Multilocus Resolution of Phylogeny and Timescale in the Extant Adaptive Radiation of Hawaiian Honeycreepers

Adaptive radiation is the rapid evolution of morphologically and ecologically diverse species from a single ancestor. The two classic examples of adaptive radiation are Darwin's finches and the Hawaiian honeycreepers, which evolved remarkable levels of adaptive cranial morphological variation. To gain new insights into the nature of their diversification, we performed comparative three-dimensional geometric morphometric analyses based on X-ray microcomputed tomography (µCT) scanning of dried cranial skeletons. We show that cranial shapes in both Hawaiian honeycreepers and Coerebinae (Darwin's finches and their close relatives) are much more diverse than in their respective outgroups, but Hawaiian honeycreepers as a group display the highest diversity and disparity of all other bird groups studied. We also report a significant contribution of allometry to skull shape variation, and distinct patterns of evolutionary change in skull morphology in the two lineages of songbirds that underwent adaptive radiation on oceanic islands. These findings help to better understand the nature of adaptive radiations in general and provide a foundation for future investigations on the developmental and molecular mechanisms underlying diversification of these morphologically distinguished groups of birds.This article is part of the themed issue ‘Evo-devo in the genomics era, and the origins of morphological diversity’.

The monophyly and phylogeny of the adaptive radiation of Hawaiian finches (Fringillidae: Drepanidini; honeycreepers, auct.) were studied using parsimony analysis of comparative osteology, combined with Templeton (Wilcoxon signed-ranks) tests of alternative phylogenetic hypotheses. Eighty-four osteological characters were scored in 59 terminal taxa of drepanidines, including 24 fossil forms, and in 30 outgroup species. The optimal phylogenetic trees show considerable agreement, and some conflict, with independently derived ideas about drepanidine evolution. The monophyly of a large Hawaiian radiation was upheld, although one fossil taxon from Maui fell outside the drepanidine clade. The finch-billed species were placed as basal drepanidine taxa, and continental cardueline finches (Carduelini) were identified as the radiation’s closest outgroups. The study found anatomical as well as phylogenetic evidence that the radiation had a finch-billed ancestor. The optimal trees identify the red-and-black plumage group as a clade, and suggest that the tubular tongue evolved only once in the radiation. Because comparative osteology provides too few characters to strongly support all the nodes of the tree, it was helpful to evaluate statistical support for alternative hypotheses about drepanidine relationships using the Templeton test. Among the alternatives that received significant statistical support are a relationship of the drepanidines with cardueline finches rather than with the Neotropical honeycreepers (Thraupini), classification of the controversial genera Paroreomyza and Melamprosops as drepanidines, and a secondary loss of the tubular tongue in Loxops mana. The hypothesis of monophyly for all the Hawaiian taxa in the study was not rejected statistically. The study provides a framework for incorporating morphological and palaeontological information in evolutionary studies of the Drepanidini. © 2004 The Linnean Society of London, Zoological Journal of the Linnean Society, 2004, 141, 207‐255. ADDITIONAL KEYWORDS: adaptive radiation ‐ comparative anatomy ‐ evolution ‐ fossils ‐ Hawaiian honeycreepers ‐ island biogeography ‐ morphology ‐ palaeontology ‐ Passeriformes.

Evolutionary ecology and radiation of Hawaiian passerine birds

Reappraising adaptive radiation

The Hawaiian archipelago is isolated in the central Pacific and consists of 7 large islands and a chain of low coral atolls and small rocky islets that extend in a long arc from Hawaii Island in the southeast to Kure Atoll in the northwest. The archipelago is the most isolated island system in the world, separated from the nearest continental landmass by more than 2000 miles of ocean.1 The islands range in elevation from just above sea level for atolls in the Northwestern Hawaiian Islands to peaks that exceed 4000 m on Hawaii Island. The interaction of extreme topographic relief, trade winds, and local climatic patterns creates a wide diversity of habitats, ranging from alpine deserts on the highest peaks to montane rain forests with precipitation exceeding 7600 mm per year. The endemic passerine avifauna of the Hawaiian Islands, particularly the endemic Hawaiian hon eycreepers (subfamily Drepanidinae) is often heralded as an outstanding example of adaptive radiation, equal to Darwin's finches from the Galapagos Islands in terms of diversity of bill types and number of species that descended from a common founder.2 From an initial colonization by only a few individuals of a single ancestral cardueline finch, this group radiated throughout the diverse habitats on the islands, specializing on a variety of food resources that included nectar, fruits, and insects. Based on recent studies of subfossils, the diversity of this group may have reached 20 genera with more than 50 species prior to human contact with the islands.3 Remarkably, a new genus and species of honeycreeper, the po'ouli {Melamprosops phaeosoma), was described in the 1970s from remote rain forests on Maui.4 Today, the endemic Hawaiian avifauna faces one of the highest rates of extinction in the world.

