Table of Contents

13 December 2018, Volume 56 Issue 6
Cover illustration: Dispersal pattern of Melicope (Rutaceae, Citrus family) in the Pacifi c. Colors of arrows are according to the main phylogenetic lineages within the genus. See Appelhans et al., pp. 576–599 in this issue.
  • Marc S. Appelhans, Jun Wen, and Warren L. Wagner
    J Syst Evol. 2018, 56(6): 573-575.
  • Research Article
  • Marc S. Appelhans, Jun Wen, Marco Duretto, Darren Crayn, Warren L. Wagner
    J Syst Evol. 2018, 56(6): 576-599.
    The genus Melicope (Rutaceae) occurs on most Pacific archipelagos and is perfectly suited to study Pacific biogeography. The main goal was to infer the age, geographic origin and colonization patterns of Melicope and its relatives. We sequenced three nuclear and two plastid markers for 332 specimens that represent 164 species in 16 genera of Rutaceae. Phylogenetic reconstruction, molecular dating, ancestral area reconstruction and diversification analyses were carried out. The two main clades (Acronychia‐Melicope and Euodia) originated in Australasia and their crown ages are dated to the Miocene. Diversification rates differed among the subclades and were lowest in the Euodia lineage and highest in the Hawaiian Melicope lineage. The Malagasy and Mascarene species form a clade, which split from its SE Asian relatives in the Pliocene/Pleistocene. At least eight colonizations to the Pacific islands occurred. The timing of all colonizations except for the Hawaiian group is congruent with age of the island ages. Australia, New Guinea and New Caledonia have been the source of colonizations into the Pacific islands in the Melicope clade. Melicope shows high dispersability and has colonized remote archipelagos such as the Austral and Marquesas Islands each twice. Colonization of islands of the Hawaiian‐Emperor seamount chain likely predates the ages of the current main islands, and the initial colonization to Kaua'i occurred after the splitting of the Hawaiian lineage into two subclades. Wider ecological niches and adaptations to bird‐dispersal likely account for the much higher species richness in the Acronychia‐Melicope clade compared to the Euodia clade.
  • Review
  • Jonathan P. Price and Warren L. Wagner
    J Syst Evol. 2018, 56(6): 600-620.
    Botanists have long considered the origins of the Hawaiian flora in terms of long‐distance dispersal from particular source areas. We extensively reviewed phylogenetic studies of the Hawaiian angiosperm flora to determine the most likely region of origin for each lineage from a defined set of source areas. We also evaluated dispersal modes of each lineage to assess whether certain dispersal modes are associated with a given source area. The largest source category was Widespread (involving related taxa that extend across more than one region), although many of these comprised native non‐endemic species, and accounted for little of the total species diversity (after accounting for in situ speciation). The next largest source regions were Indo‐Malayan and Neotropical. Comparatively few lineages originated from the East Asian region, although these include the single largest lineage. Lineages originating in the Indo‐Malayan region predominantly arrived via Pacific Islands, whereas dispersal from all other regions appears to have been mostly direct. Compared with previous analyses, we found a higher proportion of lineages originating in the Neotropics and temperate North America. Widespread origins were positively associated with dispersal via flotation on water, whereas other origins were associated with dispersal by birds, either through internal transport or external adhesion. We identified thirty‐one potential cases of dispersal out of Hawaii to other islands. Our assessment is complicated by lineages with ancient origins, with further complications likely stemming from hybridization events. Overall, numerous lineages including some distinctive endemic genera have not had sufficient phylogenetic study to determine an origin.
  • Research Articles
  • Claudia Paetzold, Michael Kiehn, Kenneth R. Wood, Warren L. Wagner, and Marc S. Appelhans
    J Syst Evol. 2018, 56(6): 621-636.
    Oceanic islands are unique in their species composition, which is defined by arrival of colonizers via long distance dispersal followed by establishment of species followed in some cases by adaptive radiation. Evolutionary biologists identified traits facilitating successful colonization of islands as including polyploidy, self‐compatibility, herbaceousness and ability for long‐distance dispersal. Successful establishment and evolutionary diversification of lineages on islands often involves shifts to woodiness and shifts in methods of outcrossing as well as changes in dispersal ability. 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. The wide distributional range of the genus illustrates its high dispersibility, most likely due to adaption to bird dispersal. Here we investigate ploidy in the genus using flow cytometry and chromosome counting. We find the genus to be paleopolyploid with 2n = 4x = 36, a ploidy level characterizing the entire subfamily Amyridoideae and dating back to at least the Palaeocene. Therefore Hawaiian Melicope have not undergone recent polyploidization prior to colonization of the islands. Thus Melicope retained colonization success while exhibiting a combination of traits that typically characterize well established island specialists while lacking some traits associated to successful colonizers.
