Phytogeographical relations of the Andean dry valleys of Bolivia

Aim The objective of this study is to examine the phytogeographical affinities of the Andean dry valleys of Bolivia in order to contribute to a better understanding of the Andean dry flora's distribution, origin and diversity. Particular emphasis is given to the analysis of the floristic connections of this flora with more austral parts of South America. Location The dry valleys of Bolivia are located in the Andes of the southern half of the country, at elevations between 1300 and 3200 m. Methods An extensive floristic list compiled by the author to evaluate plant diversity in these Andean regions was used as the base for this study. To accomplish this, all recorded genera and species were assigned, respectively, to 11 and 12 phytogeographical elements established previously by the author. Two phytogeographical spectra were thus obtained and analysed. Results At the genus level, the Andean dry valleys of Bolivia are clearly dominated by genera that have widespread distributions (cosmoplitan and subtropical genera). Many of these reached the Andes from the lowland region of the Chaco. At species level, Andean elements constitute more than 60% of the species total, most of which are restricted to the central‐southern Andes. This suggests that Chaco‐related and Andean genera had considerable levels of speciation in these valleys. Many genera and more than half the species have their northernmost distribution in the dry valleys of Bolivia, thereby underlining strong relationships with central‐southern South America (mainly Argentina, Paraguay and southern Brazil). The data supports the belief of the existence, in central‐southern Peru, of a floristic disjunction in dry to arid environments that separates a tropical dry flora north of this limit from a dry subtropical/warm temperate flora south of it. Main conclusions The Andean dry valleys of Bolivia are diverse plant communities with high levels of endemism (c. 18% of the species). The species of this region are more related to those present in central‐southern South America than to the flora of northern South America that ranges southwards to Peru. Many of the species have restricted distributions in the dry Andes of Bolivia and Argentina, and many genera of these dry valleys have their northernmost distribution in Bolivia/southern Peru, too. The data point to high levels of speciation also in the central Andes.     

The lichen genus Umbilicaria in Ecuador

Nine species of the saxicolous lichen genus Umbilicaria from Ecuador are reported and a key is provided: U. africana, U. aprina, U. cinereorufescens, U. decussata, U. dendrophora, U. haplocarpa, U. leprosa, U. nylanderiana and U. vellea. The species diversity of this genus on the equatorial high mountains of South America is low compared to North America and Eurasia, but similar to that found on high African peaks close to the equator. The species mostly belong to a high‐alpine element with worldwide distribution. Two species belong to an Andean endemic element, viz U. haplocarpa and U. leprosa. The low diversity and low percentage of endemism may reflect the fairly recent uplift of the Andes and the comparatively small geographic extent of the alpine Andean biota. The dense rainforests of Ecuador leave few suitable open rock habitats for the establishment of these light‐craving lichens below the tree‐line (4000–4200 m a.s.l.), and thus their equatorial habitat is almost exclusively restricted to the alpine zone between the tree‐line and the snow‐line (4700–4800 m a.s.l.). The equatorial species mainly reproduce asexually by thalloconidia, and the adaptive significance of this type of reproduction in high altitude habitats is discussed.     

A revision of the brassoline genus Catoblepia (Lepidoptera: Rhopalocera)

The brassoline genus Catoblepia is revised: eight new subspecies are described, nine new synonymies established, the status of four species and subspecies is revised, one species ( singularis ) is returned to the genus Opsiphanes and ten lectotypes are designated. All known holotypes (20) and lectotypes (10) are figured, many for the first time. The refined taxonomy reveals an interesting pairing of the Andean subspecies of xanthicles and berecynthia and it is suggested that diese may be mimetic pairs. There is close agreement in the distribution of the Andean species and subspecies with die endemic centres previously recognized; east of the Andes, however, diere is no close correspondence and each of the subspecies has, in general, a distribution across several endemic centres.     

