Control of sudden death in cultivated proteas from the Southwest of Western Australia

Control of Sudden Death in Cultivated Proteas from
the Southwest of Western Australia

Christopher Dunne

Thesis presented to the School of Biology and
Biotechnology, Murdoch University, Western
Australia, for the fulfillment of the requirements of a
Ph.D.

March 2004

Acknowledgements

I wish to thank the whole WA protea industry, particularly those growers that offered their plantations for the disease surveys and that assisted in the establishment of the field trials. In particular, I would like to thank to Wally and Dawn Lewis at Anniebrook flower farm in Carbanup River. Without your help, assistance and input, this project would not have been possible.

The partnership with the Department of Agriculture has proved very rewarding and I would like to acknowledge the contribution of Mark Heap, Gilly Brown, Lachlan Duncan, Chris Newell, Gerry Parvliet and Digby Gowns.

A number of Phytophthora cultures used in the current study were obtained from a culture collection at Conservation and Land Management in Western Australia. I would like to recognize Jeff Boersma for providing the VHSC 8105 P. cactorum isolate.

I would like to acknowledge Malatesta Greenwaste Recyclers and Custom Composts for providing the mulch and compost used in the field trials. Also, thanks to Agseed Research and Wrightson seeds for providing the Brassica varieties used in the current study.

Thank you to all the staff at Murdoch University who assisted me in this project. I would like to pay particular thanks to Dr Giles Hardy who has been an inspiration and mentor during this period. Also, I would like to recognize Associate Professor Bernie Dell for his supervision. Thanks to the members of the “thesis writing group” (Tania Jackson, Aaron Maxwell and Sarah Collins) for helping with the finer details of the thesis.

Thanks to all my friends and family who have supported me during this time. I would like to recognise the support of my mother, Margaret Dunne.

Finally, and most importantly, thank you to my wife Janelle Martin for putting up with me during this time. I couldn't have done it without you and I will always be indebted to you.

This study funded by an ARC-SPIRT grant (C19940004).

Table of Contents

ABSTRACT

6

CHAPTER 1 – LITERATURE REVIEW 9

1. .0. Introduction 9
   1. The cultivated protea industry 9
   2. Growing proteas 10
   3. Pests and diseases of proteas 11
   4. Phytophthora cinnamomi 11
      1. Phytophthora root rot of proteas 15
      2. Symptoms 16
      3. The disease cycle in the southwest of Western Australia 16
      4. Control of Phytophthora 19
   5. General conclusions 28
   6. Project objectives 29

CHAPTER 2 – PROTEA DEATH AND DECLINE IN PLANTATIONS FROM

THE SOUTHWEST OF WESTERN AUSTRALIA 31

• 1. Introduction 32
  2. Methods 32
     1. Plantation surveys 32
     2. Isolation of fungi associated with diseased proteas 32
     3. Inorganic nutrient analysis 33
     4. Pathogenicity trials 34
  3. Results 36
     1. Plantation surveys 36
     2. Phytophthora cinnamomi pathogenicity trial 38
     3. Fusarium pathogenicity trial 42
  4. Discussion 42
     1. Plantation visits 42
     2. . Pathogenicity tests 43

CHAPTER 3 – DO BIOFUMIGANTS SUPPRESS THE VEGETATIVE

GROWTH OF FIVE PHYTOPHTHORA SPECIES IN VITRO? 45

• 1. Introduction 46
  2. Methods 47
     1. Experimental design 47
     2. Experimental isolates 47
     3. Preparation of biofumigant tissues 47
     4. Determination of suppression (growth tests) 48
     5. Quantification of biofumigant tissue 50
     6. Post treatment isolate viability 50
     7. Statistical analysis 51
  3. Results 51
     1. Suppression by the root and shoot tissues of Brassica juncea and B. napus 51
     2. Suppression by combining Brassica tissues 54
     3. Suppression using synthetic PE-ITC 56
     4. ITC content of the Brassica tissues 60
     5. Viability of isolates after exposure to volatiles 60
  4. Discussion 61

CHAPTER 4 – DO BIOFUMIGANTS AFFECT THE SPORULATION AND

SURVIVAL OF PHYTOPHTHORA CINNAMOMI? 64

• 1. Introduction 64
  2. Methods 65
     1. Experimental design 65
     2. Phytophthora cinnamomi isolate 65
     3. Monitoring of infective and survival structures 66
     4. Statistical analysis 67
  3. Results 69
     1. Quantification of the biofumigant tissues 69
     2. The effect of biofumigants on Phytophthora cinnamomi sporangia 69
     3. The effect of biofumigants on Phytophthora cinnamomi chlamydospores 72
     4. The effect of biofumigants on the infectivity of Phytophthora cinnamomi 75
  4. Discussion 76

CHAPTER 5 – DO BIOFUMIGANTS AFFECT INOCULUM POTENTIAL, INFECTIVITY AND DISEASE INCIDENCE IN PHYTOPHTHORA

CINNAMOMI? 79

• 1. Introduction 80
  2. Methods 80
     1. Experimental Design 80
     2. Biological materials 81
     3. Monitoring of inoculum potential, infectivity and disease incidence 82
     4. Statistical analysis 83
  3. Results 84
     1. Inoculum potential and infectivity in soil cores 84
     2. Inoculum potential and infectivity in soil leachate 87
     3. Disease incidence in Lupinus angustifolius 88
  4. Discussion 90

CHAPTER 6 – CAN SOIL SOLARISATION, FUMIGATION AND

BIOFUMIGATION REDUCE PHYTOPHTHORA CINNAMOMI INFECTION OF

LEUCADENDRON SAFARI SUNSET? 93

• 1. Introduction 94
  2. Methods 94
     1. Field trial design 94
     2. Biological materials 95
     3. Soil treatments 99
     4. Soil analysis 104
     5. Inoculation 109
     6. Monitoring of plant death 109
     7. Rainfall and temperature data 109
     8. Statistical analysis 109
  3. Results 110
     1. The soil treatments 110
     2. Post-treatment analysis 116
  4. Discussion 128

CHAPTER 7 – CAN BRASSICA JUNCEA OR B. NAPUS REDUCE

PHYTOPHTHORA CINNAMOMI INFECTION OF LEUCADENDRON SAFARI

SUNSET? 134

• 1. Introduction 134
  2. Methods 134
     1. Experimental design 134
     2. Biological materials 134
     3. Soil analysis 136
     4. Inoculation 140
     5. Monitoring of disease incidence 140
     6. Statistical analysis 140
  3. Results 141
     1. Treatment analysis 141
     2. Post-treatment analyses 144
  4. Discussion 154