  • Liang Zhao, Daniel Potter, Yuan Xu, Pei-Liang Liu, Gabriel Johnson, Zhao-Yang Chang, Jun Wen
    J Syst Evol. 2018, 56(6): 637-651.
    The goals of this study were to reconstruct the phylogeny of Prunus subgenus Laurocerasus section Mesopygeum and to provide a preliminary assessment of its spatio‐temporal diversification in the Malesian region. We inferred the phylogeny using nuclear ITS and ETS and plastid psbA‐trnH, rps16, rpl16, and trnC‐petN sequences. Our analyses support the monophyly of sect. Mesopygeum. Within sect. Mesopygeum, we identified four main subclades: (i) Prunus lancilimba from continental Asia; (ii) Prunus ruthii from Malay Peninsula; (iii) a subclade comprising species from areas centered on the Sunda shelf and also a few species from continental Asia and Wallacea; and (iv) a subclade composed of species from areas of the Sahul shelf, with a small number of taxa also from areas of the Sunda shelf, continental Asia, and the Philippines. We estimated that sect. Mesopygeum originated in continental Asia at c. 44.71 Mya (95% HPD: 31.66–46.90). Nine dispersals between major geographic areas were inferred. From continental Asia, three and two dispersals were inferred to the Sunda shelf and the Sahul shelf in the mid‐Oligocene, respectively. Two dispersals were inferred from the Sahul shelf region to the Sunda shelf in late Oligecene and early Miocene, respectively. There were also two dispersals inferred from the Sunda shelf region, one to the Philippines and one to Wallacea, in the middle and late Miocene, respectively. The diversification in sect. Mesopygeum was likely driven by active geologic events and orogenies in the Neogene in the Malesian region.
  • Anne L.J. Rutgrink, Michaël Visser, Peter C. van Welzen
    J Syst Evol. 2018, 56(6): 652-662.
    The north and south Moluccas (Indonesia) have very different geotectonic origins and, due to that, a difference in flora is to be expected. The north Moluccas moved westwards along the north coast of New Guinea to their present position, the south Moluccas moved north from Australia. On the other hand, a comparable climate in both areas and (partial) submergence during tectonic movement may have equalized both floras. Collection data from Naturalis Biodiversity Center on 1559 species in 121 families treated in Flora Malesiana were collected for the Moluccas, Sulawesi, and Western New Guinea (latitudes 9.2°S and 5.6°N and longitudes 118.8°E and 141°E) and georeferenced. Species Distribution Models (SDMs) were made, based on least correlated climate and edaphic variables, using only those species that were present in 5 or more grid cells of 5-arc minutes and models were tested for deviation from random. Both areas differ significantly and share only 50%-65% of their species. The 348 significant SDMs differ much less, though still significantly, sharing 91% of the species. Despite strong climatic and edaphic similarities between the North and South Moluccas, they differ greatly in species composition, which is in support of geotectonic reconstructions. The differences between the North and South Moluccas suggest that the continuous dispersal barriers and tectonic backgrounds have influenced their current flora.
  • Joseph T. Miler, Garry Jolley-Rogers, Brent D. Mishler, and Andrew H. Thornhill
    J Syst Evol. 2018, 56(6): 663-667.

    Biodiversity is most commonly measured in taxonomic richness. For example, it is common to describe how diverse a genus or a geographic area is by counting the number of species within them. Phylogenetic diversity (PD), a measurement of the branch lengths in a phylogenetic tree, is a better measure of biodiversity that provides a comparable, evolutionary measure of biodiversity not possible with species counts. Despite its advantages, PD is rarely used as the primary measure of biodiversity. We developed a genus-level phylogeny for nearly 90% of taxonomically described Australian land plants and compared PD to genus richness in multiple clades. The proportion of PD per genera was skewed among clades. Non-angiosperm clades had more PD than expected given the number of genera while angiosperm clades had less PD than expected. For example, ferns comprised only 4.7% of the genera yet 13.0% of the PD, while the angiosperms as a whole comprised 78.9% of the genera but only 62.7% of the PD. It is likely that cultural reasons, such as taxonomic biases, are more important than methodological and biological phenomena in explaining these discrepancies. Regardless of reasons for the observed results, we conclude that a shift towards the use of PD as the primary descriptor of biodiversity will promote an important conceptual shift in biodiversity studies as a quantitative science.