At the top of the mountain

Rise of the Andes explains the diversity of parrots in South America

WILLIAM COOPER, "PARROTS OF THE WORLD", 1973Distant relatives: Pionus corallinus (left) from the Andes, and P. menstruus, from the Amazon regionWILLIAM COOPER, "PARROTS OF THE WORLD", 1973

If one could point a video camera at the north of the Andes and show in just a few minutes what occurred over six million years, the film would show the mountains rising up high and taking with them some of the various parrots that were spread throughout the northern part of the continent. In the scenes corresponding to the last two million years, the colorful birds, already isolated from their relatives in the lowlands, would start to accrue differences amongst them until they became different species. This version of history, which goes against the most often accepted hypothesis, is the result of the work of biologist Camila Ribas from the Laboratory of Genetics and Molecular Evolution of Birds of the University of São Paulo (USP). She reconstructed the evolutionary history of the parrots of the Pionus genus with the help of biogeography, an area of specialization that analyzes the geographical distribution of biological diversity. This type of focus goes back a long way: it was biogeographical patterns that led Britons Charles Darwin and Alfred Russel Wallace to develop the theory of evolution.

A century and a half after Darwin’s and Wallace’s observations, biogeography now enjoys new techniques, such as the analysis of genetic material, which helps to tell the history of the Pionus parrots published this month in the Proceedings of the Royal Society, B, a British journal. Camila thought her work was modest, until she got to the American Museum of Natural History in New York for her post-doctoral work and showed her data to her supervisor Joel Cracraft. The experienced expert in bird evolution immediately recognized the value of her material in terms of elucidating the relationship between South American geological and evolutionary history and encouraged the Brazilian researcher to expand the sampling and the depth of the analyses.

More easily said than done. The Andean parrots are rare, hard to capture, and have not been studied much. In order to obtain samples of blood or other tissues from which to draw genetic material, one must find nestlings or shoot  adult birds. As a result, Camila decided to look for museum specimens and realized that she was in the right place: the New York museum has one of the most complete collections in the world, including the only preserved sample of Pionus ponsi, a parrot with dark green feathers, a bit bluer around the throat and yellower around the back, that was collected in 1949 in the northwest of Venezuela. It also has two rare Pionus saturatus skins, with their turquoise neck, captured in Colombia in 1899. Additionally, the laboratory in which the museum’s researchers conduct their genetic analyses combines conditions and knowledge such that it is among the best in the world where extracting genetic material from old specimens is concerned.

The genetic material drawn from the samples of US and Brazilian zoology museums enabled  the building of a genealogical tree – or phylogeny – that reveals the kinship between the Pionus species. Camila applied to this genealogy a method for estimating when the different species appeared. It is as if the quantity of differences between the DNA sequences in the two species, represented by the length of each branch, allowed one to calculate when the grandfather, great-grandfather and great-great-grandfather of a person living today was born. The idea of estimating divergence dates based on the length of the branches of a phylogenic tree is known as the molecular clock, but its high degree of imprecision means that the method is not always highly regarded by researchers. That is why Camila and her collaborators used a succession of analyses.

The first step consisted of estimating again the size of the branches of the phylogenetic trees, so as to reflect evolutionary time – the graphic representation of phylogeny includes a measurement axis, analogous to a map’s scale, that provides some idea of when each evolutionary event took place. The team compared two different methods that generated very similar results, making the estimates more reliable. The next step was to calibrate the tree: ascribe a known date to some point on it, after which it would be possible to infer the others. “To define this date, one has to have fossils whose age is known or geological events that can be linked to a ramification of the tree”, explained Camila. “But there are few psittacine fossils of the family that includes parrots, macaws and parakeets.” The only geological event she was sure about associated with the history of parrots took place about 85 million years ago, long before the Pionus genus arose: the separation of New Zealand and Antarctica. This left, on one side, a lineage that led to the Nestor genus, which only exists in New Zealand, and on the other side the source of all other psittacine birds. Based on this date, the researchers estimated that the Pionus appeared some 6.9 million years ago; this date was used to provide a time measuring scale for the genus’s genealogy.