Patterns of diversification in the high Andean Ponderacris grasshoppers (Orthoptera: Acrididae: Melanoplinae)

The Andes, the world's longest mountain chain, harbours great taxonomic and ecological diversity. Despite their young age, the tropical Andes are highly diverse due to recent geological uplift. Speciation either followed the orogeny closely or occurred after the Andean uplift, as a result of subsequent climatic changes. Different scenarios have been proposed to explain the diversification of high Andean taxa. The Melanoplinae grasshopper Ponderacris Ronderos & Cigliano is endemic to the eastern slopes of the Andes of Peru and Bolivia, mostly distributed between 1000 and 4000 m above sea level. Diversification in several montane habitats of Bolivia and Peru allows tests via cladistic analysis of distinct possible geographic modes of speciation. Eight species are recognized, with three described here as new with revised diagnostic morphological characters provided: Ponderacris carlcarbonelli sp.n., P. chulumaniensis sp.n. and P. amboroensis sp.n. Cladistic analyses of 15 species (8 ingroup and 7 outgroup) and 38 morphological characters, under equal and implied weighting, confirm the monophyly of Ponderacris. Characters from the external morphology and colour pattern provided less phylogenetic information than did the male abdominal terminalia and phallic complex. Species distributed in the Peruvian Andes constituted a monophyletic group, whereas those from the Bolivian Andes formed a basal paraphyletic grade. Dispersal–vicariance analysis resulted in one ancestral distribution reconstruction indicating that the most recent common ancestor was distributed in the Lower Montane Yungas of Bolivia. Eleven dispersal and one vicariant events are postulated, with a South‐to‐North speciation pattern coincident with progressive Andean uplift. Vicariance could relate to fragmentation of montane forest during the dry intervals of the late Cenozoic. From the Bolivian area, ancestral Peruvian Ponderacris may have dispersed northward, coinciding with the rise of the Andes. Ten of 11 dispersal events occurred at terminal taxa and are likely to be recent. However, diversification of Ponderacris cannot be explained solely by the South‐to‐North speciation hypothesis, but may also include both vicariance and dispersal across barriers influenced by Pleistocene climatic cycles.     

An estimation of the floristic richness of Bolivia’s Andean dry valleys

We estimate plant species richness of Bolivian Andean dry valleys applying the species/genus ratio of Compositae to all reported genera for those regions. The results suggest the existence of around 1500 species in the Bolivian dry valleys (12.6% more than the species recorded). A brief biogeographical analysis of our results suggests that the use of species/genus ratio of Compositae to estimate the size of the flora in Bolivian Andean dry valleys should be considered with caution, since the central Andes are geologically very young, and explosive speciation processes have taken place only in some taxa (mainly in Bromeliaceae and Cactaceae). We offer some evidence pointing to the possibility that the Andean dry valleys may possess many more species than suggested by the Compositae species/genus ratio. We discuss the implications of our estimation for conservation of the Andean dry valleys, which only represent 0.7% of the protected area of Bolivia.     

Frogs at the summits: phylogeny of the Andean frogs of the genus Telmatobius (Anura,Telmatobiidae) based on phenotypic characters

A phylogenetic hypothesis for the frogs of the genus Telmatobius that includes a comprehensive sample of the morphological and geographical variation is lacking. Obtaining such a hypothesis constitutes the main focus of this contribution. A phylogenetic matrix was generated based on 97 phenotypic characters and 56 terminals. A parsimony analysis of this matrix was performed with TNT. Telmatobius is found to be monophyletic and well supported by 11 synapomorphies. Although the consensus tree shows several polytomies, four main groups have been recovered. The well‐supported T. verrucosus Group includes forest and sub‐paramo species from Bolivia and Peru, and is the sister group of the remaining species. The T. bolivianus Group includes forest and inter‐Andean valley species from Argentina and Bolivia but it is poorly supported. Two supported high‐altitude groups have been recovered, the T. macrostomus Group from the Central Andes of Peru, and the T. marmoratus Group from the Altiplano‐Puna Plateau of Argentina, Bolivia, Peru and Chile and its adjacent Pacific and Northern slopes. The synapomorphies proposed for Telmatobius are discussed as well as the evolution of some of these synapomorphies and other characters within the genus.     

Biogeography of theOxalis tuberosa alliance

TheOxalis tuberosa alliance is a group of morphologically similarOxalis species allied to the Andean tuber crop oca,O. tuberosa. Originally described by cytologists as a dozen species sharing a base chromosome number rare inOxalis (x = 8), the alliance as defined here includes additional species for which cytological information is not yet available but which are supported as members on molecular and/or morphological grounds. The alliance includes members found in the Andean region from Venezuela to northern Argentina, with one species at high elevations in Central America. They occur from the high Andean steppes (páramo and puna) to the cloud forests of middle elevations and include both restricted endemics and variable widespread species complexes. Geographical and altitudinal distributions of members of the alliance and selectedOxalis species outside the alliance were compared with a combined phylogenetic analysis of DNA sequence data of ITS and ncpGS (chloroplast-expressed glutamine synthetase). Groups within the alliance (i.e., major clades on the molecular trees) occur across widespread, overlapping regions in the Andes, with only partial ecological separation. The hypothesis that theO. tuberosa alliance may have developed in the Andes of southern Peru and northwestern Bolivia and radiated southward and, especially, northward along the Andean axis is suggested by patterns of distributions of members of the alliance and outgroups. In spite of uncertain species delimitations, it is clear that the alliance includes many endemic species and ecotypes that have very restricted distributions. As relatives of the Andean tuber cropOxalis tuberosa, the genetic diversity represented by this geographical variability should be a high priority for conservation.     

Taxonomic revision of the olingos (Bassaricyon), with description of a new species,the Olinguito

We present the first comprehensive taxonomic revision and review the biology of the olingos, the endemic Neotropical procyonid genus Bassaricyon, based on most specimens available in museums, and with data derived from anatomy, morphometrics, mitochondrial and nuclear DNA, field observations, and geographic range modeling. Species of Bassaricyon are primarily forest-living, arboreal, nocturnal, frugivorous, and solitary, and have one young at a time. We demonstrate that four olingo species can be recognized, including a Central American species (Bassaricyon gabbii), lowland species with eastern, cis-Andean (Bassaricyon alleni) and western, trans-Andean (Bassaricyon medius) distributions, and a species endemic to cloud forests in the Andes. The oldest evolutionary divergence in the genus is between this last species, endemic to the Andes of Colombia and Ecuador, and all other species, which occur in lower elevation habitats. Surprisingly, this Andean endemic species, which we call the Olinguito, has never been previously described; it represents a new species in the order Carnivora and is the smallest living member of the family Procyonidae. We report on the biology of this new species based on information from museum specimens, niche modeling, and fieldwork in western Ecuador, and describe four Olinguito subspecies based on morphological distinctions across different regions of the Northern Andes.     

Andean uplift promotes lowland speciation through vicariance and dispersal in Dendrocincla woodcreepers

Andean uplift contributed importantly to the build-up of high Neotropical diversity. Final uplift of the Eastern Cordillera of Colombia separated once-contiguous lowland faunas east and west of the Andes between 5 and 3.5 million years ago (Ma hereafter). We used DNA sequences from several moderate- to fast-evolving mitochondrial and two slow-evolving nuclear genes to generate a well-supported phylogeny of Dendrocincla woodcreepers, a genus with multiple species endemic to lowland regions both east and west of the Andes. A time-calibrated phylogeny and dispersal-vicariance analysis indicated that uplift of the Eastern Cordillera of Colombia resulted in the initial vicariant separation of a widespread lowland form east and west of the Andes at c. 3.6 Ma. This was followed by two separate east-to-west dispersal events over or around the completed Andes, each producing a genetically distinct lineage. Our analysis suggests that Andean uplift promoted the build-up of biodiversity in lowland Neotropical faunas both through vicariance-based speciation during uplift and through dispersal-based speciation following uplift. In contrast to the multiple colonizations of the trans-Andean region by Dendrocincla, the Atlantic Forest was colonized from the Amazon only once, followed by in situ diversification.     