CHAPTER 8 – GENERAL DISCUSSION 158

• 1. Plantation visits 158
  2. Effect of biofumigants on Phytophthora cinnamomi 159
  3. Field trials 160
  4. Biofumigation in protea cultivation 162
     1. Mechanisms of suppression 162
     2. Optimising biofumigation 162
     3. Incorporation methods 163
     4. Enhanced biodegradation 163
  5. Integrated management of Phytophthora cinnamomi 164
  6. Limitations of the current study 168
  7. Future research directions 169
  8. Conclusions 170

REFERENCES 171

Abstract

Phytophthora cinnamomi Rands is a common and devastating pathogen of cultivated proteas worldwide. Webb (1997) described a Sudden Death plant disease of proteas in Western Australia (WA) protea plantations. Proteas that suffer the syndrome display symptoms such as stunted growth, wilting, chlorosis and often death. In the current study, a number of protea plantations in the southwest of WA were visited to quantify the extent that P. cinnamomi was attributing to deaths of cultivated proteas. The survey indicated that P. cinnamomi is the major cause of Sudden Death in proteas. A range of other fungi (Fusarium, Botryosphaeria, Pestalotiopsis, Alternaria) and pests (nematodes, mealy bug, scale insects) were also identified to be contributing to protea death and decline in WA plantations. In many cases the factors contributing to protea disease appeared complex, with a range of physical factors or nutritional imbalances commonly associated with these pathogens and pests. As P. cinnamomi was the major cause of death of cultivated proteas the remainder of the experiments described in this dissertation investigated its control in horticultural plantings.

Biofumigation has the potential to become an important technique in an overall integrated management approach to P. cinnamomi. In this thesis, biofumigation refers to the suppression of pathogens and pests by the incorporation of Brassica plants into the soil. Two biofumigants (Brassica juncea (L.) Czern., B. napus L.) were screened for their effect on the in vitro growth of five common Phytophthora species (P. cinnamomi, P. cactorum (Lebert & Colin) Schroeter., P. citricola Sawada, P. cryptogea Pethyb. & Laff. and P. megasperma Drechsler). Growth was determined by the measuring dry weight and radial growth of vegetative hyphae. B. juncea was found to be superior in its suppressive effect compared to B. napus. There was also significant variation in the sensitivity of the Phytophthora species to the suppressive effects of the biofumigants. P. cinnamomi was the most sensitive of the five species investigated. Where the rates of the biofumigant were sufficient to suppress growth of Phytophthora, the suppressive effect was mostly fungicidal.

To determine how B. juncea and B. napus affect the infective ability and survival of P. cinnamomi, their effects on sporangia and chlamydospores production in soil was

investigated in vitro. P. cinnamomi colonised Miracloth discs were added to soil amended with the two Brassica species, before being removed every two days over an eight day period for the determination of sporangia production, chlamydospore production and infective ability. Only the soils amended with B. juncea significantly reduced sporangia production in P. cinnamomi. Both Brassica species increased the percentage of aborted or immature sporangia and reduced the infective ability of the pathogen. Neither Brassica species had any effect on zoospore release or chlamydospore production in P. cinnamomi.

Soil cores and soil leachate were collected from biofumigant-amended field soils to determine the inoculum potential and infective ability of the pathogen under glasshouse conditions. Amending the soil with both Brassica species had an immediate suppressive effect on the inoculum potential and infective ability of the P. cinnamomi. However, after this initial suppression there was a gradual increase in the recovery of the pathogen over the monitoring period of four weeks. To determine if the suppression would result in decreased disease incidence in a susceptible host, Lupinus angustifolius L. seeds were planted in the biofumigant amended soil. B. juncea amended soils reduced the disease incidence of P. cinnamomi by 25%. B. napus had no effect on disease incidence in L. angustifolius.

Although the current study had demonstrated that biofumigants could suppress the growth, sporulation and infection of P. cinnamomi, it was unclear if this would equate to a reduction in disease incidence when applied in the field. A field trial was conducted on a protea plantation in the southwest of Western Australia that compared biofumigation with B. juncea to chemical fumigation (metham sodium) and soil solarisation. The three soil treatments were used in an integrated management approach to control P. cinnamomi that included the use of a hardwood compost, mulch and water sterilisation. All treatments were monitored during their application to ensure the treatments were conducted successfully. The three soil treatments significantly reduced the recovery of the pathogen and the infective ability of the pathogen to a soil depth of 20 cm. Metham sodium was the most suppressive soil treatment and soil solarisation was the least suppressive treatment. Only the metham sodium treatment resulted in a

significant reduction in the incidence of root rot in Leucadendron salignum P.J. Bergius x laureolum (Lam.) Fourc (c.v. Safari Sunset) over the monitoring period of three years.

Another field trial was conducted on the same protea plantation to compare the effectiveness of B. juncea and B. napus, without the use of other control strategies, to reduce the incidence of P. cinnamomi infection of Leucadendron Safari Sunset. The concentration of isothiocyanates was monitored for seven days after the incorporation of the biofumigants. Although both Brassica species reduced the recovery and infective ability of the pathogen, neither biofumigant reduced the incidence of root rot in Leucadendron Safari Sunset.

In conclusion, P. cinnamomi is the most common and devastating pathogen in WA protea plantations. The current study demonstrated that P. cinnamomi is sensitive to the suppressive nature of biofumigants. Biofumigants can suppress the in vitro growth, sporulation, infective ability of P. cinnamomi and reduce the incidence of the disease caused by the pathogen in the glasshouse. Of the two Brassica species investigated, B. juncea was superior in its ability to control P. cinnamomi compared to B. napus. When applied in the field, biofumigation using B. juncea was found to be more suppressive that soil solarisation, but not as effective as metham sodium.

Chapter 1 – Literature review

1. .0. Introduction

Phytophthora cinnamomi Rands is a common, devastating pathogen of cultivated proteas in Western Australia (WA) (Boersma et al. 2000). It's imperative that the pathogen is managed if the WA industry is to continue to grow and prosper. The current study was a collaboration of the Western Australian protea growers, Department of Agriculture and Murdoch University. The study was initiated to investigate different horticultural practices that could be used to control the pathogen, thereby reducing the physical and economic impact of it in WA plantations. This chapter provides background into the WA protea industry, how P. cinnamomi causes disease within plantations and an overview of the different control techniques that have the potential to ameliorate the impact of the pathogen.

• 1. The cultivated protea industry

The word ‘protea' is commonly used to refer to cultivated members of the Proteaceae that include the genera: Protea, Leucadendron, Leucospermum and Serruria. Most cultivated proteas are indigenous to the Cape region of South Africa, although a limited number are native to Australia (e.g. Banksia, Telopea). Protea plantations occur in Australia, California, Chile, Hawaii, Israel, Korea, New Zealand, Portugal, South Africa, South America, Spain and Zimbabwe(Rabie 2000).

Proteas are grown for their colourful foliage or flowers: Leucadendrons are generally grown for their foliage, while Proteas and Leucospermums are grown for their flowers. Proteas make an ideal cut flower with an extremely long vase life and strong straight stems (Webb 1997). Proteas have been grown in managed horticultural plantations since the 1970's (Mathews and Mathews 1997). Prior to this time the only proteas sold were collected from the native plant communities. The establishment of plantations has allowed for the production of high quality blooms. This improvement in quality can largely be attributed to improved genetic stock from selections and breeding programs.

There are over 750 protea growers in Australia and the industry represents 93% of the total value of flower exports from Australia (RIRDC 1997). The industry doubled between 1989 and 1996 to over $AUS 40 million in product sold to export (RIRDC 1997). Western Australia is the largest exporting state with 53% in terms of value from 32% of the cultivated area of Proteaceae (RIRDC 1997). In WA, the industry consists of 143 farms, 597 Ha and is estimated to be worth $21 million p.a. ($16 million p.a. in product sold to export) (RIRDC 1997, Boersma et al. 2000, AgWA 2001). Protea plantations in the southwest of WA range in size from less than an acre to 43 Ha (AgWA 2001).

The majority of exported Australian proteas are sold to the Japanese market and to a lesser extent other Asian markets including Taiwan, Hong Kong, Singapore, Korea and Malaysia (Mathews and Mathews 1997). Furthermore, some product is exported to the European and American markets by a small number of growers, however, the increased cost of airfreight often prohibit these markets. Locally, a small number of flowers are sold through supermarkets, florists and small retail stores.

• 1. Growing proteas

Proteas are a very diverse group of plants that show a wide range in their growth habits. They grow best in mild climates and prefer a deep, well-drained soil (pH 5-6) for optimum growth (Webb 1997). Proteas are mostly grown from rooted cuttings for their superior form and only a few species are still grown from seed (e.g. Protea cynaroides). When selecting varieties it is important that the market requirements, growing conditions and management constraints are taken into account. Similarly, when choosing a new site, good access to the potential market from the area of planting is desirable to reduce associated traveling time and costs. For more information, a number of excellent guides exist for growing Proteas in Australia, including: McLennan (1993), Mathews and Mathews (1997), and Webb (1997).

Limited research has been conducted into the nutritional requirements of the Proteaceae. As a consequence the requirements of a limited number of cultivated species/varieties is known (Maier et al. 1995, Barlow and Haigh 1997, Silber et al. 1998). Proteas require

regular feeding with nitrogen and potassium but their phosphorus requirements are very low (Webb 1997). Nitrate application has been linked to increased number of stems and shoots length, thereby increasing productivity in leucadendron and protea species (Duncan pers comm.). Trace elements should be applied every couple of years or as required. Leaf analysis should be conducted at regular intervals on the major species grown to ensure adequate nutrient status and soil analysis should always be conducted before planting on a new site (Maier et al. 1995).

• 1. Pests and diseases of proteas

Over the past few decades the number of pathogens described from proteas has increased dramatically. The increasing knowledge of protea diseases aids early disease diagnosis and should improve the management of these diseases. Table 1.1 outlines some common pests and diseases encountered worldwide on cultivated proteas. The diseases differ in terms of their causes, symptoms, host species and epidemiology. While some pests and diseases may only reduce the value of the season's crop, others can stop whole export consignments or, worse still, cause widespread plant death in the plantation.

• 1. Phytophthora cinnamomi

Phytophthora cinnamomi is a soil-borne pathogen belonging to the Class Oomycetes or water moulds in the Kingdom Chromista (Bold et al. 1987). Oomycetes are not true fungi, rather they are a member of the water moulds with many characteristics similar to algae. Though morphology and biochemical composition may vary during changes in the life cycle, cell walls of Phytophthora species contain no chitin, but are composed mainly of non-cellulosic and cellulosic β-glucans with proteins, lipids, sugars and other polysaccharides (Bartnicki-Garcia and Wang 1983). The presence of cellulosic content in the cell walls of P. cinnamomi decreases the effectiveness of histological stains normally used for chitinous fungi when infected plant tissue sections (also cellulosic) are examined for the presence of the pathogen. For the purpose of the current dissertation P. cinnamomi is discussed as a fungus.

Table 1.1. The diseases and pests of cultivated and native proteas reported worldwide.

Phytophthora cinnamomi is a destructive soil borne pathogen that causes root rot and collar rot in susceptible host species. P. cinnamomi was first described by Rands (1922) in Sumatra as the cause of stripe canker in cinnamon trees. Since then the pathogen has been reported to occur in 67 countries including Europe, Africa, North America, Central America, South America, Asia and Australia (Zentmyer 1980, Finlay and McCraken 1991). In Australia, the pathogen has spread with devastating consequences into indigenous plant communities (Wills 1992) and economically significant horticultural crops (Hardy and Sivasithamparam 1988). The Commonwealth Endangered Species Protection Act of 1999 listed disease caused by Phytophthora as one of 14 ‘Key Threatening Processes' endangering Australian species and ecological communities (Hardy et al. 2001).

It is commonly believed that European settlers introduced P. cinnamomi to the southwest of WA in fruit trees and garden plants (Podger et al. 1965, Colquhoun and Hardy 2000). It was only as recently as the 1960's that the pathogen was actually isolated and identified (Podger et al. 1965). Today, the disease caused by the pathogen in Eucalyptus marginata Donn. Ex. Sm. (jarrah), Banksia and many other understorey species is recognised as a serious threat to the future of the native plant communities in the southwest of WA (Weste and Taylor 1971, Podger 1972, Shearer and Dillon 1995). Since its introduction, P. cinnamomi now affects over 14% of the northern jarrah forest with over 2000 of the 9000 plant species described in the south west of WA at risk (Wills 1992). The most susceptible plant families are the Proteaceae, Papilionaceae, Epacridaceae and Dilleniaceae (Wills 1992). Wills (1992) reported that 85% of the native Proteaceae from these ecosystems were susceptible to P. cinnamomi. Within these native plant communities the pathogen causes a decline in the abundance of susceptible species and subsequently a shift in the structure and composition (McComb et al. 1991, Wills 1992). This has resulted in resistant species, such as herbaceous perennials, becoming increasingly abundant. P. cinnamomi has also had a profound impact on a range of activities including timber harvesting, mining, tourism, conservation, fauna, and potable water supplies in and around infected areas (Wills 1992).

