E v o l u t i o n o f t h e B i o d i v e r s i t y H o t s p o t o f M a d a g a s c a r f r o m t h e E y e o f D i v i n g B e e t l e s – p h y l o g e n y , c o l o n i -z a t i o n a n d s p e c i a t i o n
Evolution of the Biodiversity Hotspot
of Madagascar from the Eye of
Div-ing Beetles
Phylogeny, colonization and speciation
Rasa Bukontaite
©Rasa Bukontaite, Stockholm University 2015 ISBN 978-91-7649-223-9
Cover Illustration Kim Taylor/Warren Photographic, with modifications by Rasa Bukontaite
Printed in Sweden by Holmbergs, Malmö 2015
Why diving beetles? The ... “answer is simple: They are cool!” Donald A. Yee
Abstract
Diving beetles, Dytiscidae, are distributed worldwide and can be found in both permanent and temporary, standing and running, water bodies. Dytis-cidae contains numerous endemic and non-endemic species on Madagascar. However, their evolutionary history is largely unknown on the island. Here-in, I use molecular data and analyses to infer phylogenetic relationship among groups of diving beetles, with a focus on the subfamily Dytiscinae and endemic species in two other groups of Dytiscidae on Madagascar. Pa-per I represents the first phylogenetic reconstruction focusing on the tribe Aciliini based on molecular data. Several commonly used molecular mark-ers, as well as a new marker for Hydradephagan beetles, were evaluated in this study. Our analyses confirmed monophyly of Aciliini with Eretini as its sister group. Each of the six genera within these tribes were also supported as monophyletic. The most basal clades with Neotropical and Afrotropical taxa suggest a possible Gondwanan origin, but this was dependent on the position of a fossil calibration point. Evaluation of gene fragments indicated CAD to be the most informative marker followed by another nuclear protein coding gene, WNT. Paper II focuses on colonization and radiation events of large bodied endemic diving beetles on Madagascar. This study was based on two previously published datasets, where we added new taxa from Mada-gascar. Colonization events of the tribes Cybistrini and Hydaticini were in-ferred from dated phylogenetic trees and ancestral biogeographical recon-structions. Our results suggest both multiple colonizations, and out-of-Madagascar dispersal events, mostly during the Miocene and Oligocene. In paper III, we revised the Rhantus species of Madagascar, which are all re-stricted to the central highland plateau. We used both molecular and mor-phological data to evaluate species hypothesis and emphasized the value of Manjakatompo – one of the last remaining fragments of central highland forests. In Paper IV we reconstruct the phylogeny and use Species Distribu-tion Modelling (SDM) for the endemic river-dwelling genus Pachynectes in Madagascar. Our sampling has discovered that the species diversity of
Pach-ynectes is at least three times higher than previously believed. Integrating
phylogeny with SDM we aim to better understand the mode of speciation within endemic radiations. It seems that allopatric speciation was the main driver, which led to the diversity of Pachynectes. Moreover, we tested two main hypotheses explaining microendemic patterns of distribution on Mada-gascar. Our results suggest that climatic gradients and the five main biomes
were a better predictor than rivers and watershed systems in explaining the distribution pattern and speciation between sister species.
Keywords: Dytiscidae, diving beetles, Biogeography, Phylogeny, Dating, phylogenetic informativeness, colonization, speciation, microendemism, Madagascar
List of papers
This thesis is based on the following papers, referred to by their Roman nu-merals in the text:
I. Bukontaite R., Miller K.B. and Bergsten J. 2014. The utility of CAD in recovering Gondwanan vicariance events and the evolutionary history of Aciliini (Coleoptera: Dytiscidae). BMC Evolutionary Biology 2014, 14:5, DOI: 10.1186/1471-2148-14-5.
II. Bukontaite R., Ranarilalatiana T., Randriamihaja J. H., Bergsten J. 2015. In or out-of-Madagascar? - colonization patterns for large-bodied diving
beetles (Coleoptera: Dytiscidae). PLoS ONE,
10(3):e0120777.doi:10.1371/journal.pone.0120777
III. Hjalmarsson E. A., Bukontaite R., Ranarilalatiana T., Randriamihaja J. H., Bergsten J. 2013. Taxonomic revision of Madagascan Rhantus (Coleop-tera, Dytiscidae, Colymbetinae) with an emphasis on Manjakatompo as a conservation priority. ZooKeys 350: 21-45. doi:10.3897/zookeys 350.6127 IV. Bukontaite R., Naimi B., Svensson, E. I., Bergsten J. Phylogeny, distri-bution and speciation in the endemic diving beetle genus Pachynectes on Madagascar. Manuscript
All three published papers are open access and are herein reproduced under the permission of the creative commons by attribution license (CC-BY).
Related article that is not included in the thesis, but have been prepared during the course of the PhD studies:
Strandberg J., Bukontaite R., Malm T. 2015. Partial degeneration – a solution to LBA and saturation problems. Manuscript
Contents
Introduction ... 101
Why water beetles?...11
Madagascar water beetles………12
Why Madagascar?...13
Aims of the Thesis………...15
Material and methods……….16
Collecting………...16
Molecular data……….17
Phylogeny……….19
Dating………...19
Biogeography………..20
Species distribution modeling………..21
Summaries of the papers……….23
Discussion and conclusions.………26
References……….29
Introduction
Why water beetles?
More than one million animal species have been formally described in the world, but it has been estimated that as many as 10 million might exist (Mora et al., 2011). 10% of all known species live in freshwater, a habitat which covers less than 1% of the Earth’s surface. Six out of ten of those are insects and recently Dijikstra et al. (2014) presented aquatic insects as mod-el organisms for studying speciation and diversification.
Aquatic insects have played a major role in understanding the relationship between habitat stability and selection for dispersal capacity. The capacity to disperse is an important factor influencing rates of species diversification and rates of local or global extinction. Immobile species, more prominent in running water, have their highest diversification rate when the environment is stable, whereas mobile species, more prominent in standing waters, show highest diversification rate in a moderately frequently changing environ-ment. However, both mobile and immobile species respond with high extinc-tion rates to frequent or large habitat changes (Dijikstra et al., 2014). Due to habitat loss, pollution, diseases and invasive species, this has led to freshwa-ter organisms being rated as the most threatened “habitat-group” on Earth (National Wildlife Federation report, 2013).
Beetles (Coleoptera) are the most diverse group of insects and during their evolutionary history, beetles have entered the aquatic medium at least ten times independently (Hunt et al., 2007). The most successful group in terms of number of species, and most highly adapted to swimming in water, are the Adephagan water beetles (Jäch & Balke, 2008). They belong to the beetle suborder Adephaga with about 40 000 known species in the world. Adephagan water beetles was dated to around 240 Ma (Hunt et al., 2007), thus the group spans the entire history of the Gondwanan break-up. Adephaga is the second largest suborder of Coleoptera and it is usually di-vided into two main groups: Geadephaga includes the terrestrial ground bee-tles and Hydradephaga contains eight aquatic families of beebee-tles. The Hy-dradephaga vary in length from very small (<1 mm) to very large (5 cm). Of the Hydradephaga, the family Dytiscidae, or predaceous diving beetles, are the most species-rich with over 4000 species (Alarie & Michat, 2014; Nilsson, 2001). Both adults and larvae of diving beetles are aquatic and predatory. The majority of diving-beetles, at least most lentic species, are
capable of flight although this is poorly documented for many species (Jackson, 1956). But the group is often a dominant part of the ecosystem in smaller ephemeral water bodies. Diving beetles are distributed worldwide and inhabit all kinds of both lentic (pools, ponds, lakes, swamps and bogs) and lotic (rivers, streams, springs) aquatic habitats.
However, the Index of Effort (IE) placed Dytiscids near the bottom of a list together with other aquatic groups and other aquatic predators, suggesting it is a very poorly studied group of aquatic insects with only 1 publication per every ten species, the majority of which are related to taxonomy or systemat-ics (Yee, 2014).
Phylogenetic studies of Dytiscidae with morphological (Miller, 2000, 2001), molecular (e.g. Ribera et al., 2008) or combined data (Balke et al. 2004; Miller, et al., 2009; Miller & Bergsten, 2014) have been published with vari-ous resolution and branch support but the evolutionary relationship within this family is still poorly known (Miller & Bergsten, 2014; Ribera et al., 2008). Many large genera are still waiting for comprehensive revision (e.g.
Copelatus, Laccophilus) (Miller & Bergsten, 2014) and recent discoveries in
new habitats, like for example hygropetric rocks or remote areas will likely increase species number as these habitats become better collected (Nilsson-Örtman & Nilsson, 2010).
Madagascar water beetles
About 160 species of Adephagan water beetles are known from Madagascar (Rocchi, 1991), including numerous endemic and non-endemic species, as well as a few endemic genera with species radiations on Madagascar. Diving beetles are important as indicators for freshwater biodiversity assessment which is severely needed for conservation efforts of remaining aquatic habi-tats of Madagascar (paper III). However, aquatic insects on Madagascar are still extremely poorly studied – only a few studies have been done to date and probably 95% of all research focusing on the Malagasy biota is on ver-tebrates and plants. Therefore the understanding of the evolutionary history of freshwater beetles on Madagascar is still in its infancy. How many times did they colonize Madagascar, and when? Did they arrive by rafting on blocks of vegetation washed out from river deltas and helped to the shores of Madagascar by sea currents that during part of the Cenozoic were favorable for crossing the Mozambique channel (Ali & Huber, 2010)? Or did they fly? What level of flight capacity is sufficient for upholding gene-flow across an island the size of Madagascar?
Why Madagascar?
About 400 km west of southern Africa, 4000 km from India, 5000 km from Antarctica and 6000 km from Australia, lies Madagascar, the world’s 4th largest island. It is separated from Africa by the Mozambique Channel and from other landmasses by the Indian Ocean.
It comprises less than 0,4% of Earth’s land surface, yet the level of ende-mism is extremely high from a global perspec-tive: 83% of plants, 99% of amphibians, 92% of reptiles, 52% of birds, 93% of freshwater fishes, 92-100% of terrestrial mammals and 86% of invertebrates on the island are endemic (Goodman & Benstead, 2005). The number of newly discovered species from the island is increasing each year (Vieites et al., 2009). It is believed that Madagascar is filled with high levels of unique biodiversity because of its long history and isolation from the major continents. Madagascar was part of the Gondwana super-continent. The first split between In-do/Madagascar and Africa is thought to have started around 165 Ma, but close-range faunal interchange between the two separating land-masses may still have been possible for a num-ber of million years, perhaps until 140 Ma when marine conditions were clearly present along the entire western coast of Madagascar (Yoder & Nowak, 2006). Madagascar and India remained in close contact, forming the Indo/Madagascar subcontinent, until India split away and headed on to a collision with Asia. According to Yoder & Nowak (2006) the precise timing of India’s separation from Madagascar has been variably dated from 100–95 Ma to 97.6–80.3 Ma, 91.2±0.2 Ma and 87.6+0.6 Ma. However, recent research (Torsvik et al., 2013) proposed that during the opening of the Mascarene Basin (83,2-70 Ma), Mauritius and parts of the micro-continent Mauritia were attached to Madagascar but later fragmented and relocated to the Indian Plate, suggesting a possible connection between Madagascar and India until 61 Ma (Fig. 1). Fig. 1 Geological
Fig 3. Forest cover based on Harper et al., 2007. Vegetation data pro-vided by the Royal Botanic Garden, KEW
The large size of the island and the geo-graphical position of Madagascar throughout history may have played an important role in the formation of dif-ferent habitats (Fig 2). A mountain chain runs for 1 300 km from north to south and separates the tropical wet forest on the eastern slope from the western areas of dry forest embedded in open savannah type vegetation. Central Madagascar contains montane forest and savannah type of vegetation, where-as the west is covered by deciduous forest and the southwest by shrub and sclerophyllus vegetation. Such an enor-mous variety of environments from the very wet to the very dry makes a wide range of environments into which spe-cies can adapt.
The isolation of Madagascar was com-plete already at the end of the Creta-ceous (or the very beginning of Paleo-gene), marked by a global mass extinc-tion at the Cretaceous – Tertiary (K-T) boundary around 65 Ma (Crottini et al., 2012). This is consistent with the ab-sence on Madagascar of fossil records of extant taxa from the whole Tertiary period, while the late Cretaceous fauna including gondwanatheres, multituber-culates, and marsupials is richly repre-sented (Krause, 2001). The lack of fos-sils of post-Cretaceous fauna makes it difficult to understand how and when the extant clades of organisms made it to the island and what environmental factor(s) influenced them to radiate there (Crottini et al., 2012).
Madagascar is ranked number one among the eight “hottest biodiversity
Fig. 2 Ecoregions and Biomes of Madagascar
hotspots” in the world, based on the level of endemism combined with the degree of threat (Goodman & Benstead, 2005; Myers et al., 2000). Slash and burn agriculture, in Madagascar known as ‘tavy’, puts enormous pressures on forests. Green & Sussman (1990) argued that deforestation is most rapid in areas with low topographic relief and high human population density. There is a correlation between human population increase and forest loss on Madagascar (Sussman et al., 1994). 10 % of the primary vegetation remain today on Madagascar (Hanski et al., 2007). More than 40% of the remaining forest was lost between 1950 and 2000 committing to extinction of about 9% of all species as a result of habitat loss during 50 years (Allnutt et al., 2008) (Fig 3). Moreover, WWF reported that in March 2009, after political unrest, the rainforests were pillaged for hardwoods such as rosewood, destroying tens of thousands of hectares of some of the island's most biologically di-verse national parks – including Marojejy, Masoala, Makira and Mananara.
Aims of the Thesis
The main objective of my thesis was to analyze the evolutionary history of diving beetles on Madagascar for a better understanding of the origin and diversity of freshwater organisms on the island.
In my thesis I specifically focused on:
• Understanding the importance of the Gondwana split-up or post break- up oversea dispersal for the evolution of diving beetle clades using dated phylogenetic trees and biogeographic analyses (papers I, II). • Identifying the geographical and ecological role of speciation in an
endemic river-dwelling radiation by combining phylogeny, distribu-tion and previous hypotheses on the microendemism pattern of Madagascan biota (paper IV).
• Inferring distributional ranges for known and newly discovered en-demic species with limited distribution on the island, carrying im-portance for conservation (papers III, IV).
Fig. 4 Collecting localities 2006-2014
Material and methods
Collecting
The material for all my papers came from three sources: 1) existing material from previous expeditions and colleagues; 2) museum collections in Paris (MNHN), London (BMNH), California Academy of Sciences (CAS) and Vienna (NMW); and 3) new field work in Madagascar (Fig. 4).
Sampling was done using GB water nets (Fig. 5 (a)), sieves (Fig.5 (b)) and aspira-tors (Fig.5 (c)), as well as using bottle traps and crayfish traps, baited with fish, chicken liver, cat food or light sticks. Visited habitats included a variety of water bodies: hygropetric rocks, rock-pools, forest creeks, rivers, lakes, marsh-es, ponds and wetlands (Fig. 4). The prioritized season has been either before (October-December), or after (May), the rainy season, mainly due to impossible road conditions during the rainy season for many locations. Each collecting event was documented by recording date, lati-tude and longilati-tude, altilati-tude, landscape, type of waterbody and disturbance such as presence of Zebu cattle, deforestation, villages and agriculture. All localities were also documented with photographs.