Monograph of the neotropical species of Callipteris with anastomosing veins (Woodsiaceae)

Callipteris is a pantropical fern genus defined by two synapomorphies of its rhizome scales: 1) dark-castaneous to black borders, and 2) bifid marginal teeth. This monograph treats 15 neotropical species ofCallipteris that are further defined by the synapomorphy of anastomosing veins (free-veined species of the genus occur in the Neotropics and elsewhere). Three species are newly described here; the remaining 12 were previously classified inDiplazium. All the species ofCallipteris grow on wet forest floors at 100–2300 m. In the Neotropics the genus occurs from Guatemala to Bolivia, northern Brazil, and the Lesser Antilles. It has two centers of species richness. The first is the Chocó region on the western side of the Andes in Colombia and northwestern Ecuador. This region contains eight species, five of which are endemic. The second center is the mountains of Costa Rica and Panama, where six species occur, three of which are endemic. Only one species, an endemic, occurs in the Lesser Antilles in Martinique, St. Vincent, and Grenada. Four species occur only on the eastern side of the Andes, and these apparently represent at least two separate dispersal events from the western side of the Andes.     

Population status of Andean Condors in central and southern Bolivia

Andean Condors (Vultur gryphus) are a Near Threatened species that was formerly distributed along the entire length of the Andes from western Venezuela to Tierra del Fuego. Populations have been severely reduced north of Peru, but several thousand Andean Condors still exist in the southern portion of their range in Argentina and Chile. Little is known, however, about the size of the Andean Condor population in the central part of their range in Peru and Bolivia. From June to September 2012, we used feeding stations to attract Andean Condors and estimate the size and structure of the population in the eastern Andes of central and southern Bolivia. We estimated a minimum population of 253 condors, an adult male‐to‐female ratio of 1:0.6, an immature male‐to‐female ratio of 1:0.9, and an adult‐to‐immature ratio of 1:1.1. At our five survey areas, estimated abundance ranged from 15 to 100 condors per area. Males outnumbered females in three areas and the opposite was true in two areas. Our estimated adult‐to‐immature ratio, overall and in each area, suggests that the populations could be reproducing at a high rate. As previously observed in other Andean Condor populations, skewed sex ratios could be associated with differences between sexes and age classes in habitat selection. Although our results suggest that Bolivian populations of Andean Condors are still reasonably large, population monitoring is urgently needed, including use of feeding stations throughout the entire Bolivian range of the species and intensive searches for roosting and nesting sites.     

Evolution in the high Andes: the phylogenetics of Muscisaxicola ground-tyrants

Phylogenetic relationships within the genus Muscisaxicola, a primarily Andean group of tyrant-flycatchers, were studied using complete sequences of the mitochondrial genes COII and ND3. Relationships among Muscisaxicola species were found to differ substantially from those of previous views, suggesting convergence in traditional avian taxonomic characters within the genus. The 11 species of large, gray, "typical" Muscisaxicola flycatchers (including M. grisea, newly restored to species status) formed a distinct clade, consisting of two major groups: a clade of 6 species breeding primarily in the central Andes and a clade of 5 species breeding primarily in the southern Andes. The other 2 species traditionally placed in this genus, M. fluviatilis, an Amazonian species, and M. maculirostris, were both rather divergent genetically from the typical species, although M. maculirostris may be the sister taxon to the typical clade. The patterns of sympatry exhibited by Muscisaxicola species in the high Andes appear to be the consequence of speciation and secondary contact within regions of the Andes, rather than a result of dispersal between regions. Diversification of the typical Muscisaxicola species appears to have occurred during the middle and late Pleistocene, suggesting generally that taxa of the high Andes and Patagonia may have been greatly influenced by mid-to-late Pleistocene events. There were likely several independent developments of migration within this genus, but migration is probably ancestral in the southern clade, with subsequent loss of migration in two taxa.     

Batrachochytrium dendrobatidis in the live frog trade of Telmatobius (Anura: Ceratophryidae) in the tropical Andes

Species of frogs in the genus Telmatobius are traded and sold for human consumption in the Andes and in coastal cities of Peru and Bolivia. These frogs are harvested from wild populations. We report high prevalence of infection by the pathogenic fungus Batrachochytrium dendrobatidis in live frogs purchased at the main market in Cusco, Peru, from January 2008 to January 2010. We suggest that the transport of native anurans through the live frog trade could facilitate the spread of this fungus among Andean frogs. The tropical Andes are the most important biodiversity hotspot for amphibians. Because many neotropical taxa are known to be susceptible to chytridiomycosis, the presence of a large reservoir of infection in the frog trade poses a significant threat to amphibian conservation.     