Phytophthora species are responsible for significant economic losses across many horticultural, ornamental and pasture crops in Australia. A nationally conducted survey in Australia (Cahill 1994) estimated that the direct loss due to Phytophthora disease was at least $223 million in all these industries combined for the year 1991-1992. In 1993, 84% of the diseases caused by the pathogen were considered to be capable of getting worse (increased losses) or unable to be controlled (Cahill 1994). Some of these areas that were classified unable to be controlled included the disease in native plant species and in cultivated proteas. The diseases caused by Phytophthora impose severe limitations to the future of these industries.

• • 1. Phytophthora root rot of proteas

Phytophthora cinnamomi is the most serious and costly plant pathogen of commercially cultivated proteas worldwide (McLennan 1993). P. cinnamomi is a pathogen of many economically significant Proteaceae genera, including: Banksia, Leucadendron, Leucospermum, Protea and Serruria (Von Broembsen 1984, Von Broembsen and Brits 1985, Von Broembsen and Kruger 1985, Shivas 1989, Irwin et al. 1995). The pathogen occurs within South Africa where many proteas are native (Von Broembsen 1984, Von Broembsen and Brits 1985, Von Broembsen and Kruger 1985). P. cinnamomi is also a pathogen of other wildflower species in WA, including Boronia and Chamelaucium (Shivas 1989, Boersma et al. 2000, Tesoriero et al. 2001).

In Australia, the pathogen has been reported to cause losses in cultivated proteas up to 50% and greater depending on the location (Cahill 1994). The pathogen is present in both nurseries and field plantings (Hardy and Sivasithamparam 1988, Horsman 2000). A survey conducted by Boersma et al. (2000) approximated that 50% of wildflower farms visited in the southwest of WA had one or more Phytophthora species present. The costs associated with lost production, use of fungicides and replanting due to the pathogen are high. In WA, the magnitude of the problem is such that a number of growers have either left the industry or moved to new land to escape the pathogen.

A number of different Phytophthora species have been reported to be pathogenic to cultivated proteas. Although isolated less frequently, P. nicotianae van Breda de Haan has been reported to cause losses up to 50% (Hardy and Sivasithamparam 1988,

Forsberg 1993, Cahill 1994). Also, P. citricola, P. cactorum, P. drechsleri Tucker and P. megasperma var. megasperma have been isolated from members of the Proteaceae, mostly from nurseries (Hardy and Sivasithamparam 1988, Shivas 1989, Boersma et al. 2000).

• • 1. Symptoms

Proteas infected by Phytophthora cinnamomi often die suddenly and as a consequence the disease has become commonly referred to in WA as the ‘Sudden Death' syndrome (Webb 1997, Duncan 1998). P. cinnamomi is primarily a root rotting pathogen that infects the fine roots of the host, but in some cases can girdle major tree roots or collars (Marks et al. 1981). Once the pathogen enters the root system of susceptible plants, primary symptoms of infection are evident as advancing fronts of necrosis (lesions) in the inner bark of roots and stems (Shearer 1994). Marks and Smith (1992) reported that these lesions developed in Leucadendron Silvan Red three days after inoculation. The above ground symptoms of infected proteas include stunted growth, wilting, chlorosis, decline and often death (Von Broembsen and Brits 1985). The pathogen ultimately kills the host by destroying the roots and girdling the base of the stem, thereby depriving the plant access to nutrients and water (Shearer 1994). As a consequence, symptoms of P. cinnamomi infection are often confused with symptoms of drought (Weste and Marks 1987, Wills 1992).

• • 1. The disease cycle in the southwest of Western Australia

The details of the life cycle and general biology of the pathogen have been well described (Zentmyer 1980, Weste and Marks 1987, Weste 1994) (Figure 1.1). As expected for a water mould, the growth, reproduction and spread of the pathogen are favored by excess water in the soil or ponding on the soil surface (Hardy et al. 2001). As a consequence, the movement of infested water and soil has played a major role in the spread of the pathogen. For example, human activity is recognised as the major mode of dissemination of the pathogen throughout the southwest of WA and this dispersal has been mostly facilitated by wet conditions (Shearer and Tippett 1989).

Mycelia, sporangia, zoospores and chlamydospores have all been identified to have a role in the disease process. Of these, zoospores are the major infection propagules of

P. cinnamomi in the southwest of WA (Zentmyer 1980, Shearer and Tippett 1989). They have been shown to have motility in laboratory conditions, have strong chemotactic response to roots at the zone of elongation (Tippett and O'Brien 1976), encyst to roots before the spore germinates (Hardham and Gubler 1990, Irwin et al. 1995) and infect primary tissue causing necrosis (Malajczuk et al. 1977).

Figure 1.1. Phytophthora cinnamomi infects its host's root systems using zoospores released predominantly during spring and autumn months in West Australian soils. Chlamydospores allow for the long-term survival during periods of adverse environmental conditions. Created using the descriptions of Zentmyer (1980), Ribeiro (1978), Shearer and Tippett (1989) and Erwin and Ribeiro (1996). Not to scale.

Several factors have been shown to influence sporangia and therefore zoospore production in P. cinnamomi, including: aeration (Duniway 1975), soil pH (Blaker and MacDonald 1983); soil moisture (Nesbitt et al. 1979, Gisi et al. 1980, Zentmyer 1980); temperature (Nesbitt et al. 1979, Zentmyer 1980, Weste 1983); and soil microflora (Zentmyer and Marshall 1959, Kennedy and Erwin 1961, Hardy and Sivasithamparam 1991, Erwin and Ribeiro 1996). Of these factors soil moisture and temperature are the two most critical requirements for sporangia production. Nesbitt et al. (1979) reported that P. cinnamomi sporangia were only formed in soils near field capacity and at temperatures over 15 °C (maximum sporangia were produced at 25 °C). They hypothesized that the poor sporulation in soils greater than field capacity was related to a decrease in the oxygen concentration of the soil.

Zoospores swim, or are passively transported, in flowing soil water to new host plants (Shearer 1994). As a consequence, P. cinnamomi disease fronts can rapidly extend down slope (Hill et al. 1994). The Mediterranean climate (cool wet winters, hot dry summers) of southern WA causes zoospore production to be highest when soil conditions are warm and moist (i.e. autumn & spring) (Shearer 1994). This results in a rapid increase in the inoculum density of the pathogen and consequently a high proportion of new infections occur during this time (Shea et al. 1980).

