251 Sixteenth Australian Weeds Conference Summary The coelomycetous stem canker pathogen Septocyta ruborum causes stem die-back of European blackberry (Rubus fruticosus agg.). The fungus infects its host by penetrating stomata, stem wounds and possibly crevices associated with hair bases. Inva- sion of vascular bundles by hyphae disrupted internal transportation processes and appears to be the main mechanism for stem decline. Multiple cutting of the stem surface increased infection in whole plants, but the epidermis was vital to infection. Host-testing and field evaluation of S. ruborum as a potential biological control agent for European blackberry will involve histological evidence of infection to assess the risks it poses to non-target plant species. Keywords Rubus fruticosus, biocontrol, weed. INTRODUCTION In southern Australia, European blackberry (Rubus fruticosus L. agg.) causes substantial economic losses in agriculturally productive areas as well as declines in biodiversity values in native vegetation. Although herbicide-based control techniques can provide ef- fective suppression, these have limited application (Bruzzese and Lane 1996). A classical biological control program for blackberry commenced in 1973 and, although issues associated with conflicts of interest delayed implementation, the defoliating rust fungus Phragmidium violaceum (Schultz) Winter was officially released in 1991. Phragmidium violaceum can cause severe defoliation of European blackberry taxa, but its impact is limited by climatic, genetic and taxonomic factors (Adair and Bruzzese 2006). Large areas of blackberry infestations remain without adequate biological control, particularly in native vegetation and regions where the annual rainfall is less than 750 mm per annum. Additional biological control agents are required to provide more comprehensive suppression of black- berry. The coelomycetous stem pathogen, Septocyta ruborum (Lib.) Petrak (purple blotch disease), is one of several organisms identified with potential for bio- logical control (Sagliocco and Bruzzese 2003). It is widespread on R. fruticosus in Europe and can cause significant declines in the health and productivity of Infection process of Septocyta ruborum, a coelomycetous fungus with potential for biological control of European blackberry in Australia Janita Baguant 1 , Robin J. Adair 2,3 and Ann C. Lawrie 1 1 RMIT University, School of Applied Sciences, PO Box 71, Bundoora West, Victoria 3083, Australia 2 Department of Primary Industries, PO Box 48, Frankston, Victoria 3199, Australia 3 CRC for Australian Weed Management Email: robin.adair@dpi.vic.gov.au infected plants (Punithalingam 1980). Wild popula- tions of R. fruticosus and some commercial black- berry cultivars with strong R. fruticosus pedigrees are susceptible to infection (Koellreuter 1950, Kövics 1980). In this paper we report on infection processes of S. ruborum, particularly the modes of penetration, the mechanisms responsible for cane dieback and the symptomatic host response to infection by this fungus. Understanding infection processes, particu- larly internal tissue sampling to determine invasion success, assisted in the design and evaluation of host- specificity tests. MATERIALS AND METHODS In studies using stem pieces, conidia of S. ruborum were harvested from cultures grown on PDA plates. Rubus anglocandicans A.Newton and R. ulmifolius Schott primocanes from laboratory-grown plants were cut into stem pieces (3–4 cm long and 2.0–2.5 mm diameter), surface-sterilised and dipped in a suspen- sion containing 10 7 conidia mL −1 of S. ruborum. Stem ends were removed and stem pieces were placed in a sterile Petri dish with moist filter paper and cultured in a phytotron at 20–25°C with a 12 h photoperiod for up to 35 days. Every three days, stems were harvested and 0.5 cm stem lengths fixed in a solution of 4% glu- taraldehyde. Uninfected controls were dipped in sterile distilled water and cultured under the same conditions. There were three replicates for each sampling time for each Rubus species. In preparation for sectioning, stem pieces were rinsed with 0.05M phosphate buffer (pH 7.0) and de- hydrated through increasing concentrations of ethanol before embedding in paraffin wax. Transverse and longitudinal sections were cut 4 μm thick with a Leica rotary microtome and stained with 1% acid fuchsin. In addition, samples of R. anglocandicans at 24 days and R. ulmifolius at 35 days after inoculation were stained with 0.5% toluidine blue. Stained sections were mounted in DPX. Infected stem sections were also prepared for scanning electron microscopy (Jeol JSM-35CF). Stems were rinsed with 0.05M phosphate buffer, dehydrated