Andropogon gayanus (gamba grass, Poaceae) is a large, tropical grass native to Africa, but introduced as a pasture plant across the tropics worldwide. It is considered invasive in northern Australia, and a research program into the potential of biological control of the species has recently been initiated. Because biological control agents are more likely to pose risks to non-target species closely related to the target weed than to distantly related species, risk management and experimental design in weed biological control require information on the evolutionary relationships of the target weed. Here, we inferred target sequence capture phylogenies of Poaceae tribe Andropogoneae with a focus on Australian lineages of the tribe, combining newly generated and previously published data. Our results place some lineages treated as incertae sedis in the most recent classification of Andropogoneae, including Australian endemics. Clausospicula extensa was retrieved as sister to Chrysopogon, and Spathia neurosa was nested in Dichanthium, but the position of Pseudopogonatherum was inconsistent across our analyses. Our analyses imply that subtribes Apludinae and Anthistiriinae as currently recognised may be non-monophyletic. Only 2 of 25 replicate specimen pairs of the same species failed to group together, confirming a high degree of reproducibility despite differences in bioinformatic approaches between our study and that used for previously published data. The revised phylogeny presented here supports the development of risk assessment plans for biological control and provides additional resources for broader Andropogoneae phylogeny.
Alpinia cumingii was described by Karl Moritz Schumann from the Philippines in 1904. Two remarkably similar species, also from the Philippines, Alpinia penduliflora and Alpinia pendula, were described respectively in 1909 by Henry Nicholas Ridley and 1928 by Eduardo Quisumbing and Elmer Drew Merrill, without any reference to A. cumingii, and placed in different sections of Alpinia. A fourth morphologically similar species was collected in New Guinea during fieldwork in 2008. Using molecular data, we aim to determine the phylogenetic relationships of the four species and integrate morphological data to assess the rank at which these should be recognised. We collected new material close to the three type localities in the Philippines and used these new collections as a basis for an updated morphological description, including the previously undescribed species from New Guinea. Molecular analyses revealed that all four species grouped within the Alpinia eubractea clade both in the internal transcribed spacer (ITS) and trnK–matK regions with strong support. The undescribed species from New Guinea appeared as a sister taxon to the Philippine species in the ITS tree, whereas using the trnK–matK region, it grouped with Vanoverberghia, both with weak support. We name the species Alpinia oktediensis, referring to the type locality. Our data support A. pendula as a distinct species, whereas A. cumingii and A. penduliflora are considered conspecific. We also report a collection from Karekelong Island, as evidence for a new record of A. cumingii in the flora of Indonesia. We provide provisional conservation assessments for each species.
Capparis comprises ~145 species, of which ~21 occur in Australia; however, the relationships and taxonomic status of Australian Capparis taxa remain to be tested. We present phylogenies of all Australian Capparis taxa by analysing Angiosperms353 loci using coalescent and concatenated approaches. All trees resolve Capparis as monophyletic, with a whole-genome duplication (WGD) event detected at the crown. Capparis sect. Capparis and sect. Busbeckea are monophyletic but sect. Monostichocalyx is non-monophyletic. The relationships of species within sect. Busbeckea are poorly supported due to rapid radiation following ancient WGD. The relationships of taxa within sect. Capparis and the clades of sect. Monostichocalyx are well supported, with some incomplete lineage sorting. Capparis spinosa is geographically, morphologically and phylogenetically structured across northern Australia. Based on these results, we describe five new species and two new subspecies of Capparis, bringing the total number of species in Australia to 26. Capparis xylocarpa, C. megacarpa, C. loxophleba and C. splendidissima are newly described. Capparis loranthifolia var. bancroftii is raised to species level as C. bancroftii. Capparis spinosa subsp. nummularia is re-circumscribed, and two new subspecies, C. spinosa subsp. formicosa and C. spinosa subsp. insularis are described. We provide a key for all Australian taxa.
Dinckleria opaca M.A.M.Renner & M.Paulsen is described from Anglesea amber for fossils previously reported in the literature as an undetermined liverwort of the family Plagiochilaceae. The fossils are fragmentary and critical character systems are not preserved, complicating generic attribution. Membership in Dinckleria is supported by the ovate-elliptic and nearly symmetrical leaf shape, the trilobed leaf apex, and the vestigial, unlobed, triangular underleaves. Fossil preservation is poor, with no evident impregnation of plant material by resins, such that the fossil comprises the amber-air interface capturing the volume of the included plant material, that in turn contains highly degraded and disaggregated remnants of the plant itself. The amber-air interface has trace impressions of internal cell walls standing proud above the collapsed and concave external walls, typical of dehydrated plant material, and appears to have experienced minor deformation and distortion during amber formation, such that these leaf cell impressions and leaf cell surfaces are incompletely preserved. The age of the fossil, at 42 Ma, contrasts with the shallow phylogenetic divergences within D. pleurata, the species currently extant in Victoria and also south-east Queensland and New Zealand, suggesting a complex biogeographic history for the Dinckleria lineage in south-east Australia involving extinction and recolonisation by long-distance dispersal.