Biological diversification in a complex region: a spatial analysis of faunistic diversity and biogeography of the Andes of Colombia

Aim Understanding large‐scale patterns of beta diversity and endemism is essential for ecoregional conservation planning. We present a study of spatial patterns of faunal diversification and biogeographical relationships in the Andean region of Colombia. This region has a great geomorphological complexity, as it is formed by several mountain ranges with different geologic origins. We hypothesize that this complexity results in a high turnover in species composition among subregions. Location The Andean region of Colombia, including the Santa Marta and Macarena mountain ranges. Methods The region was divided into subregions, represented by the eastern and western slopes of each of the three Andean cordilleras, the Cauca and Magdalena valley bottoms, and the peripheral mountain ranges of Perijá, Macarena and Sierra Nevada de Santa Marta. Species lists for five animal taxa (rodents, bats, birds, frogs and butterflies) were compiled for each subregion and similarities in species composition were determined by cluster analysis. To explore biogeographical relationships, species were classified into one of four distributional categories: endemic, tropical Andean, Andean‐Central American and wide continental distribution. Results The highest species richness in the region was found in the Pacific and eastern versants of the Andes, and the lowest in the Cauca and Magdalena valley bottoms. Inter‐Andean slopes were intermediate in species richness. However, when species richness was calculated per unit area, the most diverse regions were the Santa Marta and Macarena ranges, the Cauca Valley watershed and the Pacific slope. Although each taxonomic group had a different branching pattern, dendrograms indicated five common subregional clusterings: (1) Perijá‐Sierra Nevada, (2) the Pacific slope, (3) the eastern Andean slope, (4) the Cauca and Magdalena valley bottoms, and (5) the inter‐Andean slopes. Clustering patterns of inter‐Andean slopes varied among taxa. In birds, bats and rodents, grouping was by opposite slopes of the same valley, whereas frogs were grouped by mountain ranges and butterflies by valleys and their respective slopes. Seventy‐five per cent of species in all taxa were found in less than five subregions. The fauna of the Magdalena and Cauca valley bottoms was composed mostly of lowland species with wide geographical distributions, whereas the cordilleran fauna was mostly restricted to the tropical Andes. Main conclusions The western and eastern versants of the Andes have the highest species richness, but are also the largest subregions. On a per unit area basis, the peripheral ranges (Santa Marta and Macarena) are the richest, followed by the western portion of the Andes (the Cauca Valley watershed and the Pacific versant). Clustering patterns in dendrograms suggest two major patterns of differentiation of the Andean fauna: one elevational (lowlands vs. highlands) and one horizontal (among ranges and/or slopes). Biogeographical affinities of the inter‐Andean valley bottoms are with the lowland faunas of tropical America. In contrast, Andean faunas diversified locally, resulting in the evolution of a large number of endemic species, particularly among the less vagile taxa. Three different main branches of Andean fauna can be recognized, one confined to the Pacific, another to the eastern (Amazonian‐Llanos) versant of the Andes, and the third one composed by the inter‐Andean slopes of the Cauca and Magdalena valleys. The identification of five main biogeographical units in the Andean region of Colombia has important implications for the conservation of the regional biota. Conservation initiatives that seek to preserve representative samples of the regional biodiversity should take into account the patterns of diversification described here, and the evolutionary processes that gave rise to these patterns.     

