Predicting the distribution of Endophyllum osteospermi (Uredinales, Pucciniaceae) in Australia based on its climatic requirements and distribution in South Africa

Wood A.R. ; Crous P.W. ; Lennox C.L. (2004)

Article

The perennial bush Chrysanthemoides monilifera ssp. monilifera (Asteraceae) is infected by the autoecious, microcyclic rust fungus Endophyllum osteospermi. Both organisms are native to South Africa, whilst the plant has also become naturalised in Australia where it is the target of a biological control program. E. osteospermi is under consideration as a biocontrol agent for this weed. Temperature and light requirements for aecidioid teliospore germination and basidiospore development were studied, as was the nuclear cycle during germination. Aecidioid teliospores germinated between 10 and 20°C, with 15°C as optimum temperature. Light, and particularly near-UV light, stimulated germination whereas germination was poor under dark conditions. A period of 6-8 h of light was the minimum needed to obtain germination levels equivalent to continuous light. The temperature requirements for basidiospore development differed from that for aecidioid teliospore germination. Optimal basidiospore production was at 15°C, but a rapid decrease occurred at higher temperatures, with few developing at 19°C, despite a high germination rate at this temperature. Two nuclear divisions occurred within 12 h of germination initiation to produce a metabasidium with three or four nuclei. A third nuclear division occurred in the basidiospores that then germinated between 24 and 48 h. Plants inoculated under controlled conditions took 5 to 24 months after inoculation for witches' broom symptoms to begin to develop. The detailed life cycle of E. osteospermi is presented. A Geographic Information System (GIS) approach was used to develop a model of the potential distribution of E. osteospermi in South Africa. This was based on monthly average climate surfaces with parameters derived from the above experiments. The parameters were modified so that the majority of all recorded localities of E. osteospermi in South Africa were included, whilst at the same time including only the minimum geographic area. The same model was applied to Australia to suggest a potential distribution of the rust fungus if released in Australia for the biological control of C. monilifera ssp. monilifera. This potential distribution was similar to one generated using the climate matching computer program CLIMEX, but gave greater spatial accuracy, at least in South Africa. Both approaches indicate that E. osteospermi should establish in temperate south-eastern Australia where C. monilifera ssp. monilifera is an invasive weed.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/10883
This item appears in the following collections: