Multilocus tests of Pleistocene refugia and ancient divergence in a pair of Atlantic Forest antbirds (Myrmeciza) F  ABIO RAPOSO DO AMARAL,* 1 PATRICK K. ALBERS, † 2 SCOTT V. EDWARDS † and CRISTINA Y. MIYAKI* *Departamento de Gen  etica e Biologia Evolutiva, Universidade de Sa ˜o Paulo, Rua do Mata ˜o, 277, Cidade Universit aria, Sa ˜o Paulo, SP CEP 05508-090, Brazil, †Department of Organismic and Evolutionary Biology and Museum of Comparative Zoology, Harvard University, 26 Oxford Street, Cambridge, MA 02138, USA Abstract The Atlantic Forest (AF) harbours one of the most diverse vertebrate faunas of the world, including 199 endemic species of birds. Understanding the evolutionary pro- cesses behind such diversity has become the focus of many recent, primarily single locus, phylogeographic studies. These studies suggest that isolation in forest refugia may have been a major mechanism promoting diversification, although there is also support for a role of riverine and geotectonic barriers, two sets of hypotheses that can best be tested with multilocus data. Here we combined multilocus data (one mtDNA marker and eight anonymous nuclear loci) from two species of parapatric antbirds, Myrmeciza loricata and M. squamosa, and Approximate Bayesian Computation to determine whether isolation in refugia explains current patterns of genetic variation and their status as independent evolutionary units. Patterns of population structure, differences in intraspecific levels of divergence and coalescent estimates of historical demography fit the predictions of a recently proposed model of refuge isolation in which climatic stability in the northern AF sustains higher diversity and demographic stability than in the southern AF. However, a pre-Pleistocene divergence associated with their abutting range limits in a region of past tectonic activity also suggests a role for rivers or geotectonic barriers. Little or no gene flow between these species suggests the development of reproductive barriers or competitive exclusion. Our results sug- gests that limited marker sampling in recent AF studies may compromise estimates of divergence times and historical demography, and we discuss the effects of such sam- pling on this and other studies. Keywords: anonymous loci, Atlantic Forest, multilocus phylogeography, refugia theory Received 4 January 2013; revision received 24 March 2013; accepted 28 March 2013 Introduction Neotropical forests are well known for their astonishing diversity, and defining which evolutionary processes originated such diversity has been the goal of evolu- tionary biologists for more than a century. Several early naturalists and modern biologists have sought to iden- tify patterns and processes, such as the presence of rivers, cycles of climate variation or geologic events, that may have contributed to the build-up of Neotropi- cal diversity (e.g. Wallace 1852; Haffer 1969, 1993, 1997; Bush 1994; Moritz et al. 2000; Bates et al. 2008; Rull 2008; Carnaval et al. 2009; Hoorn et al. 2010; Thom e et al. 2010). However, complexity seems to be the rule in the Neotropics, and it has become clear that a single model alone or a specific geological period cannot Correspondence: F abio Raposo do Amaral, Fax: +55 11 33196428; E-mail: fabioraposo@gmail.com 1 Current address: Departamento de Ci^ encias Biol ogicas, Uni- versidade Federal de Sa ˜o Paulo, Campus Diadema. Rua Prof. Artur Riedel, 275, Jardim Eldorado, Diadema, SP CEP 09972- 270, Brazil 2 Current address: Wellcome Trust Centre for Human Genetics, Roosevelt Drive, Oxford, OX3 7BN, UK © 2013 John Wiley & Sons Ltd Molecular Ecology (2013) doi: 10.1111/mec.12361