Evolutionary islands in the Andes: persistence and isolation explain high endemism in Andean dry tropical forests

Aim The tropical Andes are a world biodiversity hotspot. With diverse biomes and dramatic, geologically recent mountain uplift, they offer a system to study the relative contributions of geological and biome history to species richness. There are preliminary indications that historical species assembly in the Andes has been influenced by physiographical heterogeneity and that distinct biomes have evolved in relative isolation despite physical proximity. Here we test this ‘Andean biotic separation hypothesis’ by focusing on the low‐elevation, seasonally dry tropical forest (SDTF) biome to determine whether patterns of plant diversification within the SDTF differ from those in mid‐ and high‐elevation biomes. Location Tropical Andes, South America. Methods Densely sampled time‐calibrated phylogenies for five legume genera (Amicia, Coursetia, Cyathostegia, Mimosa and Poissonia) containing species endemic to the Andean SDTF biome were used to investigate divergence times and levels of geographical structure. Geographical structure was measured using isolation‐by‐distance methods. Meta‐analysis of time‐calibrated phylogenies of Andean plant groups was used to compare the pattern and tempo of endemic species diversification between the major Andean biomes. Results Long‐term persistence of SDTF in the Andes is suggested by old stem ages (5–27 Ma) of endemic genera/clades within genera, and deep divergences coupled with strong geographical structure among and within species. Comparison of species diversification patterns among different biomes shows that the relatively old, geographically confined pattern of species diversification in SDTF contrasts with the high‐elevation grasslands that show rapid and recent radiations driven by ecological opportunities. Main conclusions The SDTF biome has a long history in the Andes. We suggest that the diverse SDTF flora has been assembled gradually over the past c. 19 Ma from lineages exhibiting strong phylogenetic niche conservatism. These patterns suggest that Andean SDTFs have formed stable and strongly isolated ‘islands’ despite the upheavals of Andean uplift. Indeed, the Andean SDTFs may represent some of the most isolated and evolutionarily persistent continental plant communities, similar in many respects to floras of remote oceanic islands.     

Nutritional evaluation of three underexploited andean tubers:Oxalis tuberosa (Oxalidaceae),Ullucus tuberosus (Basellaceae), andTropaeolum tuberosum (Tropaeolaceae)

Three species of edible tubers endemic to and domesticated in the Andes were studied for their nutritional value. Collected samples ofOxalis tuberosa, Ullucus tuberosus, andTropaeolum tuberosum show a high amount of variation in both percent protein and quality of essential amino acids. A protein difference of120% is present amongT. tuberosum cultivars and a protein difference of 300% is present among the three species. The data indicate that previous published Andean tuber crop food values may need revision. The introduction of "improved" crop varieties and less nutritious foodstuffs threatens the base ofcultivar diversity that has been selected by Andean agriculturists over centuries. This rapid erosion of Andean tuber diversity indicates the importance of identifying and conserving Andean tuber cul-tivars throughout the Andes.     

A south-to-north biogeographic hypothesis for Andean speciation: evidence from the lizard genus Proctoporus (Reptilia, Gymnophthalmidae)

Aim The lizard genus Proctoporus Tschudi, 1845 was used as a model to test the South‐to‐North Speciation Hypothesis (SNSH) for species groups occurring in the Andes Mountains of South America. This hypothesis proposes that speciation of high Andean taxa followed a south‐to‐north pattern, generally coinciding with the progression of final uplift of the Andes. According to SNSH, a phylogenetic hypothesis of relationships of a taxonomic group occurring in the high Andes would show a branching pattern in which the southernmost species diverged first, followed by the more northern species, and so on in a northerly pattern. Location The central and northern Andes Mountains in South America. Methods A phylogenetic hypothesis was reconstructed for all species of the lizard genus Proctoporus by examining the external morphology of 341 individuals. This phylogeny was then examined to determine monophyly of the genus, distribution patterns of species groups, and congruence with SNSH. Results The genus Proctoporus did appear to be monophyletic and, therefore, it was valid to use this group to assess SNSH. The southernmost species were found to be the most basal, which was consistent with SNSH. The species occurring in the northern Andes did not exactly match the SNSH prediction. The Venezuelan and Trinidadian species did appear to be highly derived, as predicted by the hypothesis, but the Ecuadorian and Colombian species did not form a particular pattern in relation to the hypothesis. Main conclusions The SNSH does appear to have predictive power with regard to large‐scale distribution patterns. The finer‐scale patterns of speciation in the Andes, however, appear to be a more complex phenomenon that cannot be fully explained by a simple hypothesis. It is important to have a testable hypothesis in hand with which to compare data from disparate species groups. The incorporation of phylogenetic data of other high Andean taxa with similar distribution patterns is necessary to determine the full utility of SNSH in explaining evolutionary patterns in the Andes of South America.     

Areas of endemism and distribution patterns for Neotropical Piper species (Piperaceae)

Aim The study aimed to establish areas of endemism and distribution patterns for Neotropical species of the genus Piper in the Neotropical and Andean regions by means of parsimony analysis of endemicity (PAE) and track‐compatibility analysis. Location The study area includes the Neotropical region and the Northern Andean region (Páramo‐Punan subregion). Methods We used distribution information from herbarium specimens and recent monographic revisions for 1152 species of Piper from the Neotropics. First, a PAE was attempted in order to delimit the areas of endemism. Second, we performed a track‐compatibility analysis to establish distribution patterns for Neotropical species of Piper. Terminology for grouping Piper is based on recent phylogenetic analyses. Results The PAE yielded 104 small endemic areas for the genus Piper, 80 of which are in the Caribbean, Amazonian and Paranensis subregions of the Neotropical region, and 24 in the Páramo‐Punan subregion of the Andean region. Track‐compatibility analysis revealed 26 generalized tracks, one in the Páramo‐Punan subregion (Andean region), 19 in the Neotropical region, and six connecting the Andean and Neotropical regions. Both the generalized tracks and endemic areas indicate that distribution of Piper species is restricted to forest areas in the Andes, Amazonia, Chocó, Central America, the Guayana Shield and the Brazilian Atlantic coast. Main conclusions Piper should not be considered an Andean‐centred group as it represents two large species components with distributions centred in the Amazonian and Andean regions. Furthermore, areas of greater species richness and/or endemism are restricted to lowland habitats belonging to the Neotropical region. The distribution patterns of Neotropical species of Piper could be explained by recent events in the Neotropical region, as is the case for the track connecting Chocó and Central America, where most of the species rich groups of the genus are found. Two kinds of event could explain the biogeography of a large part of the Piper taxa with Andean–Amazonian distribution: pre‐Andean and post‐Andean events.     

Inferring the diversification of the epiphytic fern genus Serpocaulon (Polypodiaceae) in South America using chloroplast sequences and amplified fragment length polymorphisms

In this study, we infer the phylogeny of the recently described epiphytic fern genus Serpocaulon. Four regions of the plastid genome were sequenced for 68 samples, representing 31 of ca. 40 currently accepted species of this genus. The reconstructed phylogeny supports most of the previously proposed clades, but more exhaustive studies are needed to improve species delimitation in several terminal clades. A further objective of this study was to determine the utility of amplified fragment length polymorphism (AFLP) data to study the diversification of species complexes. Independent analyses of data sets based on chloroplast DNA sequences or AFLPs resulted in phylogenetic trees with similar topologies, but showed also some notable differences. We present an explicit hypothesis of the biogeographic history of Serpocaulon. All reconstructed phylogenies suggest an origin of this genus in the Bolivian–Brazilian region and indicate a major role of the Bolivian Andes as a stepping-stone in the colonization of northern regions of the Andes. The majority of the extant species diversity of this almost exclusively epiphytic fern clade is likely the result of an adaptive radiation that was triggered by the colonization of Andean mountain forest habitats (above 2,000 m). This is the first report of a Bolivian origin of fern diversification in Andean mountain habitats using phylogenetic evidence.     

Variability in Andeannuña common beans (Phaseolus vulgaris,Fabaceae)

Nuñas are a group of common bean (Phaseolus vulgaris, Fabaceae) varieties, whose grains are consumed after toasting in their original Andean habitats. Nowadays, these varieties are restricted to certain parts of the highlands of Peru and Bolivia. Linguistic, ethnobotanical, and archaeological data suggest that they were grown in that zone duringpre-Hispanic times. SDS-PAGE electrophoresis shows a wide range ofphaseolin types amongnuñas, many of which are present in sympatric wild forms. This paper discusses the possibility thatnuñas resulted from an early and widely applied selection pressure during bean domestication in the Andes.