Specimens were in situ put in 95% ethanol and later transferred to absolute ethanol and stored in -20 freezers. Tissue from large specimens were either extracted from the thorax region already in the field, or ethanol was injected to the hameolymph post-mortem with a syringe, as DNA degradation other-wise can occur due to the slow permeability of their exoskeleton and re-placement of body fluids with ethanol. Vouchers are deposited at the Swe-dish Museum of Natural History, Stockholm.
During my field work 2011-2014 in Madagascar many new species were found, especially of the endemic genera Pachynectes (paper IV),
Fig. 5 Collecting methods and habitats (GB water nets (a), sieves and kitchen strainers (b), aspirator (c))
Madaglymbus and Hovahydrus as well as new samples of species like Hy-daticus madagascariensis (known only from old museum collections) and
the endemic Hydaticus limnetes and Hydaticus saecularis (known only from type material), (see paper II).
However, so far, we have not managed to find a few already known species:
Hydaticus plagiatus, which is known only from a single unique type
speci-men, and Cybister dytiscoides, the largest Malagasy diving beetle. This could be due to the habitat loss and species extinction (Hanski et al., 2007).
Molecular data
Molecular data is commonly used to infer the evolutionary relationship among organisms. Rapidly evolving genes are useful for reconstructing phy-logenetic relationship among closely related species, but as the organisms become less closely related, mutations at the third codon position of protein coding genes might become saturated. This can lead to misinterpretation as homology statements between sequences become more difficult and multiple changes in the same position are disguised. For example, mitochondrial pro-tein coding genes are often used among closely related organisms as it might suffer from saturation at higher levels in the phylogeny. In contrast, nuclear
(a)
(c)
protein-coding genes may show high informativeness at higher levels of inferred trees. In paper I we evaluated two mitochondrial protein-coding genes, cytochrome c oxidase subunit (COI) and II (COII), two ribosomal genes, mitochondrial 16S and nuclear 28S and three nuclear protein-coding genes, histone 3 (H3), wingless (Wnt) and rudimentary (CAD) for their utili-ty and informativeness over evolutionary time of diving beetles. We found that CAD and Wnt are the most informative genes along the represented evolutionary history, with mitochondrial markers being the most useful at younger ages.
In paper III we used mitochondrial COI for species delimitation analyses to test the status of two allopatric populations in Madagascar.
During my PhD project I optimized molecular lab protocols for the diving beetles I was working on. I usually extracted DNA from the hindlegs or head and prothorax of ethanol-preserved or dry-mounted material, depending on the size of the specimens, and incubated in the lysis buffer overnight. Mo-leStrips™ DNA Tissue and KingFisher Cell and Tissue DNA kit were then used for isolation of DNA with an elution volume of 100 μl. DNA of old and dry-mounted specimens was extracted using the DNEasy DNA extraction kit following the manufacturer’s recommendation, except for that 20 μl of 1 molar DTT (dithiothreitol) was added during the lysis stage. After extrac-tion, I retained the tissue (legs or head with prothorax) and reunited them with the vouchers.
For DNA amplification I used illustra TM Hot Start Mix “Ready-to-go” PCR beads from GE Healthcare following manufacturing protocols. Each 25 µl reaction contained 1 µl of 10 µM primer pair mix (x2), 2 µl of DNA tem-plate and 21 µl of water. The PCR cycling profile followed 940C for 5min, followed by 40 cycles of 940C 30s, 50-560C for 30 s (depending on the pri-mers’ annealing temperature), 720C for 1-1,5 min (depending on the length of the gene fragment) and 720C for 8 min. Amplified products were purified using ExoFast Cleanup mix (Fermentas) and run in the program set to 370C for 30 min followed by 800C for 15 min.
Sequencing reactions were obtained with ABI BigDye Terminator kit (Ap-plied Biosystems). Each sequencing reaction contained 1 μl of BigDye™ (Applied Biosystems), 1 μl of 1,6 μM primer and 2-6 μl of PCR product. The sequence cycling profile was 95°C for 1 min and then 25 cycles of 95°C 30 s, 50°C 15 s and 60°C 4 min. Sequencing products were purified using the DyeEx 96 kit and fragments were analyzed on an ABI377xl analyzer from Applied Biosystems at the Molecular Systematics Laboratory, Swedish Mu-seum of Natural History.
Gene regions were sequenced in both directions. The contigs were assem-bled from the forward and reverse reads, and primers were trimmed, in Se-quencher 5.0.
Phylogeny
Several commonly used methods are available to infer evolutionary relation-ships of a group of interest. The Parsimony method prefers the evolutionary tree with the least number of evolutionary changes to explain the observed differences among taxa (Edwards, 2009 and references therein). However, because of its simplicity, it does not account for rate variation among line-ages nor does it account for unseen changes. Maximum Likelihood (ML) is based on a statistical approach that provides the tree with the highest likeli-hood of having given rise to the observed data given a model (Stamatakis, 2006). It results in the “maximum likelihood” tree with bootstrap analysis commonly used to assess branch support. However, it might be very compu-tationally intensive and slow. The third and, probably the most oftenly used method – the Bayesian approach (Ronquist, 2004) - perhaps provides the most realistic results due to the way uncertainties surrounding all model parameters in the analysis are treated simultaneously (Ronquist, 2004). Simi-lar to ML, it is also based on a likelihood function, but with branch support displayed as posterior probabilities of nodes.
In all my papers I applied the Bayesian approach to infer the phylogenetic relationships. However, I utilized both Bayesian and ML in papers II and IV, in order to assess stability of results to different methods of inference.
Dating
The divergence times, or age, of species and clades can be estimated by as-suming that the evolutionary rate is either a constant (Strict molecular clock) or varies (Relaxed molecular clock) over time and across evolutionary line-ages. Under the strict molecular clock the rate of molecular evolution is equal to the neutral mutation rate and is independent of environmental or evolutionary changes. However, there are many additional attributes in the biology of species that might affect substitution rates (Martin & Palumbi, 1993).
Under the relaxed molecular clock two assumptions are possible: 1) the rates across branches can evolve over time and rates are inherited, i.e. the autocor-relation assumption (Yang & Yoder, 2003); 2) no corautocor-relation for adjacent branches is assumed under the uncorrelated relaxed clock (Drummond et al., 2006). In the case of the autocorrelation assumption, the model is based on the hypothesis that closely related lineages share similar biological proper-ties. Thus, they are expected to have similar evolutionary rates (Drummond et al., 2006).
In contrast, for the uncorrelated relaxed molecular clock, the mutation rates are allowed to vary independently under the chosen rate distribution, for example under an exponential or lognormal distribution.
Here (in paper I and paper II) I use the uncorrelated lognormal clock to relax the equal rate constraints across the phylogeny, and to date my trees with the BEAST software (Drummond et al., 2006). But dating does not only need a clock function, there is also a need for one or more calibration points on which to attach the phylogeny and convert lineage rates to a “real” time-scale.
For calibrating the trees I use two types of calibration points: fossil records and geological events.
Fossil records are the most commonly used external information. However, fossils are usually incomplete and fragmentary and can easily be misidenti-fied and incorrectly placed on the phylogeny. In paper I we tested two pos-sible positions of the fossil Acilius florissantensis. Based on the descriptions of the fossil it is possible to either associate it with the monophyletic Nearc-tic clade of Acilius or to whole genus.
If there is no fossil known for the group of interest, some authors suggest using geological events as calibration points (Heads, 2010). As there were no suitable fossil record to use as calibration for Cybistrini, I used the age of the island Mauritius (7-8 Ma, Warren et al., 2003) as external information for the Cybistrini analyses in paper II.
Biogeography
While phylogenetic trees represent the relationships between species and dated trees show the times of origin, questions such as where ancestral spe-cies were distributed and how their distribution changed over time are left to be considered. Several methods have been developed to study ancestral dis-tributions, such as Dispersal-Vicariance Analyses (DIVA) (Ronquist 1997), Dispersal-Extinction-Cladogenesis (DEC) (Ree et al., 2005; Ree & Smith, 2008) and a Bayesian Binary Method (BBM) (Yu et al., 2015).
