Molly A. Whalen

Systematics of the Rubus fruticosus aggregate (Rosaceae) and other exotic Rubus taxa in Australia

Exotic Rubus taxa in Australia have been revised following consultation with European and North American experts in Rubus, allied with studies of variation in patterns of DNA restriction fragments and morphology. Many of these taxa have names that are applied for the first time in Australia (prefaced with a †). The major focus of the work was the Rubus fruticosus L. aggregate and taxa of this aggregate covered here are R. anglocandicans A. Newton, R. cissburiensis W.C. Barton & Ridd., †R. echinatus Lindl., †R. erythrops Edees & A. Newton, R. laciniatus Willd., R. leightonii Lees ex Leight. †R. leucostachys Schleich. ex Sm., †R. phaeocarpus W.C.R. Watson, R. polyanthemus Lindeb., †R. riddelsdellii Rilstone, †R. rubritinctus W.C.R. Watson, R. ulmifolius Schott (including R. ulmifolius var. ulmifolius and †R. ulmifolius var. anoplothyrsus Sudre), and R. vestitus Weihe, along with two undescribed taxa, Rubus sp. Scott Creek (D.E. Symon 16504) and Rubus sp. Tasmania (J.R. Hosking 1551). Other naturalised taxa are R. alceifolius Poir., R. ellipticus Sm., R. idaeus L., †R. laudatus A. Berger, †R. loganobaccus L.H. Bailey, †R. philadelphicus Blanch., R. roribaccus (L.H. Bailey) Rydb. and R. rugosus Sm. Patterns of morphological and molecular variation among individuals of the R. fruticosus agg. in Australia were examined. In phenetic analyses based on examination of 137 herbarium specimens and 27 morphological characters, taxa showed varying degrees of separation. Some taxa, for example R. anglocandicans and the two varieties of R. ulmifolius, formed distinct groups in these analyses whereas there was considerable overlap among individuals of other species. Fifty M13/HaeIII DNA-banding patterns (phenotypes) were identified among 198 collections from the R. fruticosus agg. across Australia. Thirty-five DNA phenotypes were correlated with 15 taxa of the R. fruticosus agg.; the remaining 15 DNA types correlated poorly or were determined with only a moderate level of confidence. R. anglocandicans, R. echinatus, R. leightonii, R. leucostachys, R. sp. Tasmania, R. ulmifolius and R. vestitus had two or more DNA phenotypes whereas only one DNA phenotype was observed for the remaining eight taxa. Taxa that were more distinct with respect to their DNA phenotypes also tended to be more distinct with respect to morphology based on a Mantel matrix correlation test. Within taxa that were difficult to tell apart morphologically, those sharing the same DNA phenotype were considered members of the same Rubus taxon. These results are discussed in the context of the evolution and ecology of the R. fruticosus agg. in Australia and in relation to the incomplete taxonomy of Rubus in Europe and North America.

A putative hybrid of Eucalyptus largiflorens growing on salt- and drought-affected floodplains has reduced specific leaf area and leaf nitrogen

AputativehybridbetweenEucalyptuslargiflorensF.Muell.andEucalyptusgracilisF.Muell.,calledgreenbox, hasattractedattentionforitsabilitytogrowonthesalt-anddrought-affectedChowilla floodplainoftheMurrayRiverinSouth Australia. Relationships between carbon isotope discrimination (D 13 C) and the ratio of substomatal to ambient CO2 (ci/ca) indicated that green box was not as water use efficient as E. largiflorens. Specific leaf area of green box and E. gracilis was significantly lower compared with E. largiflorens (38.38 and 36.96 versus 43.71cm 2 g -1 ). Leaf nitrogen for green box and E.graciliswassignificantlylowercomparedwithE.largiflorens(12.66and11.35versus15.07mgg -1 dryweight,P=0.004 and 0.001,respectively) and leaf carbon ofE.gracilis was significantly higher compared with green box andE.largiflorens (541.75versus514.90and519.82mgg -1 dryweight,P=0.002and0.011respectively).Thereweresignificantly(P=0.016) more occurrences of elevated ci/ca below a minimum gs in E. gracilis compared with E. largiflorens, with green box being intermediate (means=21.6, 6.8 and 9.4). After 10 years, E. largiflorens trunk circumference had significantly increased (P=0.017)andheighthadsignificantlydecreased(P=0.026)duetovisibledieback.GreenboxandE.gracilisgrewslower, conserving resources, illustrating a useful strategy to consider when choosing plants for revegetation efforts.

