The CSIRO breeding program at the Australian Cotton Research Institute at Narrabri aims to develop locally adapted varieties for use by Australian farmers. The program has been very successful in developing varieties for all cotton growing areas with ten conventional and five INGARD varieties (the latter subject to regulatory approval) available through Cotton Seed Distributors (CSD) for 1996 planting. The products of the breeding program have brought many millions of extra dollars to farmers and have been vital in maintaining the strength of the cotton industry. Key advances from the CSIRO breeding program include the development varieties with okra leaf (for insect and mite tolerance), better disease resistance (particularly bacterial blight and Verticillium wilt), improved fibre quality and good adaptation to cool growing areas. For 1996 planting CSD have available two very promising new CSIRO varieties -Sicot 189 and Siokra S-101.

Weed management systems in cotton are continually evolving, particularly in response to reduced cultivation and chipping and increased use of herbicides. The use of permanent beds and reduced tillage is likely to cause a shift in the problem weed spectrum, towards perennial weeds and those with rhizomatous root systems, which have previously been controlled with cultivation. A survey of 23 properties found that most weeds were much worse on fully cultivated fields than permanent bed fields. The exceptions were rhyncho, datura, nutgrass and polymeria, which were much worse on permanent bed fields. These data support the use of permanent beds as a way of reducing over-all weed pressure, but emphasise the need to .develop control strategies targeted at perennial and rhizomatous weeds.

Fungal diseases, in particular Verticillium wilt, and, more recently, Fusarium wilt, have the potential to cause major losses in Australian cotton production. Research conducted by our group is focussed on genetic approaches to the improvement of cotton resistance to fungal attack.

A new project at Narrabri aims to develop methods of breeding cotton varieties that per/ orm better under dryland and limited water situations.

The cotton monocultural system in most of the production areas in Australia strongly discriminates against natural enemies and favours the development of pest outbreaks. Pest outbreaks in monocultural systems occur because insect predators and parasitoids usually have more complex food requirements than Helicoverpa and most other phytophagous insects. The latter usually mate and oviposit without any feeding, relying only on reserves transferred from their larval food whereas predatory insects and parasitoids require different sources of food in larval and adult stages to develop and survive through the season. Thus pests like Helicoverpa can rapidly infest crops through migration and lay their eggs with little opposition from natural enemies. To solve this problem the cotton system should be diversified by interplanting cotton with other crops. The alternate crop could serve as a refuge to beneficial insects. Crops like safflower, sunflower, sorghum, corn, tomato and Lucerne have been studied for the past 4 years and Lucerne was found to be suited for the cotton production system. We report here studies to integrate lucerne into the Envirofeast IPM program to serve as a refuge for beneficial insects.

The Australian cotton industry is at the dawn of a new age in cotton production: an age where most, if not all, of the cultivars in use will carry the products of gene technology. The expected release of INGARD cotton this season is specially satisfying to us as the beginning to the early promises of genetic engineering for cotton improvement. While the task is nowhere near completed, the first commercial release of a transgenic cotton plant represents a concerted effort on the part of the industry's researchers, research collaborators, seed producers, consultants and growers to provide the industry with a new product that will significantly reduce chemical pesticide usage. INGARD cotton should be followed in rapid succession by a variety of new cultivars with enhanced production attributes that together will revolutionise the way you grow cotton.

The wild Australian Gossypium species arei subdivided into three sections. Sturtia. Hibiscoidea and Grandicalyx. The first two sections together contain five species. and include the widespread and locally abundant G. sturtianum and C. australe, respectively. Section Grandicalyx contains a further twelve species restricted to the Kimberley region or northwestern Australia and the Cobourg Peninsula in the Northern Territory. Until recently. the Grandicalyx taxa were poorly known. and phytochemical data have not previously been reported. Indeed. the phytochemistry or species belonging to sections Sturtia and Hibiscoidea is only now being investigated closely.

Australian cotton production relies heavily on insecticides for the control of the major pests, Helicoverpa spp. , mites and sucking insects (Fitt 1994). An over-reliance on insecticides results in problems of insecticide resistance, disruption of natural enemies and environmental contamination and has cast doubt on the long term viability of reliance on synthetic insecticides. The efforts of the cotton industry is to reduce the dependence on insecticide. This can be achieved by developing control programs that integrate minimal use of pesticides with other forms of control, especially predation by natural enemies of Helicoverpa spp. Despite widespread use of economic thresholds and the so called &quote;soft options&quote; in the current production systems, little emphasis has been placed on beneficial insects, while the reliance on chemicals negates the use of the term integrated pest management (IPM) for such a system. In cotton crops in Australia an average of 8 to 12 insecticide sprays are applied each season to control Helicoverpa spp. and other pests. A true IPM system should conserve natural enemies of the pests using appropriate techniques and utilize them as basic components in the management of these pests. An IPM program is usually a package consisting of different components of pest control which are integrated in stages during the development of the program. Since 1992, we have been developing an IPM program for cotton and have reached a stage where we have produced cotton yields similar to those obtained from crops which has been managed with conventional insecticides. We report here stage 5 of our IPM program where we integrated Envirofeast product and Iucerne crop refugia with Gemstar virus (NPV), a product developed by Biosys Inc.

Most pest management practitioners agree that our current approaches are not sustainable and integrated pest management (IPM) is the way of the future. IPM in cotton means different things to different people and there is some element of thought that our present lPM approach is more of an ideal than a reality. True integration, involving modifications to make diverse tactics compatible and reduce counter effects, is the essence of the approach (Pedigo 1989). Attitudes must change! The idea that &quote;the only good bug is a dead bug&quote; must give way to an appreciation of the ecological role of a pest species and acceptance of their presence. This change can only come about by education at all levels.

Resistance is a serious threat to transgenic cottons. Australian cotton growers are very familiar with the devastation that the industry can suffer from insecticide resistance, as in the Ord in the 1970s. Similarly, cotton growers are also well aware of the importance of transgenic cotton for reducing pesticide use in the industry. Unfortunately, transgenic plants are no less subject to selection for resistance than classical insecticides. A major pest of cabbage, the diamondback moth, has already evolved resistance to Bacillus thuringiensis (Bt) sprays in many cropping areas around the world. Bt-resistant diamondback moth larvae can be selected by and are completely resistant to transgenic plants that carry exactly the same Bt gene (&quote;CrylA&quote;) as is used in transgenic (INGARD) cotton (Metz, Roush, Tang, SheIton, and Earle, 1995 Molecular Breeding I: 309-317). There is every reason to suspect that a few cotton bollworms carry a resistance gene that is very similar to that in diamondback moth. Further, in contrast to insecticide sprays, the Bt toxin is continuously expressed in transgenic plants, which means that every insect feeding on them will be selected for resistance. This persistent exposure offers the potential of even stronger selection for resistance than would come from sprays. There is also evidence that genes for resistance to Bt may be more common than were genes for resistance to chemical insecticides, which could also give faster resistance. Thus, resistance management is at least as critical to transgenic crops as it has been for chemical insecticides.