Irrigation affects the disease incidence of P. cinnamomi with the highest levels of activity of the pathogen recorded on irrigated areas during summer (Oudemans 1999). This is of importance as it is common practice to regularly irrigate proteas over their first few summers to aid establishment. The method of irrigation contributes to disease incidence with a study by Xie et al. (1999) recommending that drip irrigation be used every 1-3 days. Not only does the water provided by irrigation create favorable conditions for zoospore formation, but if the water source is infested irrigation also provides a means to spread the pathogen within the plantation.

Temperature also has a role in the progression of the disease within already infected plants. When Turnbull et al. (1995) investigated deaths in protea plantations from Queensland caused by P. cinnamomi, they reported that the dominant effect of soil

Control of Sudden Death in cultivated proteas from the southwest of Western Australia temperature was to alter the rate of hyphal extension within the root systems of plants already infected by the pathogen and only a small reduction in infection was recorded.

Chlamydospores also have infection potential (Malajczuk et al. 1977, Hwang and Ko 1978) and have been frequently recorded in infected plant tissue (Tippett and O'Brien 1976, Ho and Zentmyer 1977, Malajczuk et al. 1977). More importantly, chlamydospores allow for the long-term survival of this species during periods of adverse environmental conditions (low soil moisture, absence of susceptible tissue or high microbial activity). Research conducted in Australia demonstrates that chlamydospores of P. cinnamomi persist under conditions of low soil moisture during summer (Weste and Vithanage 1978, 1979). When favorable environmental conditions return, the chlamydospores can germinate allowing an increase in the density of the pathogen within the soil. Chlamydospores may also serve to reactivate current infections that were presently being contained within infected host tissue by the plant's defense mechanisms (Tippett and Hill 1984).

Phytophthora cinnamomi is predominantly heterothallic and requires the ‘pairing' of an A1 and A2 mating strain for sexual recombination (Galindo and Zentmyer 1964). Due to the A1 mating strain being rare in Australia (Pratt et al. 1972), it is unlikely that sexual reproduction is occurring to a significant degree within disease sites (Shearer 1994). However, it is unclear what role oospores have in the reproduction and persistence of P. cinnamomi.

• • 1. Control of Phytophthora

There are a number of characteristics of P. cinnamomi that make it a difficult pathogen to control (Munnecke 1972, 1984, Erwin and Ribeiro 1996). Ribeiro (1978) summarized these as follows:

■

■

■

its wide host range;

its ability to survive in symptomless hosts;

the ability of the pathogen to invade the soil profile to considerable depths (unlike many other root infecting fungi), thereby escaping bacterial antagonists or chemical fumigants (Kinal et al. 1993);

• the production of several types of inoculum that have varying degrees of resistance to environmental conditions (e.g. chlamydospores and zoospores); and
• the rapid spread of the pathogen in poorly drained or waterlogged soils.

A range of control measures is commonly used to control the pathogen. These are often used concurrently and include: hygienic nursery management, quarantine, crop rotation, resistant varieties, root stock development, improvement of drainage, chemical and biological control (Ribeiro 1978). To date there is no single control application that can sustain control in cultivated proteas (Irwin et al. 1995). Current control measures of P. cinnamomi in proteas are often inadequate when environmental conditions are favorable for the pathogen (Marks and Smith 1988).

Physical control

Phytophthora cinnamomi is often introduced into disease-free plantations by infested soil or infected plant material from nurseries (Hardy and Sivasithamparam 1988, McLennan 1993). It's also possible that new infections may arise from the movement of soil, plant material, soil water or stream water from surrounding properties and native plant communities. When viewing perspective land for a protea plantation it is important to determine if Phytophthora is present on the property, or in any surrounding areas. It is recommended growers implement a number of physical strategies aiming to prevent new introductions, reducing the population levels of the pathogen and inhibiting disease development (Shearer and Tippett 1989). Effective strategies include restricting soil movement within a property, avoiding the introduction of soil from external sources, manipulating soil drainage to reduce water logging and surface water events, and implementing sound hygienic practices such as raised benches within nursery areas and the cleaning of mobile farm equipment between paddocks.

As the pathogen is difficult to control once infection has occurred, prevention of new infections is the highest priority. It is essential that all new plant material brought into a grower's nurseries/plantations should be checked for symptoms of P. cinnamomi infection. Obvious symptoms include a staining or softness of the stem, collar and roots (Duncan 1998). There is increasing evidence to suggest that the use of some fungicides

in nurseries can suppress symptoms of disease, therefore all new stock must be treated as potentially infected. It is highly recommended that new plant material be purchased from nurseries with a Nursery Industry Association accreditation for site hygiene. This provides reassurance that the nursery uses practices such as sterilisation of potting media, general hygiene, water filtration/disinfection, and use of fungicides (Cahill 1994). Protea growers should adopt similar practices within their own nursery areas and screen any new material before planting it into their plantation.

Chemical control

A range of chemicals are used to control P. cinnamomi in horticultural settings, including the sterilisation of water, use of fumigants and fungicides. As described above the pathogen can multiply in open water supplies such as dams, streams and open tanks (McLennan 1993). Therefore, it is important to filter or sterilize any water that is to be used for irrigation. Often the sterilization of water can provide added advantages of providing algae-free pathways and drains, cleaner benches/equipments and reduction of blockages developing in pipes/ sprinklers (Lake 2000).

Chemical fumigants, such as methyl bromide and metham sodium have been widely used in nurseries and intensive horticulture as a pre-planting sterilisation technique (Munnecke 1972, Cohen and Coffey 1986, Porter et al. 1999, Duniway 2002). The use of methyl bromide and metham sodium can have the added benefit of increasing yield in some horticultural crops (Stephens et al. 1999, Utkhede and Smith 2000). However, these fumigants are expensive, extremely toxic and can have a significant environmental impact. The Montreal Protocol identified methyl bromide as a major ozone depletor and set a time frame for phasing its use out worldwide by 2005-2010 (Porter et al. 1999).

Metham sodium is also a highly toxic fumigant that requires appropriate caution and safely equipment. Metham sodium has been shown to suppress fungal diseases, pests and weed species (Kreutzer 1963, Gerstl et al. 1977, Duniway 2002). Utkhede and Smith (2000) reported that metham sodium reduces infection of apple roots by Phytophthora cactorum. The active compound in metham sodium is called methyl isothiocyanate (ITC) and is produced when the fumigant is applied to moist soil.

Members of the plant genus, Brassica contain glucosinolates, which upon hydrolysis yield a range of biologically active products, including ITCs. The liberation of these products from Brassica species is described later.