Adaptive Radiations: Insights From Evo-Devo

Adaptive radiation describes the divergence of an ancestral taxon into multiple, phenotypically diverse species, adapted to a range of ecological niches by means of natural selection. The process is recognized as a fundamental reason for the origin of biodiversity. The main driver of adaptive radiation is ecological opportunity, though the specific agents are often poorly understood with the exception of some iconic lineages. Many well-studied adaptive radiations are island endemics, which makes island systems an ideal study system for adaptive radiation. Oceanic islands represent discrete replicates of the evolutionary process, as they are isolated, comparatively small, and often topographically complex. Species communities are formed by colonization and in situ diversification. The Hawaiian Islands are the most isolated archipelago on earth and home to a range of adaptively radiating lineages. The islands form as the Pacific plate passes over a magmatic hotspot with the eight current high islands originating within the last ca. 5-6 million years and the majority of the native biodiversity diverging within that time. The genus Melicope colonized numerous archipelagos throughout the Pacific including the Hawaiian Islands, where the lineage comprises currently 54 endemic species and represents the largest radiation of woody plants on the islands. Most species are single-island endemics and adapted to a variety of habitat types and elevational ranges. The lineage is monophyletic with an estimated crown age predating the rise of the current high islands, the oldest of which originated approximately 5 million years ago. As for many adaptively radiating lineages, phylogenetic inference based on Sanger sequencing has not been sufficient to resolve species or deeper level relationships in Hawaiian Melicope. Recent years have seen development of high throughput sequencing methods and their increasing application to solve recalcitrant relationships. In this thesis, I examined the evolutionary trajectory of the Hawaiian Melicope adaptive radiation. I investigated the so-called ‘island syndrome’, which describes a set of traits commonly characterizing successful island colonizers, including recent polyploidy and shifts associated with subsequent establishment, in Hawaiian Melicope. I utilized restriction site-associated high throughput sequencing (RAD-seq) to reconstruct species relationships and historical biogeography in the lineage and estimate diversification rates and the impact of habitat adaption on species divergence. RAD-seq datasets provided unprecedented resolution of species relationships in Hawaiian Melicope. However, the size and complexity of high throughput sequencing datasets require a high computational effort, which currently limits the applicability of algorithms for phylogenetic inference to concatenated analysis or site-specific coalescence-based methods. I employed both methods and found them to result in incongruent relationships for the backbone of the Hawaiian Melicope topology. Concatenation violates the assumptions of the multispecies coalescent model, while site-based methods are statistically inconsistent but less accurate in simulated and empirical datasets. Considering the increased accuracy of the concatenated approaches as evaluated by quartet concordance methods and the synergistic effect of concatenation, I concluded that results of concatenated analysis reflect the relationships of Hawaiian Melicope best. Results of flow cytometric screening of 32 Hawaiian species, representing 66% of the described diversity, and literature searches indicate that the ancestor of Hawaiian Melicope did not show traits associated with successful colonizers. The genus seemingly retained colonization success while exhibiting a combination of traits that typically characterize well-established island specialists. In particular, the ancestral Melicope colonist was not a recent polyploid. Neopolyploidy increases evolutionary flexibility and thus enhances chances for establishment and adaption. In Hawaiian Melicope flexibility is possibly facilitated by introgressive hybridization events. Phylogenetic reconstruction based on RAD-seq datasets provides evidence for two ancient and several recent introgression events. Extant Hawaiian Melicope are divided into five fully supported main clades, two of which correspond to morphologically circumscribed infrageneric groups, whereas three morphologically defined taxonomic units are not monophyletic. All in all, 24 species were included with multiple samples, four of which were resolved as non-monophyletic. Finally, I confirmed that the Melicope radiation endemic to the Marquesas Islands originated from the Hawaiian radiation. These results highlight the necessity for a taxonomic revision in the lineage. Estimated divergence times revealed that the Hawaiian archipelago was colonized prior to the origin of the current high islands. Inter-Island colonization patterns largely follow the progression rule from older to younger islands, but back colonizations to older islands occurred. Extant diversity results from recent divergence of a small number of taxa prevailing through the bottlenecks represented by the origin and colonization of the high islands. Long internal branches and estimated diversification rates indicate a high extinction rate, possibly related to the consequences of volcanic activity and the impact of glacial cycles. Consequently habitat types that are more vulnerable to climatic changes, i.e. dry ranges and bogs show high speciation and extinction rates. Increased rates of diversification are linked to habitat dissection and frequent ecological trait shifts.

