapid growth before receiving additional immigrants, and therefore a founder effect may persist for thousands of generations (Boileau, Hebert and Schwartz, 1992; Freeland, Rimmer and Okamura, 2004). At the other extreme, several species of Haircap mosses (genus Polytrichum) provide an example of how the rapid simultaneous dispersal of multiple genotypes led to postglacial dispersal without founder effects. During the ice ages, Haircap mosses were confined to southern refugia, but extensive spore dispersal allowed them to quickly colonize more northerly latitudes once the ice had retreated. This rapid dispersal has resulted in very low levels of genetic differentiation among populations, and no decrease in genetic variation with increasing latitude (Van der Velde and Bijlsma, 2003).

Given these two contrasting scenarios, we must now ask whether one predominates over the other, in other words whether there is any evidence for concordant distributions of genetic diversity along a latitudinal gradient from southerly refugia to northerly postglacial sites. A survey of the literature suggests that many northerly populations have indeed retained a postglacial founder effect that has reduced the genetic diversity of populations at high latitudes. In one review, data from 42 species of freshwater and anadromous fishes in North America yielded a significant inverse relationship between nucleotide diversity and latitude over formerly glaciated regions. Similar patterns of relatively low genetic diversity in formerly glaciated versus refugial areas have been found in many other taxa, including North American voles (Microtus longicaudus; Conroy and Cook, 2000), dragonflies (Anax junius; Freeland et al., 2003), mosquitoes (Wyeomyia smithii; Armbruster, Bradshaw and Holzapfel, 1998), herbs (Asclepias exaltata; Broyles, 1998), and black spruce (Picea mariana; Gamache et al., 2003), and European rock ferns (Asplenium spp.; Vogel et al., 1999), grasshoppers (Chorthippus parallelus; Hewitt, 1999), woodmice (Apodemus sylvaticus; Michaux et al., 2003), bryophytes (Leucodon sciuroides; Cronberg, 2000) and butterflies (Polyommatus coridon; Schmitt and Seitz, 2002). Thus it would appear that, although there are exceptions to the rule, there is a degree of phylogeographic concordance in the form of an inverse correlation between genetic diversity and latitude.

Was this article helpful?

0 0

Post a comment