Isolation of microsatellites from Baumea juncea (Cyperaceae)

We developed 13 polymorphic microsatellite loci for the rhizomatous perennial sedge Baumea juncea. The number of alleles per locus ranged from two to six, with averages of 3.0 and 3.2 at two South Australian populations. Observed and expected heterozygosity at variable loci within each population ranged from 0.034 to 1.000 and 0.034 to 0.766, respectively. These markers will allow the identification of genets, as well as the study of spatial genetic structure and genetic differentiation among populations.

Morphological variation and phylogenetic relationships within Eucalyptus series Subulatae (Myrtaceae) of southern Australia

Morphological variation within Eucalyptus series Subulatae, a group of mallee and woodland tree taxa distributed across southern Australia, was assessed by adult and seedling characteristics. A phenetic study included a total of 51 adult morphological characters and 37 seedling characters, which were assessed for 564 individuals representing 163 populations, covering the broad geographical distribution of the series. All taxa included in the series by either Johnson and Hill (1999) or Brooker (2000) were included together with E. angustissima, E. cooperiana, E. falcata and E. salmonophloia, all of which have been at times included in, or considered closely related to, E. ser. Subulatae. The phenetic analyses indicate that the four subseries of Brooker (2000) are morphologically distinct, although their distinctiveness is only evident from seedling characters. The two subseries that Johnson and Hill (1999) recognise (subser. Flocktonianosae and Transcontinentalosae) corresponding to Brooker’s (2000) subser. Decurrentes, are more weakly defined. Southern populations of E. dolichocera are not considered here to belong to this species but rather to belong to a different subseries based on seedling morphology. A phylogenetic analysis of 44 morphological characters and 23 species of E. ser. Subulatae and 24 species from variously related taxa suggest that E. ser. Subulatae may not be monophyletic. Eucalyptus brockwayi and E. salmonophloia are basal to E. ser. Subulatae and all of the other taxa included in the analysis. Within E. ser. Subulatae, subsers Oleaginae and Spirales are both monophyletic. Both the phylogenetic and phenetic analyses strongly suggest that E. brockwayi is unique in several characters, including some not previously recognised, and is best placed in a monotypic series. The position of E. aspersa remains unresolved, but is probably best retained in E. subser. Decussatae. A key to the subseries of E. ser. Subulatae is presented and putative intersubserial, interserial and intersectional hybrids involving the series are cited.

A taxonomic revision and morphological variation within Eucalyptus series Subulatae subseries Oleaginae (Myrtaceae), including the oil mallee complex, of south-western Australia

Morphological variation within Eucalyptus series Subulatae subseries Oleaginae, a group of mallee taxa distributed in south-western Australia, was assessed by adult and seedling characteristics. A total of 36 adult morphological characters and 20 seedling characters was included in phenetic analyses of 82 individuals representing 29 populations covering all the taxa and the broad geographical distribution of the series. The phenetic analyses indicate relatively weak separation of all previously recognised taxa within E. subser. Oleaginae, except for E. longissima (previously referred to as E. grasbyi), which is distinctive in its seedling morphology, and to a lesser degree, E. ultima, which is also most easily distinguished by seedling morphology. E. eremicola, E. peeneri and E. sublucida are not well differentiated from one another, differing significantly only in adult leaf colour and gloss. The three previously recognised species of the oil mallee complex (E. kochii, E. plenissima and E. ‘horistes’) are also only weakly differentiated from one another. Two variants within E. kochii were observed to differ from each other in several adult vegetative and floral characteristics and are recognised at the subspecific level. A new taxonomy for E. subser. Oleaginae is presented, based on phenetic analyses combined with extensive field, glasshouse and herbarium examination of all taxa in the subseries. A new species is described (E. longissima Nicolle sp. nov.) to accommodate populations of mallees previously erroneously referred to E. grasbyi. New combinations are made for E. peeneri [E. eremicola subsp. peeneri (Blakely) Nicolle comb. nov.] and E. oleosa var. borealis [E. kochii subsp. borealis (C. Gardner) Nicolle comb. nov.] and two new subspecies of E. kochii are described (subspp. amaryssia Nicolle and yellowdinensis Nicolle). Eucalyptus sublucida is considered to be synonymous with E. eremicola subsp. peeneri. Keys to the taxa within the subseries are presented.

Geographic variation in extrafloral nectaries in the ant-associated plant genus Adriana (Euphorbiaceae) and its relationship to water availability

The relationship between geographic variation in extrafloral nectary size and climatic variation was assessed in three taxa within Adriana (Euphorbiaceae). In each taxon, there was a significant association between gland size and moisture-related climatic variables. A glasshouse experiment was conducted to examine the effect of water stress on nectary longevity, and it was found that the secretory activity of nectaries declined more rapidly on drought-stressed plants than on watered plants.