DIVA is a simple, event-based approach, which reconstructs the ancestral distribution of a group of organisms by minimizing the dispersal events needed for explaining the present-day distribution. In this approach, the vi-cariance events have no cost, while dispersals and extinctions cost one per area unit added to the distribution (Ronquist, 1997). However, the main drawback of the original DIVA method is that it can only handle fully bifur-cated trees and that it reconstructs biogeography on a fixed tree without un-derlying topological uncertainty. Nylander et al. (2008) and Harris and Xiang (2009) proposed Bayesian approaches that account for phylogenetic uncertainty and thereby allow a more accurate analysis of the biogeographic history of lineages. Both methods build upon using the original program
DIVA on every individual tree in a pool of trees sampled from the posterior distribution of topologies.
The DEC model described by Ree et al. (2005) and Ree & Smith (2008) is based on a complex but flexible stochastic model, in which evolution of geographic ranges both along a branch and at the nodes are modelled explic-itly. It attempts to estimate how ancestral areas were inherited by daughter lineages, in contrast to DIVA. This model is evaluated in a likelihood framework, and the ancestral range inheritance scenario with the best likeli-hood of observing the current species distributions is chosen as optimal. The DEC model also allows imposing dispersal constraints during certain points in time.
BBM is the most complex model implemented in RASP (Yu & Harris, 2013). It is a Bayesian approach of character evolution (Ronquist, 2004) with relative rates of change among character states in the model of range evolution inferred during analysis. In paper I I used BBM method to infer the ancestral areas of Aciliini. In paper II I tested Bayes-DIVA by Nylander et al., (2008), revised Bayes-DIVA by Harris and Xiang (2009) and DEC methods for both Hydaticini and Cybistrini datasets. All the methods result-ed in very similar ancestral areas.
Species distribution modeling
Species distribution modeling (SDM) (also known as “Ecological niche modeling” (ENM)) has become a great tool for mapping the geographical distributions of species abiotic niches (Peterson et al., 2011). It is an im-portant tool to bridge the gap in our knowledge and enable predictions of species distributions over large unsampled areas (Raxworthy et al., 2003). SDM uses environmental conditions drawn from known georeferenced lo-calities to build a distribution model of a species’ realized ecological niche (Elith et al., 2006; Phillips et al., 2006; Phillips & Dudı, 2008). It has been used for different purposes, such as alarming for extinction risks and the conservations of species (Jenkins et al., 2013; Kremen et al., 2008; Pimm et al., 2014), detection of unknown species areas (Raxworthy et al., 2003) and predicting future distributions of invasive species (Peterson, 2003). Moreo-ver, integrating phylogenetic hypotheses with geographic and ecological data has revealed new insight into the factors that influence the evolution and distribution of species (Graham et al., 2004 and the references there in). Wiens (2007) showed that inferences of character evolution could be in-formed by GIS-based methods. The combination of niche modeling and phylogenetic analysis can provide some insight into the role of ecological divergence in speciation, regardless of the particular geographic mode of speciation: parapatric, sympatric and allopatric (see e.g. Raxworthy et al., 2007). No overlap in the potential distribution between sister-species indi-cates that speciation has been an adaptive allopatric process to local
envi-ronments. Broad overlap between sister-species potential distributions could be interpreted either as sympatric speciation where local adaptation to meas-ured climatic variables was not important, or parapatric speciation with sec-ondary dispersal and contact.
In paper IV I use SDM together with the phylogeny of the endemic radia-tion of the diving-beetle genus Pachynectes in Madagascar to test whether realized or potential distributions according to SDM overlap between en-demic sisters species and to test whether watershed or climatic gradients are responsible for the distribution patterns in Pachynectes. I applied the maxi-mum entropy (MaxEnt) model to fit the species distribution models (SDMs), a commonly used algorithm for presence-only records (Phillips et al., 2006). However, the challenge I faced in this paper was too few known occurrence points of some of the species, which is a common issue for the microendem-ic species of Madagascar (Pearson et al., 2006). Following Pearson et al., 2006, I used the Jacknife (or “leave-one-out”) procedure to evaluate the ac-curacy of the models, which has been shown to be capable of assessing the predictive ability of models built using very small sample sizes.
Summaries of the papers
Paper I
Bukontaite R., Miller K.B. and Bergsten J. 2014. The utility of CAD in re-covering Gondwanan vicariance events and the evolutionary history of Aciliini (Coleoptera: Dytiscidae). BMC Evolutionary Biology 2014, 14:5, DOI: 10.1186/1471-2148-14-5.
(Contribution: RB and JB conceived the study. RB and KBM performed the lab work. JB and KBM provided crucial samples. RB and JB performed the analysis and wrote the first draft. All authors contrib-uted to the final manuscript. All authors read and approved the final manuscript).
The tribe Aciliini belongs to the subfamily Dytiscinae of diving beetles (Dy-tiscidae) and presently includes 69 species of medium-sized water beetles distributed on all continents except Antarctica. Evolutionary relationships among Aciliini genera had never before been analyzed with molecular phy-logenetic methods. Paper I presents a molecular phyphy-logenetic analysis of the Aciliini tribe, including evolutionary relationship within and among genera, divergence times and ancestral distribution of taxa. Our combined data set included 6095 bp of DNA sequences, including both mitochondrial and nu-clear gene markers. Moreover, we introduced a new nunu-clear protein-coding gene, CAD, to the analysis of Hydradephagan beetles. Our gene evaluation analyses suggested CAD to be the most informative marker between 15 and 50 Ma, while at deeper times Wnt was more informative. Mitochondrial protein coding genes were considered the most informative for the radiation within genera.
The monophyly of Aciliini was questioned by some authors (Ribera et al., 2008), but our analyses suggested Aciliini to be monophyletic with Eretini as its sister group with high support. We recovered a fully resolved backbone between all seven genera of Aciliini, and optimized their geographical histo-ry on the resulting phylogeny. The basalmost clades were optimized to South America (Thermonectus) and Africa (Aethionectes and Tikoloshanes), which suggest a Gondwana vicariance origin. However, the uncertainty as to whether our fossil calibration should be used as a stem- or crowngroup con-straint for Acilius influenced the result. Using the fossil as a crowngroup calibration supported the Gondwana break-up theory, but used as a stem group calibration the basal nodes of the tree were too young to support this theory.
Paper II
Bukontaite R., Ranarilalatiana T., Randriamihaja J.H., Bergsten J. In or out-of-Madagascar? - colonization patterns for large-bodied diving beetles (Coleoptera:Dytiscidae). PLoS ONE 2015,
10(3):e0120777.doi:10.1371/journal.pone.0120777
(Contribution: RB and JB conceived and designed the study. RB analyzed the data. RB and JB wrote the first draft paper. RB, TR, JHR and JB did the field work. All authors read and approved the final manu-script).
While mammals, birds, amphibians and reptiles have been the main focus of much of the research into colonization and speciation on Madagascar, stud-ies on the hugely diverse insects are starting to appear. In paper II we fo-cused on diving beetles from the tribes Hydaticini and Cybistrini. These groups contain numerous both endemic and non-endemic species that make them a perfect model-group for analyses of colonization and radiation on Madagascar.
We used two previously published datasets by Miller et al., (2007, 2009) together with newly collected endemic species from Madagascar. Our results suggested that Hydaticini and Cybistrini colonized Madagascar multiple times, mainly during Miocene and Oligocene. The timing is similar to many other endemic taxa of Madagascar (Yoder & Novak, 2006), but the result was unusual in that there was no clear evidence of a colonization that was followed by species radiation on the island. Also, because the Madagascar species were widely spread out throughout the phylogeny in Hydaticus and had rather basal placements in several species groups, all three biogeograph-ic methods suggested a Madagascar origin for a large part of the genus with out-of Madagascar dispersal events to Africa and the Orient. We discuss potential explanations for this unexpected result.