Systemic fungicides that offer limited protection against P. cinnamomi include cymoxanil, carbamate (prothiocarb) (Dixon et al. 1990), fluazinam (Swart and Denman 2000), phenyl amide (metalaxyl) (Cho 1981, Pegg et al. 1987, Swart and Denman 2000) and the alkyl phosphonates (phosphite) (Marks and Smith 1992, Swart and Denman 2000). Spraying programs should determine optimum rates of application, addition of surfactants, timing between applications, timing between other chemicals and what seasonal weather conditions favour disease. None of the fungicides described above can prevent disease development after infection if environmental conditions are favorable for the pathogen (Marks and Smith 1988, 1990, Swart and Denman 2000). This was demonstrated with the use of metalaxyl and phosphite in the Leucadendron Silvan Red (Marks and Smith 1992). The fungicides were most effective when applied prior to inoculation, although a significant reduction in lesion extension was achieved if applied at the time of inoculation.

Metalaxyl

Metalaxyl has been previously used as a soil drench to control P. cinnamomi in Proteaceae species (Hill et al. 1995, Swart and Denman 2000). The fungicide caused a significant reduction in the levels of inoculum of P. cinnamomi to a soil depth greater than 1.3 m (Hill et al. 1995). The possible benefits of the metalaxyl were supported by the findings of Cho (1981), who investigated its use to prevent Banksia death in Hawaiian plantations. However, the fungicide never became widely used by the industry due to fears of a developing resistance in P. cinnamomi as reported in other Phytophthora species (Pegg et al. 1987). Additionally, some protea species have been reported to be sensitive to metalaxyl (Von Broenbsen and Brits 1985).

Phosphite

Arguably the most effective fungicide for widespread use against P. cinnamomi is the systemic fungicide phosphite. Phosphite is transported in the phloem and the xylem of plants and thus can be applied as a foliar spray, soil drench or by trunk injection (Guest

and Grant 1991). The term phosphite is widely used to describe the salts and esters of phosphonic acid, such as Fosetyl-Al® (Aluminum tris-o-ethyl phosphite) (Whiley et al. 1995). Within plant tissues, alkyl-phosphonates are rapidly hydrolysed to phosphonic acid (H3PO3) and subsequently ionized to the phosphite anion, HPO3^-2 (Ouimette and Coffey 1989, 1990). Phosphite has been used to control a number of Phytophthora species in many horticultural crops including avocado (Pegg et al. 1990), almond (Wicks and Hall 1988), cocoa (Anderson and Guest 1990), pineapple (Pegg et al. 1990), ornamentals (deBoer and Greenhalgh 1990) and tomato (Flett et al. 1990). Phosphite is relatively inexpensive, non-toxic and hypothesized to be degraded in soil by other fungi and bacteria (Guest and Grant 1991). Further, phosphite has been shown to have no adverse effects on soil microbes that may be antagonistic to the pathogen (Wongwathanarat and Sivasithamparam 1991). However, it is possible that the continued use of phosphite may cause the pathogen to become resistant and thereby reduce the effectiveness of the fungicide. Furthermore, recent work conducted by Dobrowloski (pers. comm.) suggests that the continued use of phosphite may increase the pathogenicity of P. cinnamomi.

The phosphite anion has two modes of action within the host tissue that work in combination to arrest the pathogen. Firstly, phosphite acts by directly inhibiting the pathogen within plant tissue (Fenn and Coffey 1984, Cohen and Coffey 1986, Dolan and Coffey 1988). The inhibition resulting from the direct mechanism has been associated with biochemical changes (Grant et al. 1990) and physical compartmentalisation of the pathogen (Smith et al. 1997). Secondly, phosphite can act indirectly by enhancing the host plant defense mechanisms (Grant et al. 1990, Guest and Bompeix 1990, Guest and Grant 1991).

The effectiveness of phosphite is affected by a range of factors including: host species; disease resistance; climate; site topography; soil type; soil moisture content; soil temperature; and cultural practices (e.g. trickle irrigation) (Marks and Smith 1992). Current spraying practices of phosphite in Australian protea plantations recommend 4-6 weekly applications at 1% a.i. (active ingredient) during the summer months (Turnbull et al. 1995). In Queensland, the use of phosphite as a foliar spray (1.2% a.i. @ 6 weekly intervals) resulted in 100% plant survival in Leucadendron Silvan Red and

75% survival in Leucospermum Firewheel (Turnbull et al. 1995). However, this study reported that 88% of the treated plants that survived had P. cinnamomi present in their tissue, suggesting that phosphite may only prevent disease development after infection has occurred and not actually stop new infections. However, when environmental conditions were favorable to the pathogen even decreasing the spray interval (3 weeks) and doubling the application rate did not prevent death in the Leucospermum.

Although phosphite is considered to have low toxicity (Guest and Grant 1991), phytotoxic responses have been reported in a number of different plant species (Pilbeam et al. 2000, Tynan et al. 2001, Wilkinson et al. 2001). For example, at application rates above 1% a.i., many Banksia species have unacceptable phytotoxic responses (Shearer and Fairman 1997). However, application rates of phosphite below 1% a.i. have been shown to be effective in reducing colonisation of the pathogen in Banksia (Shearer and Fairman 1997, Pilbeam et al. 2000, Wilkinson et al. 2001, Barrett et al. 2003). Phytotoxic responses include foliar necrosis, defoliation, growth abnormalities, reduced pollen fertility and chlorosis (Fairbanks et al. 2002). A number of protea growers from WA have reported limited success with the use of phosphite to control P. cinnamomi. It is possible that this limited success may be partly due to incorrect usage of the fungicide. Errors to avoid include using an over diluted solution, forgetting to include a appropriate surfactant or allowing too long between applications (Hardy pers. comm.).

Biological control

Biological control can be defined as the reduction of inoculum density or disease­producing activities of a pathogen accomplished by or through one or more organisms other than man (Cook and Baker 1983). Biological control can be adapted for use in intensive nursery applications, and for management of larger areas. Cook and Baker (1983) described two approaches to biological control, enhancing host resistance or creating an unfavourable environment for the pathogen to infect host plants, or to survive. Common biological control methods include antagonistic microbes, development of resistant rootstocks, plant breeding, plant selection, hardwood composts, soil solarisation and biofumigation. Similar to chemical control, greater control is achieved if the technique is applied prior to exposure to the pathogen.