A taxonomic revision and morphological variation within Eucalyptus series Subulatae subseries Spirales (Myrtaceae) of southern Australia

Variation in adult and seedling morphology within Eucalyptus series Subulatae subseries Spirales, a group of mallee and woodland tree taxa distributed across southern Australia, was examined. A total of 35 adult morphological characters and 13 seedling characters was included in the phenetic analyses of 150 individuals representing 40 populations covering all the taxa and the broad geographical distribution of the subseries and also including E. brockwayi and E. salmonophloia for comparative purposes. Based on phenetic analyses of adult and seedling characters, six taxa are recognised within E. subser. Spirales and seedling characters are important in delimiting these taxa. Of the seven subspecies of E. oleosa described by Johnson and Hill (1999), two subspecies, E. oleosa subsp. oleosa and subsp. repleta could not be distinguished from one another based on either adult or seedling morphology. Similarly, the three subspecies, E. oleosa subsp. ampliata, wylieana and victima, could not be distinguished. A new taxonomy for E. subser. Spirales based on phenetic analyses combined with extensive field, glasshouse and herbarium examination of all taxa in the subseries is presented. Six terminal taxa in the subseries are recognised, viz. E. delicata, E. longicornis and E. oleosa with subspp. oleosa, ampliata, corvina and cylindroidea. Within E. oleosa, E.oleosa subsp. oleosa and subsp. ampliata are not readily distinguishable on the basis of adult morphology alone. Eucalyptus oleosa subsp. repleta is synonymised with subsp. oleosa, and E. oleosa subspp. wylieana and victima with subsp. ampliata. Keys to the taxa of the subseries are presented.

An experimental study of the pollination biology of the perennial halophyte Frankenia pauciflora var. gunnii (Frankeniaceae) in a South Australian salt marsh

Relatively little is known about the pollination biology or breeding systems of entomophilous plant species in salt-marsh habitats. In the present study, the pollination biology of the perennial halophyte Frankenia pauciflora DC. var. gunnii Summerh. (Frankeniaceae) was investigated in a salt marsh near Goolwa, South Australia. The flowering peak of this species occurs in summer and flowers are visited by a diverse array of insects in the Diptera, Hymenoptera and Lepidoptera. An experiment was performed to test for the presence of pollen limitation in the field. Flowers to which outcross pollen was added produced significantly more fruits and significantly more seeds per fruit than did unmanipulated control flowers. Average seed weight per fruit did not differ significantly between control and pollen-addition flowers. In glasshouse experiments, flowers to which cross-pollen was added produced significantly more fruit than did flowers to which self-pollen was added, suggesting the presence of a facultatively xenogamous breeding system.

Microsatellite markers for the freshwater shrimp Paratya australiensis (Atyidae)

Paratya australiensis is a common freshwater shrimp found in most catchments in south-eastern Australia. Microsatellite loci were isolated from a partial genomic library created using 454 sequencing. Of the 25 tested, 13 were found to be polymorphic, however, for pooling purposes only 12 were used for genotyping. The number of alleles per locus varied from 2 to 14 in a population from Wakefield River, South Australia, and the mean (range) observed and expected heterozygosity were 0.512 (0.136–0.909) and 0.590 (0.165–0.788), respectively across all alleles. These microsatellites will be used to study the genetic structure of populations in two catchments in South Australia.

Ant defence of a dioecious shrub, Adriana quadripartita (Euphorbiaceae), with extrafloral nectaries

Extrafloral nectaries (EFNs) can function to indirectly reduce herbivory by attracting ants that interfere with or predate on herbivorous insects. So as to examine the efficacy of ants as defenders of plants with extrafloral nectaries (EFNs) against herbivores, an ant-exclusion experiment was conducted on plants of the dioecious species Adriana quadripartita (Euphorbiaceae). The experiment was conducted on Torrens Island, South Australia, and adds to previous work that examined the associations among adrianas, ants and invertebrate herbivores at several locations across Australia. The aim of the present study was to investigate whether ants that tend EFNs reduce herbivore numbers and levels of herbivory, and whether this ant–plant–herbivore interaction differs between male and female A. quadripartita plants. The abundances of ants and invertebrate herbivores were overall greater on male than on female plants. The exclusion of ants from branches resulted in an increase in herbivore abundance on male and female plants; however, ant exclusion resulted in an increase of foliar damage on male plants only. Whereas previous research on A. tomentosa shows that ants can decrease herbivore abundance, the results here provide the first evidence that EFNs on adrianas can function to decrease foliar herbivory.