The diversifications of Darwin's finches and Hawaiian honeycreepers are two text-book examples of adaptive radiation in birds. Why these two bird groups radiated while the remaining endemic birds in these two archipelagos exhibit relatively low diversity and disparity remains unexplained. Ecological factors have failed to provide a convincing answer to this phenomenon, and some intrinsic causes connected to craniofacial evolution have been hypothesized. The tight coevolution of the beak and the remainder of the skull in diurnal raptors and parrots suggests that integration may be the prevalent condition in landbirds (Inopinaves). This is in contrast with the archetypal relationship between beak shape and ecology in Darwin's finches and Hawaiian honeycreepers, which suggests that the beak can adapt as a distinct module in these birds. Modularity has therefore been proposed to underpin the adaptive radiation of these groups, allowing the beak to evolve more rapidly and freely in response to ecological opportunity. Here, using geometric morphometrics and phylogenetic comparative methods in a broad sample of landbird skulls, we show that craniofacial evolution in Darwin's finches and Hawaiian honeycreepers seems to be characterized by a tighter coevolution of the beak and the rest of the skull (cranial integration) than in most landbird lineages, with rapid and extreme morphological evolution of both skull regions along constrained directions of phenotypic space. These patterns are unique among landbirds, including other sympatric island radiations, and therefore counter previous hypotheses by showing that tighter cranial integration, not only modularity, can facilitate evolution along adaptive directions.

The Hawaiian honeycreepers are a dramatic example of adaptive radiation but contrast with the four other songbird lineages that successfully colonized the Hawaiian archipelago and failed to undergo similar diversification. To explore the processes that produced the diversity dichotomy in this insular fauna, we compared clade age and morphological diversity between the speciose honeycreepers and the comparatively depauperate Hawaiian thrushes. Mitochondrial-DNA-based genetic distances between these Hawaiian clades and their continental sister taxa indicate that the ancestral thrush colonized the Hawaiian Islands as early as the common ancestor of the honeycreepers. This similar timing of colonization indicates that the marked difference in diversity between the Hawaiian honeycreeper and thrush clades is unlikely to result from differences in these clades' tenures within the archipelago. If time cannot explain the contrasting diversities of these taxa, then an intrinsic, clade-specific trait may have fostered the honeycreeper radiation. As the honeycreepers have diversified most dramatically in morphological characters related to resource utilization, we used principal components analyses of bill characters to compare the magnitudes of morphological variation in the ancestral clades from which the Hawaiian honeycreeper and thrush lineages are derived, the Carduelini and Turdinae respectively. Although the Carduelini share a more recent common ancestor and have a lower species diversity than the Turdinae, these finch-like relatives of the honeycreepers exhibit significantly greater variation in bill morphology than do the continental relatives of the Hawaiian thrushes. The higher magnitude of morphological variation in the non-Hawaiian Carduelini suggests that the honeycreepers fall within a clade exhibiting a generally high evolutionary flexibility in bill morphology. Accordingly, although the magnitude of bill variation among the honeycreepers is similar to that of the entire passerine radiation, this dramatic morphological radiation represents an extreme manifestation of a general clade-specific ability to evolve novel morphologies.

Most studies of adaptive radiation in animals focus on resource competition as the primary driver of trait divergence. The roles of other ecological interactions in shaping divergent phenotypes during such radiations have received less attention. We evaluate natural enemies as primary agents of diversifying selection on the phenotypes of an actively diverging lineage of gall midges on tall goldenrod. In this system, the gall of the midge consists of a biotrophic fungal symbiont that develops on host-plant leaves and forms distinctly variable protective carapaces over midge larvae. Through field studies, we show that fungal gall morphology, which is induced by midges (i.e., it is an extended phenotype), is under directional and diversifying selection by parasitoid enemies. Overall, natural enemies disruptively select for either small or large galls, mainly along the axis of gall thickness. These results imply that predators are driving the evolution of phenotypic diversity in symbiotic defense traits in this system and that divergence in defensive morphology may provide ecological opportunities that help to fuel the adaptive radiation of this genus of midges on goldenrods. This enemy-driven phenotypic divergence in a diversifying lineage illustrates the potential importance of consumer-resource and symbiotic species interactions in adaptive radiation.

Adaptive radiation is the rapid evolution of morphologically and ecologically diverse species from a single ancestor. The two classic examples of adaptive radiation are Darwin's finches and the Hawaiian honeycreepers, which evolved remarkable levels of adaptive cranial morphological variation. To gain new insights into the nature of their diversification, we performed comparative three-dimensional geometric morphometric analyses based on X-ray microcomputed tomography (μCT) scanning of dried cranial skeletons. We show that cranial shapes in both Hawaiian honeycreepers and Coerebinae (Darwin's finches and their close relatives) are much more diverse than in their respective outgroups, but Hawaiian honeycreepers as a group display the highest diversity and disparity of all other bird groups studied. We also report a significant contribution of allometry to skull shape variation, and distinct patterns of evolutionary change in skull morphology in the two lineages of songbirds that underwent adaptive radiation on oceanic islands. These findings help to better understand the nature of adaptive radiations in general and provide a foundation for future investigations on the developmental and molecular mechanisms underlying diversification of these morphologically distinguished groups of birds.This article is part of the themed issue ‘Evo-devo in the genomics era, and the origins of morphological diversity’.

The Hawaiian Honeycreepers are typified by nectar feeding, their bright colouration, and canary-like songs. They are considered one of the finest examples of adaptive radiation, even more diverse than Darwin's Galapagos finches, as a wide array of different species has evolved in all the different niches provided by the Hawaiian archipelago. The book will therefore be of interest to evolutionary biologists and ecologists, as well as professional ornithologists and amateur bird watchers. As with the other books in the Bird Family of the World series, the work is divided into two main sections. Part I is an overview of the Hawaiian Honeycreeper evolution and natural history and Part II comprises accounts of each species. The author has produced his own outstanding illustrations of these birds to accompany his text.

Sexual dimorphism is widespread and substantial throughout the animal world. It is surprising, then, that such a pervasive source of biological diversity has not been integrated into studies of adaptive radiation, despite extensive and growing attention to both phenomena. Rather, most studies of adaptive radiation either group individuals without regard to sex or focus solely on one sex. Here we show that sexual differences contribute substantially to the ecomorphological diversity produced by the adaptive radiations of West Indian Anolis lizards: within anole species, males and females occupy mostly non-overlapping parts of morphological space; the overall extent of sexual variation is large relative to interspecific variation; and the degree of variation depends on ecological type. Thus, when sexual dimorphism in ecologically relevant traits is substantial, ignoring its contribution may significantly underestimate the adaptive component of evolutionary radiation. Conversely, if sexual dimorphism and interspecific divergence are alternative means of ecological diversification, then the degree of sexual dimorphism may be negatively related to the extent of adaptive radiation.

Speciation and the interactions between recently diverged species are thought to be major causes of ecological and morphological divergence in evolutionary radiations. Here, we explore the extent to which geographical overlap and time since speciation may promote divergence in marine species, which represent a small fraction of currently published studies about the patterns and processes of speciation. A time-calibrated molecular phylogeny of New World haemulid fishes, a major radiation of reef and shore fishes in the tropical West Atlantic and East Pacific, reveals 21 sister species pairs, of which eight are fully sympatric and 13 are allopatric. Sister species comparisons show a non-significant relation between most of the phenotypic traits and time since divergence in allopatric taxa. Additionally, we find no difference between sympatric and allopatric pairs in the rate of divergence in colour pattern, overall body shape, or functional morphological traits associated with locomotion or feeding. However, sympatric pairs show a significant decrease in the rate of divergence in all of these traits with increasing time since their divergence, suggesting an elevated rate of divergence at the time of speciation, the effect of which attenuates as divergence time increases. Our results are consistent with an important role for geographical overlap driving phenotypic divergence early in the speciation process, but the lack of difference in rates between sympatric and allopatric pairs indicates that the interactions between closely related species are not dominant drivers of this divergence.

First collected by Charles Darwin in the Galapagos Archipelago, the Geospizinae, or finches, have rightly been celebrated as a classic instance of the workings of evolution through natural selection. Among birds, Darwin's finches are rivaled only by the Hawaiian honeycreepers (Drepanididae) as a microcosmic exemplification of the principle of adaptive evolutionary radiation. Although the Drepanididae have undergone more evolution and adaptive radiation than the Geospizinae, the latter are in some ways more valuable to omithologists. Their special interest today, writes David Lack, in providing the best example, in birds, of an adaptive radiation into different ecological niches that is sufficiently recent, geologically speaking, for intermediate and transitional forms to have survived (1964:178). The Galapagos Archipelago, where Darwin spent five weeks collecting these finches during the voyage of H.M.S. Beagle (1831-1836), comprises sixteen principal islands located on the equator some six hundred miles west of Ecuador (Fig. 1). The islands, most of which are several million years old, are wholly volcanic in origin and have never been connected to the mainland. Darwin's finches were evidently one of the earliest colonists of the Galapagos, since their degree of evolutionary complexity thirteen species distributed among four genera is unmatched by any other avian group in this archipelago. A fourteenth species, belonging to yet another genus, inhabits Cocos Island, four hundred miles to the northeast. Unlike other endemic species of Galapagos birds, the Geospizinae no longer have any close relatives on the American mainland. They are therefore classed in their own separate tribe or subfamily, which is placed with the Emberizidae.1 Being one of the earliest colonists of the Galapagos Islands, the ancestral form of Darwin's finches found an environment in which the types of niches occupied by other, diverse birds on the continent