Paper III
Hjalmarsson E. A., Bukontaite R., Ranarilalatiana T., Randriamihaja J. H., Bergsten J. 2013. Taxonomic revision of Madagascan Rhantus (Coleop-tera, Dytiscidae, Colymbetinae) with an emphasis on Manjakatompo as a conservation priority. ZooKeys 350: 21-45. doi:10.3897/zookeys 350.6127
(Contribution: AEH and JB conceived and designed the study. RB performed the molecular lab work. TR, JHR, RB and JB did the field work. AEH performed the molecular and morphological analyses. AEH and JB wrote the first draft. All authors read and approved the final manuscript).
Here we review the diving-beetle genus Rhantus Dejean of Madagascar based on museum collection holdings and recently collected expedition ma-terial. Using both morphology and molecular data we aimed to test species boundaries by applying the general mixed Yule coalescent model to an ul-trametric gene-tree for species delimitation. We recognized three species of
Rhantus to occur in Madagascar: R. latus (Fairmaire, 1869), R. bouvieri
Regimbart, 1900 and R. manjakatompo Pederzani and Rocchi, 2009, the last described based on a single male specimen from the central Ankaratra moun-tains. All three species are endemic to Madagascar and restricted to the high-lands of the island. We also provide descriptions, a determination key, SEM-images of fine pronotal and elytral structures, distribution maps, habitus photos, and illustrations of male genitalia and pro- and mesotarsal claws of each species. As well we discuss the role of the Manjakatompo forest as a refugium for Madagascan Rhantus diversity and other endemics of the mon-tane central high plateau.
Paper IV
Bukontaite R., Naimi B., Svensson, E.I., Bergsten J. Phylogeny, distribution and speciation in the endemic diving beetle genus Pachynectes on Mada-gascar. Manuscript
(Contribution: RB and JB conceived and designed the study with later input from EIS. RB and NB ana-lysed the data. RB and JB collected the samples. RB, JB and EIS wrote the paper. All authors read and approved the final manuscript).
Madagascar is well known for its high level of endemism. But most of its endemic species also have very restricted regional distributions, making it the main aspect of the island’s biodiversity. However, what led to such mi-croendemic diversity on the island, scientists are still struggling to explain. Several hypotheses including watersheds as factors or climatic gradients have been used to explain the micro - distribution of Malagasy species. Here, we focused on the endemic diving beetle genus Pachynectes. After extensive new sampling efforts the species diversity of Pachynectes turned out to be at least three times higher than previously believed. We reconstructed a well-supported phylogeny of this genus based on five gene regions and recovered three main clades: a western clade, an eastern clade and a clade defined by vaguely keeled elytra. The western and keeled clades are both distributed along the drier western part of the country, while the eastern clade occurs in the eastern humid rainforest. Based on our SDM results, the microendemic distribution pattern of Pachynectes species follow climate and biomes rather than watersheds. In addition sister species were allopatrically distributed suggesting an allopatric mode of speciation in the group.
Discussion and conclusions
The “Tree of life” hypothesis suggests that the evolutionary history of all organisms can be traced back to a single shared common ancestor. Evidence from both morphological and molecular data suggests that all organisms on Earth are genetically related. It is fascinating for every scientist to add detail to the amazing Tree of life puzzle for a better understanding of evolution. In paper I we put together the Aciliini puzzle, which had never been studied in detail before. Based on a comprehensive molecular dataset, our analyses confirmed that Aciliini, as well as its genera, are monophyletic groups with strong support. Biogeographic history of the group likely goes back to a Gondwana break-up history with ancestors experiencing the vicarience event between S. America and Africa as the south Atlantic opened. However this hypothesis was dependent on the position of our calibration point, which could place the age of the group as early as in Eocene epoch supporting dis-persal across open ocean as an alternative hypothesis. The uncertainty in estimating the divergence time is a common issue as fossils rarely represents specific nodes or splits in a phylogeny of extant species, but rather points along a branch or side branch, making it difficult to determine an exact an-cestor of a specific clade. For increasing the precision of divergence times, future studies are needed which include multiple primary calibration points and takes full uncertainty into account. Since fossils in this group are scarce this will probably have to be done in the context of a higher level phylogeny. Another interesting piece of the puzzle I am presenting in my thesis is the biogeographic history of Hydaticini and Cybistrini species on the island of Madagascar (paper II). The predominant pattern of the present-day biota of Madagascar is comprised of the descendents of Cenozoic dispersers with African origins (Yoder & Nowak, 2006). However, our results indicated that both groups can be traced back to multiple colonizations but also potential out of Madagascar dispersal events sometime during the Eocene and Oligo-cene. Several other studies have also inferred similar “out-of-island” pattern mainly to the surrounding Mascarene and Comoros islands (e.g. Harmon et al., 2008; Vences et al., 2003; Agnarsson & Kuntner 2012) and some groups to mainland Africa (e.g Jansa, et al., 1999). However in the cases of Hy-daticini and Cybistrini, we speculate that ”out – of Madagascar” pattern, could be due to the taxonomic sampling being biased towards, and overrepresented by Madagascan species. This hypothesis remains to be
test-ed with an equally intense sampling in natural habitats on the African main-land as on Madagascar, which might twist the inferred colonization events to opposite direction. Moreover, several of the endemic species of Hydaticini and Cybistrini were very rare and found only in few localities. We speculate that the lack of species radiations for these two groups on Madagascar is due to their dispersal ability where the size of the island does not represent sig-nificant barriers to gene flow for these groups of large-bodied diving beetles. This can also be concluded from the test of populations of a similar large-bodied species of Colymbetinae (paper III). Rhantus manjakatompo, known only from populations on two mountain massifs in central and northern Madagascar (900 km apart), were still found to be conspecific based on mo-lecular and morphological data.
In contrast, the genus Pachynectes represents a species radiation on the is-land with three times higher diversity than previously believed (paper IV), with some species known only from single or very few localities. In contrast to Hydaticini, Cybistrini and Colymbetinae, Pachynectes are small beetles (3-5 mm) and are running water specialists. The phylogeny places two clades largely confined to the drier western part as basal to the third clade confined to the humid eastern part. That the invasion of the eastern humid forests happened more recently is in agreement with both studies on some lizard groups (e.g. Blair et al., 2015) and presumed ages of the biomes (Wells, 2003). Species distribution modelling based on the current known localities confirm micro- or regional endemic ranges and together with the phylogeny inferred speciation mode to be allopatric with largely non-overlapping ranges between sisters species. Finally, the climatic gradient model, simplified into bioclimatic regions of Madagascar by Cornet (1974), was a better explanatory model for the group than watershed regions pro-posed by Wilmé et al., (2006). Wilmé et al's (2006) watershed hypothesis has been successful to mainly explain the distribution of lemurs, whereas the climate gradient hypothesis seems more suitable for some reptile groups (Pearson & Raxworthy, 2009; Raxworthy et al., 2007). A consensus is now emerging that it is unlikely to find a single model that fits for all endemic groups (Blair et al., 2015; Brown et al., 2014; Pearson & Raxworthy, 2009). This is because various biotic and abiotic factors might influence speciation in different species groups differently, even among closely related species (Pearson and Raxworthy, 2009). Identifying the cause of microendemic pat-terns in Madagascar is a very challenging task also because of the rareness of many endemic species on Madagascar (Bukontaite et al., 2015; Hjalmarsson et al., 2013; Pearson et al., 2006), making it difficult to apply species distri-bution modelling on the available data (Pearson et al., 2006). Moreover, improvement of methods should be considered for modeling presence-only data and accounting for biotic interactions to be able to achieve the most realistic distributions (Elith & Leathwick, 2009).