This innovative methodology was successful. “The reviewers who evaluated the article approved its publication without questioning the method”, rejoiced the researcher. “Time estimates involve a large margin of error”, she explained, “but we are confident about the relative times”. In other words, the Pionus may not have appeared precisely 6.9 million years ago, but she knows the order of events over the course of the genealogy.

High differences
The molecular clock shows that when they rose, the Andes fragmented the distribution of parrot species, which over the course of some millions of years accrued differences and gave rise to new varieties. If the estimate is correct, the three species that existed some six million years ago, when the northern part of the Andes was 30% of its current height, evolved into ten different lineages: six in the highlands and four in the lowlands; the latter encompasses virtually all of the rest of South America.

It did not happen from one day to the next, but movements in the Earth’s crust brought about a huge range of mountains where previously there had been a forest plain. Thus, isolated groups of plants and animals appeared, such as maitacas (Pionus parrots); then local mechanisms came into play, increasing biological diversity.  In  the mountains, the climate change cycles were extreme: glaciers little by little engulfed the forest and reduced the areas the parrots could inhabit to limited patches; the glaciers then melted, allowing the birds to spread again. These processes were repeated several times and, over the last million years, they gave rise to most of the Andean maitaca parrots. Today there are ten different species , according to Camila’s work.

The uneven terrain and glacial cycles lead many people to imagine that the evolutionary process in the Andean highlands is more complex than in the lowlands. As an example, Camila mentioned Jason Weir, a Canadian who published  an article last year in Evolution, a prestigious American journal, in which he concluded that the most recent species of South-American birds are to be found on the mountain tops. “The problem is that he used a classification that doesn’t represent true diversity”, retorted the Brazilian researcher. The nine species of maitaca parrots living in the lowlands are, according to Camila’s dating system, equally recent: most also arose during the last million years.

WILLIAM COOPER, "PARROTS OF THE WORLD", 1973Clockwise: P. seniloides, from the lowlands, and chalcopterus, senilis and tumultuosus, from the AndesWILLIAM COOPER, "PARROTS OF THE WORLD", 1973

A difficulty this type of study raises is that we know very little about Brazilian biodiversity: how many species exist, what is their kinship and where they occur. In the work with the Pionus, Camila treated as different species several types of parrots that current taxonomy regards as subspecies. “They are very different animals and live in areas that are clearly separate geographically”, she explained. Now experts must decide how many species of maitaca parrots actually exist: the nine recognized today, the 19 the Camila and her co-authors consider distinct, or an intermediate number. “The Brazilian Committee of Ornithological Registries has asked to analyze the article”, the biologist told us. In Brazil, it is still impossible to answer more elaborate biological questions without  organizing the classification of the species first, according to Camila, who also had to become a systematist, i.e., an expert in the field of biology that focuses on classifying living beings according to their degree of kinship.

However, she acquired a taste for this. During the course of her doctorate, which she completed in 2004 under the guidance of Cristina Miyaki, at USP, Camila ordered the classification of several parrot genera. She examined the nine species normally considered as belonging to the genus Pionopsitta, which are distributed in northern South America, and discovered that the classification did not correspond to reality. It is as if a family’s genealogical tree were to include second cousins but ignore first cousins. A remodeling was required. In an article published in 2005 in the Journal of Biogeography, Camila resuscitated the genus Gypopsitta, which had ceased to be used, and included in it eight species of these bright green parrots with a head that is either red, yellow or green with colored spots, generally known as curicas. The Pionospsitta genus was left with a single species, known as pileata, cuiú-cuiú or caturra (red-capped parrot), with its red mask.

The biologist also recovered the history of the Gypopsitta that, like the maitaca parrot, had an evolutionary history derived from the rise of the Andes. The group that remained to the west of the mountain range generated three species that now live in Central America, Colombia and Ecuador. After this, geological events, probably connected with the movements of the Earth’s crust that produced the Andes, separated the Amazonian curicas and gave rise to western species – G. barrabandi, along the Amazon basin all the way to Peru, and G. pyrilia, in Guyana’s. The Gypopsitta representatives that remained in the eastern part of the Amazon region split into three species, which may have occurred as a result of sea level fluctuations and glaciation.

Forest from the past
The same approach can be valuable to find out past and present links between Brazilian ecosystems. Part of the biogeographical history of Brazil is recorded in the Pyrrhura parakeets, known as tiribas (a type of parakeet), another type of psittacine bird that Camila studied during her doctorate. Like maitacas, different species of tiribas can be found in the Andes and in the Amazon, Cerrado (savannah), Caatinga (brush) and Mata Atlântica (Atlantic rainforest) regions. When studying kinship among these species, Camila concluded that the ancestor of these parakeets gave rise to a branch that led to the Pyrrhura cruentata, which now inhabits the Mata Atlântica forest, and to another type that then diversified into all the other species. This second lineage, in turn, branched out and generated  several species that now inhabit various South American habitats. Contrary to what one commonly observes, the Pyrrhura species that now share a given environment  are not closely related; to the contrary, they represent lineages that diverged in the history of tiribas’s distant past. This shows, for instance, that not all species currently found in the Mata Atlântica had their evolutionary roots in that forest. Camila’s work, published in 2006 in The Auk, a specialized journal, suggests that the Mata Atlântica fauna is comprised of species whose ancestors were from there, and of other species of Amazonian origin. Studies about other animals indicate the same: the Mata Atlântica and the Amazon region were not always separate from each other as they are now. “At a recent point in time, about one million years ago, there seems to have been communication between the Amazon region and the Mata Atlântica forest, through forest corridors that existed where we now have the Cerrado and the Caatinga areas”, summarized Camila.

Moreover, the researcher showed that the Mata Atlântica rainforest has some groups that are very recent and others that are very old. The old groups appear in phylogenies  in the form of long branches with no ramifications; in other words, they have no sister-species of relatively recent origin. “This suggests that there may have been a lot of extinction around there, or fewer opportunities for diversification”, she explained. However, the presence of long branches – lineages that have existed in the Mata Atlântica for millions for millions of years – shows that the forest has been a stable environment for longer than the Amazon forest, where fairly recent environmental variations caused most of the parrots that live there to diversify over the last one or two million years. This process gave rise to species considered to be young.

Camila’s next step will be to move beyond psittacine birds and to study those that have different histories and that might help us to understand better how Brazilian forests and their biodiversity were formed. She is starting with the Amazon region and chose birds that reflect the importance of taking into account the ecological particularities of each species. The Psophia or jacamim (trumpeter) is a land bird with a short tail and dark feathers, and it is limited to Amazonian drylands – it does not live in swampy areas. This specialization seems to limit their movement, which would not be the case for parrots,  capable of flying over long distances. The result is that different Amazon regions are the home of different species of Psophias, whose diversification is recent. One must still explain what isolated the lineages and gave rise to the different species. As for the arapaçus (woodcreepers), birds with brown feathers that can get to the insects that live under the bark of trees with their long beaks, they have different ecological habits depending on the species. The Dendrocincla merula, like the Psophia, is more restricted to drylands. Camila is now taking part in a study coordinated by Alexandre Aleixo, from the Emílio Goeldi Museum in Pará, together with researchers from the Federal University of Pará, which has already shown that there are lineages separated by the large rivers of the Amazon region. This, however, does not seem to apply to the Dendrocincla fuliginosa, which is more flexible in habitat terms: a preliminary analysis indicates that the distribution of lineages is broader and encompasses large tracts of forest.

There is still a lot to learn in order to understand how past and current features – rivers, mountains, geological movements and climate changes, among others, shaped the animal and vegetable species that live there. Brazilian biological diversity carries marks that may reveal some of the mysteries surrounding the formation of South America, but biologists and geologists still have a lot of work ahead of them before they can decipher this history, which is important not only to understand how this part of the world was formed but also to draw up conservation strategies for South America’s very rich and unique fauna.

Reconstruction of the evolutionary history and phylogeographical studies of neotropical birds using molecular markers (nº 03/14106-3); Modality: Thematic Project; Coordinator: Cristina Yumi Miyaki – USP; Investment: R$ 507,359.46 (FAPESP)