Suppressive soils and antagonistic bacteria

Broadbent and Baker first reported soils suppressive to P. cinnamomi in 1974. They associated the suppressive nature of the soil with soil microflora. A large number of soil microbes have since been shown to suppress a wide range of fungal pathogens, including P. cinnamomi, in vivo and in the nursery/glasshouse environment (Weste and Vithanage 1978, Halsall 1982, Keast et al. 1985, Turnbull et al. 1989, Hardy and Sivasithamparam 1991a, El-Tarabily et al. 1996). Some antagonist bacteria can even promote growth in Proteaceae plants (El-Tarabily et al. 1996). Commonly used antagonistic bacteria include Actinomycetes, Mircomonospora, Pseudomonas and Streptomyces species. Turnbull et al. (1992) demonstrated the use of antagonistic bacteria such as Pseudomonas cepacia Burkholderia could reduce the impact of P. cinnamomi in cultivated proteas. Antagonistic bacteria can be used alone or added to composts/mulches to further enhance their suppressive effect.

The mechanisms by which antagonistic microbes control P. cinnamomi can be accounted by a number of reasons. Firstly, antagonistic bacteria can attack the mycelia, sporangia and zoospores of P. cinnamomi thereby reducing the infective capability of the pathogen (e.g. antibiotic producing bacteria) (Broadbent and Baker 1974, Malajczuk et al. 1983). Secondly, the microbes can colonise the rhizosphere around the plant and thereby create a physical barrier to infection. There is some debate to the feasibility of these bacteria sustaining control when applied on a plantation scale (Turnbull et al. 1992).

Plant breeding, selection, and root-stocks

Protea species range in their susceptibility to P. cinnamomi from highly susceptible to resistant (Cho 1981, 1983, McCredie et al. 1985, Von Broembsen and Brits 1985). Selection and breeding for resistance has been used to control Phytophthora diseases in cultivated proteas (McCredie et al. 1985, Tynan et al. 1995) and other horticultural crops, including avocado and Chamelaucium (Irwin et al. 1995, O'Sullivan et al. 1999). Proteas that are resistant to P. cinnamomi are used preferentially for breeding new varieties or as a source for rootstock development. Plant selections for breeding and rootstock development can significantly increase product yield and quality. Swart and

Denman (2000) demonstrated that combining the use of resistant plants and fungicides completely suppressed symptom development of P. cinnamomi.

Resistance is hypothesised to be due to the plant being able to restrict growth of the pathogen within their roots and then produce new roots above the area where the pathogen is contained (Irwin et al. 1995). The ability of the plant to restrict the pathogen could be due to physical factors such as deposition of callose, the formation of necrophylactic periderms or lignin formation (Tippett and Hill 1984, Cahill et al. 1989).

Composting

Hardy and Sivasithamparam (1991a) demonstrated that composted hardwood bark suppress a range of different Phytophthora species, including P. cinnamomi. The addition of compost to soil suppressed sporulation by P. cinnamomi, resulting in reduced propagule density and survival of the pathogen. The incorporation of inorganic mulches is a common practice used to reduce root-rot in other horticultural crops (e.g. avocado) (Leonardi et al. 1999). The suppressive effect of composts can be attributed to a number of factors, including the chemical nature of leachate, temperature achieved during the composting process and microbial suppression (Hoitink and Fahy, 1986).

Soil solarisation

Soil solarisation refers to the use of energy of solar energy to heat the topsoil and is often described as polyethylene or plastic mulching (Katan 1980). The soil is solarised by tarping the topsoil with thin (~100 µm), transparent plastic and fastened on the side. The soil temperature increases during solarisation due to two dominant factors. Firstly, the plastic sheeting prevents heat loss by evaporation and convection and thereby heats the soil (Mahrer 1979). Secondly, water droplets form on the inner surface of the plastic sheeting and heat the soil via the greenhouse effect (Mahrer and Katan 1981). To ensure the highest possible soil temperatures are achieved, solarisation is usually conducted during the summer months for a minimum of 4-6 weeks. It is essential that the soil is kept moist (Coelho et al. 2001) and is cultivated prior to tarping to reduce air pockets in the soil that can reduce the efficiency of conduction (Katan et al. 1976, Grinstein et al. 1979).

Soil solarisation has been reported to be effective in controlling P. cinnamomi (Barbercheck and Von Broembsen 1986) and a range of other Phytophthora species (Kaewruang et al. 1989, Coelho et al. 1999, McGovern et al. 2000, Mizubuti et al. 2000, Coelho et al. 2001, Pinkerton et al. 2002). The nature of suppression achieved by soil solarisation is hypothesied to be due to a combination of physical and biological factors, including: the heating of the top-soil to temperatures (>50 °C); and short-lived increase in the soil microflora after the treatment had been applied. Nesbitt et al. (1979) reported soil temperatures >36 °C cause complete hyphal lysis in P. cinnamomi. Similar suppression has been previously described in a number of Phytophthora species (Coelho et al. 1999, McGovern et al. 2000, Mizubuti et al. 2000). Furthermore, Pinkerton et al. (2000) reported solarisation could reduce disease severity of Phytophthora in snapdragons. Increased suppression of the pathogen can be achieved by combining solarisation with other control techniques, including biofumigation (Coelho et al. 1999). Soil solarisation is cheap, non-toxic and an easy alternative to chemical fumigation.

Biofumigation

The incorporation of Brassica tissue into soil is known to have a suppressive effect on some soil-borne pathogens and pests. During the past decade, this agricultural process has become known as ‘biofumigation' (Angus et al. 1994). The incorporation of biofumigants into soil has been shown to significantly reduce a range of fungal pathogens, including Fusarium (Sarwar et al. 1998), Gaeumannomyces graminis var. tritici (Sacc.) Arx & DL Olivier (Angus et al. 1994), Rhizoctonia solani Kühn (Manici et al. 1997, Manici et al. 2000), Verticillium dahliae Klebahn (Harding and Wicks 2001) and members of the oomycetes, Aphanomyces (Chan and Close 1987) and Pythium (Manici et al. 1997, Manici et al. 2000). Interest in biofumigation has increased in recent years due to restrictions on the use of synthetic fumigants and the trend in many horticultural industries to reduce the use of dangerous chemicals. Currently, there is little information on how biofumigants affect Phytophthora species, particularly P. cinnamomi. However, studies conducted on Aphanomyces and Pythium have shown that biofumigants can suppress pathogen growth and can reduce disease incidence (Lewis and Papavizas 1971, Manici et al. 1997, Smolinska et al. 1997, Sarwar et al. 1998, Charron and Sams 1999).