The restricted distribution inferred for many of the endemic species (paper III, IV) together with the severe past and ongoing deforestation (e.g. Allnutt et al., 2008) carries important conservation implications. Kremen et al. (2008) used distribution data of endemic species from several taxonomic groups to suggest areas of high priority as the government of Madagascar decided to expand the protected area network. For insect representatives they only used data on terrestrial groups such as ants and butterflies. They did not include freshwater insect diversity, despite the urgent need for freshwater conservation in Madagascar (Benstead et al., 2003), mainly because of the lack of well-curated data and knowledge of such groups.
References
Agnarsson, I. & Kuntner, M. (2012). The Generation of a Biodiversity Hotspot: Biogeography and Phylogeography of the Western Indian Ocean Islands. In Kesara Anamthawat-Jonsson (Ed.): Current Topics in Phylogenetics and Phylogeography of Terrestrial and Aquatic Systems. Rijeka: In Tech Publishers, pp. 33-82.
Alarie, Y., & Michat, M. . (2014). Bridging Ecology and Systematics: 25 years of study of larval morphology of world Dytiscidae. In D. Yee (Ed.), Ecology, Systematics, and the Natural History of Predaceous
Diving Beetles (Coleoptera: Dytiscidae) (pp. 17–47). Springer
Netherlands
Ali, J. R., & Huber, M. (2010). Mammalian biodiversity on Madagascar controlled by ocean currents. Nature, 463(7281), 653–6.
doi:10.1038/nature08706
Allnutt, T. F., Ferrier, S., Manion, G., Powell, G. V. N., Ricketts, T. H., Fisher, B. L., Harper, G. J., et al. (2008). A method for quantifying biodiversity loss and its application to a 50-year record of deforestation across Madagascar. Conservation Letters, 1(4), 173–181.
doi:10.1111/j.1755-263X.2008.00027.x
Balke, M., Ribera, I., & Vogler, A. P. (2004). MtDNA phylogeny and biogeography of Copelatinae, a highly diverse group of tropical diving beetles (Dytiscidae). Molecular phylogenetics and evolution, 32(3), 866–80. doi:10.1016/j.ympev.2004.03.014
Benstead, J. P., De Rham, P. H., Gattolliat, J.-L., Gibon, F.-M., Loiselle, P. V., Sartori, M., Sparks, J. S., et al. (2003). Conserving Madagascar’s Freshwater Biodiversity. BioScience, 53(11), 1101. doi:10.1641/0006-3568(2003)053[1101:CMFB]2.0.CO;2
Blair, C., Noonan, B. P., Brown, J. L., Raselimanana, A. P., Vences, M., & Yoder, A. D. (2015). Multilocus phylogenetic and geospatial analyses illuminate diversification patterns and the biogeographic history of Malagasy endemic plated lizards (Gerrhosauridae: Zonosaurinae).
Brown, J. L., Cameron, A., Yoder, A. D., & Vences, M. (2014). A necessarily complex model to explain the biogeography of the amphibians and reptiles of Madagascar. Nature communications, 5, 5046. doi:10.1038/ncomms6046
Bukontaite, R., Ranarilalatiana, T., Randriamihaja, J. H., & Bergsten, J. (2015). In or Out-of-Madagascar?—colonization patterns for large-bodied diving beetles (Coleoptera: Dytiscidae). PLOS ONE, 10(3), e0120777. doi:10.1371/journal.pone.0120777
Cornet, A. (1974). Essai de cartographie bioclimatique à Madagascar. Notice explicative. ORSTOM, 1–55.
Crottini, A., Madsen, O., Poux, C., Strauss, A., Vieites, D. R., & Vences, M. (2012). Vertebrate time-tree elucidates the biogeographic pattern of a major biotic change around the K-T boundary in Madagascar.
Proceedings of the National Academy of Sciences of the United States of America, 109(14), 5358–63. doi:10.1073/pnas.1112487109
Dijkstra, K.-D. B., Monaghan, M. T., & Pauls, S. U. (2014). Freshwater biodiversity and aquatic insect diversification. Annual Review of
Entomology, 59, 143–63. doi:10.1146/annurev-ento-011613-161958
Drummond, A. J., Ho, S. Y. W., Phillips, M. J., & Rambaut, A. (2006). Relaxed phylogenetics and dating with confidence. PLoS biology, 4(5), e88. doi:10.1371/journal.pbio.0040088
Edwards, A. W. F. (2009). Statistical methods for evolutionary trees.
Genetics, 183(1), 5–12. doi:10.1534/genetics.109.107847
Elith, J., Graham, C. H., Anderson, R. P., Dudı´k, M., Ferrier, S., Guisan, A., Hijmans, R. J., et al. (2006). Novel methods improve prediction of species’ distributions from occurrence data. Ecography, 29, 129–151. Elith, Jane, & Leathwick, J. R. (2009). Species Distribution Models:
Ecological Explanation and Prediction Across Space and Time. Annual
Review of Ecology, Evolution, and Systematics, 40(1), 677–697.
doi:10.1146/annurev.ecolsys.110308.120159
Goodman, S. M., & Benstead, J. P. (2005). Updated estimates of biotic diversity and endemism for Madagascar. Oryx, 39(01).
Graham, C. H., Ferrier, S., Huettman, F., Moritz, C., & Peterson, A. T. (2004). New developments in museum-based informatics and applications in biodiversity analysis. Trends in Ecology & Evolution,
19(9), 497–503. doi:10.1016/j.tree.2004.07.006
Green, G. M., & Sussman, R. W. (1990). Deforestation history of the eastern rain forests of Madagascar from satellite images. Science, 248(4952), 212–5. doi:10.1126/science.248.4952.212
Hanski, I., Koivulehto, H., Cameron, A., & Rahagalala, P. (2007). Deforestation and apparent extinctions of endemic forest beetles in Madagascar. Biology Letters, 3(3), 344–7. doi:10.1098/rsbl.2007.0043 Harmon, L. J., Melville, J., Larson, A., & Losos, J. B. (2008). The role of
geography and ecological opportunity in the diversification of day geckos (Phelsuma). Systematic Biology, 57(4), 562–73.
doi:10.1080/10635150802304779
Harris, A. J., & Xiang, Q.-Y. (J.). (2009). Estimating ancestral distributions of lineages with uncertain sister groups: a statistical approach to Dispersal-Vicariance Analysis and a case using Aesculus L.
(Sapindaceae) including fossils. Journal of Systematics and Evolution,
47(5), 349–368. doi:10.1111/j.1759-6831.2009.00044.x
Heads, M. (2010). Evolution and biogeography of primates: a new model based on molecular phylogenetics, vicariance and plate tectonics.
Zoologica Scripta, 39(2), 107–127.
doi:10.1111/j.1463-6409.2009.00411.x
Hjalmarsson, A., Bukontaite, R., Ranarilalatiana, T., Randriamihaja, J., & Bergsten, J. (2013). Taxonomic revision of Madagascan Rhantus (Coleoptera, Dytiscidae, Colymbetinae) with an emphasis on Manjakatompo as a conservation priority. ZooKeys, 350, 21–45. doi:10.3897/zookeys.350.6127
Hunt, T., Bergsten, J., Levkanicova, Z., Papadopoulou, A., John, O. S., Wild, R., Hammond, P. M., et al. (2007). A comprehensive phylogeny of beetles reveals the evolutionary origins of a superradiation. Science,
318(5858), 1913–6. doi:10.1126/science.1146954
Jäch, M.A. and Balke, M. (2008). Global diversity of water beetles (Coleoptera) in freshwater. Hydrobiologia, 595, 419-442. doi:
Jackson, D. J. (1956). Observations on flying and flightless water beetles.
Journal of the Linnean Society of London, Zoology, 43(289), 18–42.
doi:10.1111/j.1096-3642.1956.tb02505.x
Jansa, S. A., Goodman, S. M., & Tucker, P. K. (1999). Molecular phylogeny and biogeography of the native rodents of Madagascar (Muridae : Nesomyinae): a test of the single-origin hypothesis. Cladistics, 270, 253–270.
Jenkins, C. N., Pimm, S. L., & Joppa, L. N. (2013). Global patterns of terrestrial vertebrate diversity and conservation. Proceedings of the
National Academy of Sciences of the United States of America, 110(28), E2602–10. doi:10.1073/pnas.1302251110
Krause, D. W. (2001). Fossil molar from a Madagascan marsupial. Nature,
412(6846), 497–8. doi:10.1038/35087649
Kremen, C., Cameron, A., Moilanen, A., Phillips, S. J., Thomas, C. D., Beentje, H., Dransfield, J., et al. (2008). Aligning conservation priorities across taxa in Madagascar with high-resolution planning tools. Science, 320(5873), 222–6. doi:10.1126/science.1155193
Martin, P. A., & Palumbi, R. S. (1993). Body size, metabolic rate, generation time, and the molecular clock. Proceedings of National Academy of
Sciences of the United States of America, 90, 4087–4091. doi:
10.1073/pnas.90.9.4087
Miller, K. B. (2000). Cladistic analysis of the tribes of Dytiscinae and the phylogenetic position of the genus Notaticus Zimmermann
(Coleoptera: Dytiscidae). Insect Systematics & Evolution, 31(2), 165– 177. doi:10.1163/187631200X00363
Miller, K. B. (2001). On the phylogeny of the Dytiscidae (Insecta: Coleoptera) with emphasis on the morphology of the female reproductive system. Insect Systematics & Evolution, 32(1), 45–89. doi:10.1163/187631201X00029
Miller, K. B., Bergsten, J., & Whiting, M. F. (2007). Phylogeny and classification of diving beetles in the tribe Cybistrini (Coleoptera, Dytiscidae, Dytiscinae). Zoologica Scripta, 36(1), 41–59.
doi:10.1111/j.1463-6409.2006.00254.x
Miller, K. B., Bergsten, J., & Whiting, M. F. (2009). Phylogeny and
choice for Bayesian analysis with multiple nuclear and mitochondrial protein-coding genes. Zoologica Scripta, 38(6), 591–615.
doi:10.1111/j.1463-6409.2009.00393.x
Miller, K., & Bergsten, J. (2014). The phylogeny and classification of predaceous diving beetles. In A. D. Yee (Ed.), Ecology, Systematics
and the Natural History of Predaceous Diving Beetles (Coleoptera: Dytiscidae) (pp. 49–172). Springer Netherlands
Mora, C., Tittensor, D. P., Adl, S., Simpson, A. G. B., & Worm, B. (2011). How many species are there on Earth and in the ocean? PLoS biology,
9(8), e1001127. doi:10.1371/journal.pbio.1001127
Myers, N., Mittermeier, R. A., Mittermeier, C. G., Da Fonseca, G. A., & Kent, J. (2000). Biodiversity hotspots for conservation priorities.
Nature, 403(6772), 853–8. doi:10.1038/35002501
National Wildlife Federation report, 2013. http://www.nwf.org/Who-We-Are/Annual-Report/
Nilsson, A. N. (2001). Dytiscidae (Coleoptera). World Catalogue of Insects. Vol. 3. Strenstrup: Apollo Books, pp. 395
Nilsson-Örtman, V., & Nilsson, A. (2010). Using taxonomic revision data to estimate the global species richness and characteristics of underscribed species of diving beetles (Coleoptera:Dytiscidae). Biodiversity
Informatics, 7, 1–16.
Nylander, J. A. A., Olsson, U., Alström, P., & Sanmartín, I. (2008).
Accounting for phylogenetic uncertainty in biogeography: a Bayesian approach to dispersal-vicariance analysis of the thrushes (Aves: Turdus). Systematic Biology, 57(2), 257–68.
doi:10.1080/10635150802044003
Pearson, R. G., & Raxworthy, C. J. (2009). The evolution of local endemism in Madagascar: watershed versus climatic gradient hypotheses
evaluated by null biogeographic models. Evolution; international
journal of organic evolution, 63(4), 959–67.
doi:10.1111/j.1558-5646.2008.00596.x
Pearson, R. G., Raxworthy, C. J., Nakamura, M., & Townsend Peterson, a. (2006). Predicting species distributions from small numbers of occurrence records: a test case using cryptic geckos in Madagascar.
Journal of Biogeography, 34(1), 102–117.
doi:10.1111/j.1365-2699.2006.01594.x
Peterson, A. T. (2003). Predicting the Geography of Species’ Invasions via Ecological Niche Modeling. The Quarterly Review of Biology, 78(4), 419–433. doi:10.1086/378926
Peterson, T. A., Soberon, J., Pearson, R., Anderson, R. P., Martines-Meyer, E., Nakamura, M., & Araujo, M. B. (2011). Niches and Geographic
distributions. Princeton University Press.
Phillips, S. J., Anderson, R. P., & Schapire, R. E. (2006). Maximum entropy modeling of species geographic distributions. Ecological Modelling,
190(3-4), 231–259. doi:10.1016/j.ecolmodel.2005.03.026
Phillips, S. J., & Dudı, M. (2008). Modeling of species distributions with Maxent : new extensions and a comprehensive evaluation, 161–175. doi:10.1111/j.2007.0906-7590.05203.x
Pimm, S. L., Jenkins, C. N., Abell, R., Brooks, T. M., Gittleman, J. L., Joppa, L. N., Raven, P. H., et al. (2014). The biodiversity of species and their rates of extinction, distribution, and protection. Science,
344(6187), 1246752. doi:10.1126/science.1246752
Raxworthy, C. J., Ingram, C. M., Rabibisoa, N., & Pearson, R. G. (2007). Applications of ecological niche modeling for species delimitation: a review and empirical evaluation using day geckos (Phelsuma) from Madagascar. Systematic biology, 56(6), 907–23.
doi:10.1080/10635150701775111
Raxworthy, C. J., Martinez-Meyer, E., Horning, N., Nussbaum, R. A., Schneider, G. E., Ortega-Huerta, M. A., & Peterson, T. A. (2003). Predicting distributions of known and unknown reptile species in Madagascar. Nature, 426(6968), 837–41. doi:10.1038/nature02205 Ree, H. R., Moore, R. B., Webb, O. C., & Donoghue, J. M. (2005). A
likelihood framework for inferring the evolution of geographic range on phylogenetic trees. Evolution, 11(59), 2299–2311.
Ree, R. H., & Smith, S. A. (2008). Maximum likelihood inference of geographic range evolution by dispersal, local extinction, and cladogenesis. Systematic Biology, 57(1), 4–14.
Ribera, I., Vogler, A. P., & Balke, M. (2008). Phylogeny and diversification of diving beetles (Coleoptera: Dytiscidae). Cladistics, 24(4), 563–590. doi:10.1111/j.1096-0031.2007.00192.x
Rocchi, S. (1991). Contributo alla conoscenza degli Haliplidae e dei Dytiscidae del Madagascar con descrizione di due nuove specie (Coleoptera). Frustula Entomologica (N.S.), 14(27), 71–89.
Ronquist, F. (1997). Dispersal-Vicariance Analysis: A New Approach to the Quantification of Historical Biogeography. Systematic Biology, 46(1), 195–203. doi:10.1093/sysbio/46.1.195
Ronquist, F. (2004). Bayesian inference of character evolution. Trends in
ecology & evolution, 19(9), 475–81. doi:10.1016/j.tree.2004.07.002
Stamatakis, A. (2006). RAxML-VI-HPC: maximum likelihood-based phylogenetic analyses with thousands of taxa and mixed models.
Bioinformatics (Oxford, England), 22(21), 2688–90.
doi:10.1093/bioinformatics/btl446
Sussman, R. W., Green, G. M., & Sussman, L. K. (1994). Satellite imagery, human ecology, anthropology, and deforestation in Madagascar.
Human Ecology, 22(3), 333–354. doi:10.1007/BF02168856
Torsvik, T. H., Amundsen, H., Hartz, E. H., Corfu, F., Kusznir, N., Gaina, C., Doubrovine, P. V., et al. (2013). A Precambrian microcontinent in the Indian Ocean. Nature Geoscience, 6(3), 223–227.
doi:10.1038/ngeo1736
Warren, B. H., Bermingham, E., Bowie, R. C., Prys-Jones, R. P., & Thébaud, C. (2003). Molecular phylogeography reveals island colonization history and diversification of western Indian Ocean sunbirds (Nectarinia: Nectariniidae). Molecular Phylogenetics and
Evolution, 29(1), 67–85. doi:10.1016/S1055-7903(03)00063-0
Wells, N. A. (2003). Some hypotheses on the Mesozoic and Cenozoic paleoenvironmental history of Madagascar. The natural history of
Madagascar, Univ. Chicago Press Chicago.
Vences, M., Vieites, D. R., Glaw, F., Brinkmann, H., Kosuch, J., Veith, M., & Meyer, A. (2003). Multiple overseas dispersal in amphibians.
Proceedings. Biological Sciences / The Royal Society, 270(1532),
Vieites, D. R., Wollenberg, K. C., Andreone, F., Köhler, J., Glaw, F., & Vences, M. (2009). Vast underestimation of Madagascar’s biodiversity evidenced by an integrative amphibian inventory. Proceedings of the
National Academy of Sciences of the United States of America, 106(20), 8267–72. doi:10.1073/pnas.0810821106
Wiens, J. J. (2007). Species delimitation: new approaches for discovering diversity. Systematic biology, 56(6), 875–8.
doi:10.1080/10635150701748506
Wilmé, L., Goodman, S. M., & Ganzhorn, J. U. (2006). Biogeographic evolution of Madagascar’s microendemic biota. Science, 312(5776), 1063–5. doi:10.1126/science.1122806
Yang, Z., & Yoder, A. D. (2003). Comparison of likelihood and bayesian methods for estimating divergence times using multiple gene loci and calibration points, with application to a radiation of cute-looking mouse Lemur Species. Systematic Biology, 52(5), 705–716.
doi:10.1080/10635150390235557
Yee, D. A. (2014). An Introduction to the Dytiscidae: their diversity, historical importance, cultural significance, and other musings. In D.A Yee (Ed.), Ecology, Systematics, and the Natural History of
Predaceous Diving Beetles (Coleoptera: Dytiscidae) (pp. 1–16)
Springer Netherlands
Yoder, A. D., & Nowak, M. D. (2006). Has vicariance or dispersal been the predominant biogeographic force in Madagascar? Only time will tell.
Annual Review of Ecology, Evolution, and Systematics, 37(1), 405–
431. doi:10.1146/annurev.ecolsys.37.091305.110239
Yu Y., Harris A.J., He, X.J. (2013). RASP (Reconstruct Ancestral State in Phylogenies) 2.1 beta.
Yu, Y., Harris, A. J., Blair, C., & He, X. (2015). RASP (Reconstruct Ancestral State in Phylogenies): a tool for historical biogeography.
Molecular phylogenetics and evolution.
Acknowledgements
I would like to thank all the members at the Entomology department for surrounding me during my PhD journey. Without your companionships and your passion for insects this thesis would never be the way it is and I would never be as I am now. Thank you!
My warmest thanks to a very special person Julia S., who not only took care on the sugar level in my blood, but together with Niclas E. took the best care on my Nelson (Mandela (cat)) while I was away for the field work. I will always be in your debt (but the cat is mine! ).
My colleague and warmest friend Yngve B. for his never ending smile and eyes full of optimism. It was always nice to know that you were somewhere working on your insects during those dark long evenings, when I was still struggling with my analyses. Niklas A. for always fixing ethanol (99,5%) for me (I swear I wasn't drinking it ). Hege V. and Mattias F. thank you for your kindness and support. Gunvi L. for those cookies with caramel inside and Sybille H. for the gummy bears on my desk (niam niam niam)! Bo for the nice words about my lunch dishes!
MSL people: Martin, Bodil, Veronica, Keyvan, Jan, Rodrigo, without your assistants none of my papers would be done! And anonymous person(s) who used to steal my PCR robots and gel dye. If you are reading this, the preparation of the gel dye: 1 blue dye (which is usually in the fridge): 4 TBE buffer (which stands on the table) (e.g. 10 ml blue dye + 40ml TBE buffer). It is very easy to make! Trust me!
I want to express my gratitude to my dear Malagasy team mates, Tolotra R., Claudain and Jacquelin H.R, for being such great friends to work with. Jacquelin H.R, you have been a great support to me during our long and not always easy (actually never easy) collecting trips. I think we made a perfect team due to our friendship and we managed to keep it all together even when everything was falling apart.
My big thanks go to my new collecting friends, which I had the honor to work with during my last expedition: Philip P., Kelly M., Robert S., Grey G. and Sandra H. The biggest achievement from that expedition was my new friendships with you guys, especially with Sandra H., whose sweet friendship and company kept the smile on my face during the trip and Grey G., who continuously checks my language - without your corrections my work would be unreadable!!!
My biggest thanks go to my room-mates! Jeanette S. for being the light in the labyrinths of bureaucracy if not for you I would completely be lost among all those papers and requirements and this thesis would never be ac-cepted by Stockholm University! Marika K. for making me laugh to the pain of my stomach! I had never laughed so much in my life! Ah, those cra-zy days!
Seraina K. for all the fun time we spent together and for the sunny breaks we used to have in our office by just sitting in front of the window on our chairs! You have been my dearest friend and a great support I have had dur-ing my PhD. Thank you!
My long-time friend/colleague Jonas S. for making me think would I cross the river full of crocodiles and white sharks for 1 000 000£ ?! Would I prefer to have two heads or twenty fingers?! Well, I will be unemployed soon.. so..a-a.. twenty fingers might help me to swim faster.. and if I survive it I promise to sing for you: “ Jo-ja Jo-nas, i turgu jo-ja….♫ ” .
Tobias M. for opening the secrets of life - there is a life outside the work. Now I see it clearly! Thank you! Andreas, Maria and Markus for making me feel at home. Especially, I want to thank Inger and Staffan for letting me escape and hide in their little country house on Nämdö island! Thank you!
And of course, I would like to thank my mom, for her never ending wonder what I am eating. So, today there will be some thinly sliced fillet steak with sweet roasted onions, served with salad leaves, tomatoes, and roughly crum-bled salty blue cheese, dressed with some olive oil and balsamic vinegar. And of course a lot of coffee with milk!
And finally, I am extremely grateful to my supervisor, Johannes Berg-sten for a great project plan (Fig. 1). I did my best to accomplish it, may-be not always very successfully… Thank you for the opportunity to let me see things I have seen, to experi-ence what I have experiexperi-enced and just simply letting me enjoy my PhD time! I would never have been able to do 4 years of research on the top-ics I love most if not for your unlim-ited patience and professionalism. But the most grateful I am for a wonderful creature you gave me (well, you got rid off it ), little Nelson (cat), who kept me company (by snoring next to my laptop) all
those long nights while I was working on my thesis or deleting all nonsenses I wrote during the day by stepping on the keyboard! I cannot thank you enough for everything!
Finally, I would like to apologize for those I forgot to mention. Please, note, that it is scientifically proven that aging and lack of sleep result in exponen-tial decay of memory lost. Aaaa... What's my name?