The suppressive effect of Brassica species has been attributed to the plants containing significant amounts of glucosinolates (Kirkegaard and Sarwar 1999). Glucosinolates have little biological activity, however, products formed from their hydrolysis have been demonstrated to be highly biocidal (Brown and Morra 1997). Glucosinolates are hydrolysed during decomposition of Brassica plants by the plant-derived enzyme, myrosinase (Sarwar et al. 1998) (Figure 1.2). The hydrolysis reactions have been described by Larsen (1981) and Bones and Rossiter (1996). The specific hydrolysis products formed depend on the nature of the organic side chain on the glucosinolate. Brown and Morra (1997) distinguished over 100 different glucosinolates that can be distinguished on the differing nature of the organic side chain. Of the hydrolysis products, the volatile isothiocyanates (ITCs), are considered to be the most toxic (Brown and Morra 1997). However, it has been hypothesized that some water-soluble glucosinolate products and mildly toxic non-glucosinolate derived volatile compounds also produced during decomposition of Brassica's, may contribute to the biofumigant effect (Bending and Lincoln 1999). Suppression of some fungal pathogens by the ITCs liberated from Brassica plants has been shown to be superior to the synthetic fumigant metham sodium (Sarwar et al. 1998).

Isothiocyanates Thiocyanates Nitriles

Glucosinolates

Enzyme: myrosinase

Figure 1.2. The hydrolysis of glucosinolates in Brassica tissues by myrosinase can produce a range of highly toxic chemicals, including isothiocyanates. After Brown and Morra (1997) and Gardiner et al. (1999).

• 1. General conclusions

Factors that can dramatically impact on production, such as the diseases caused by P. cinnamomi, provide a hindrance to the future success of the commercial protea industry worldwide. P. cinnamomi reduces the amount of saleable product, while increasing management costs. In WA, the effect of P. cinnamomi is so severe and widespread that it threatens the future of the entire industry. Therefore, it is critical to

the industry that viable and cost effective control measures are developed to reduce the incidence and spread of the pathogen.

As discussed earlier (Section 1.4.4.), there is no single control measure that can control the pathogen in diseased plantations. The best option for growers is to apply a number of different control techniques (physical, chemical and biological) using an integrated management approach. This requires growers to be aware and have strategies in place, even if the disease is not present in their plantation. The findings of this dissertation will give growers specific information on a range of pre-planting control techniques that can be used to replant on dieback areas.

• 1. Project objectives

Sudden Death was first used to describe unexplained deaths in proteas plantations in WA during the middle of the 1990's (Heap pers comm.). Phytophthora cinnamomi has been shown to be associated with Sudden Death in proteas (Boersma et al. 2000). This dissertation aimed to compare different practical techniques that have been successfully used in other horticultural crops to control fungal diseases such as P. cinnamomi. The techniques investigated include: biofumigation (Indian mustard, fodder rape), use of compost and mulch, soil solarisation and chemical fumigation (metham sodium). These techniques utilise elements of chemical and biological control. One treatment, biofumigation, was selected for investigation into the mechanisms by which Brassica plants suppress Phytophthora species, particularly P. cinnamomi.

Prior to commencing the investigative experiments, a number of protea plantations were visited in the southwest of WA to quantify the extent that P. cinnamomi was contributing to plant deaths (Chapter 2). These visits also aimed to determine the role of other plant pathogens, pests and nutritional imbalances that may be contributing to plant decline and death.

To describe the mechanisms by which biofumigants control P. cinnamomi, a series of experiments were conducted at Murdoch University. Firstly, the effect of volatile chemicals released from decomposing biofumigant tissue on the in vitro growth of Phytophthora species was investigated (Chapter 3). Secondly, the effect of amending

soil with biofumigants on the sporulation and survival of P. cinnamomi was determined (Chapter 4). The proceeding experiment then aimed to determine whether the suppressive effect observed in the previous two chapters would equate to a reduction in disease incidence in a susceptible host (Chapter 5).

The final two experimental chapters describe two field trials conducted at Anniebrook flower farm at Carbanup River in WA. The first trial compared the ability of three different soil treatments, Indian mustard biofumigant, metham sodium and solarisation in combination with hardwood mulch to reduce P. cinnamomi infection (Chapter 6). The second trial compared the control of the pathogen by two different types of biofumigants alone, fodder rapes and Indian mustard (Chapter 7). Measurements of disease incidence and other contributing factors were used to compare the treatments.

Chapter 2 – Protea death and decline in plantations
from the southwest of Western Australia

Published:

Dunne, C.P., Dell, B. and Hardy, G.E.St J. 2003. Sudden death in proteas in the south-west of Western Australia. Acta Horticulture. 602: 39-44.

Conference proceedings:

Dunne, C.P., Dell, B. and Hardy, G.E.St J. Sudden death in proteas in the south-west of Western Australia. Proceedings of the 6^th International Protea Research Symposium, Maui, Hawaii, USA, March 2002.

Note:

Dunne conducted the research

Hardy and Dell were supervisors

• 1. Introduction

Phytophthora cinnamomi is a common and devastating pathogen of native and cultivated Proteaceae worldwide (Cho 1981, Von Broembsen 1984, Von Broembsen and Brits 1985, Von Broembsen and Kruger 1985, Shearer and Dillon 1995). The pathogen is particularly widespread throughout native plant communities, horticultural industries and nurseries in the southwest of Western Australia (WA) (Hardy and Sivasithamparam 1988, Cahill 1994, Webb 1997, Boersma et al. 2000). Although P. cinnamomi is known to be an important pathogen in the WA protea industry, it is unclear if other pathogens may account for significant losses within plantations. Worldwide, over the past decade there have been a number of new diseases recorded from diseased proteas, however, little work has been conducted to investigate if pathogens other than P. cinnamomi are causing significant losses to the industry in WA.

In the current study, a number of protea plantations in the southwest of WA were visited in order to quantify the extent that P. cinnamomi was attributing to plant death. The study aimed to determine the role of other plant pathogens, pests and nutritional imbalances that may be contributing to plant decline and death.

• 1. Methods
     1. Plantation surveys

A total of 28 plantations in the southwest of WA were visited between 1999 and 2001 to investigate unexplained death and decline in protea plantations. Plant material and soil were collected for the purpose of isolating fungal pathogens. Additionally, in some cases, soil and leaves were sampled for inorganic nutrient analyses. Site factors at each plantation (e.g. insufficient drainage) that may have influenced plant health, were also recorded. At plantations that displayed signs or symptoms of plant disease, three to five plants were sampled for each protea species.

• • 1. Isolation of fungi associated with diseased proteas

During the initial plantation visits it became evident that the majority of diseases appeared to be caused by root and stem pathogens. Leaf diseases were less prevalent and often were not causing significant economic losses to the grower. Therefore, the sampling technique used was biased towards root and stem pathogens. However, where

foliar diseases (leaf spots or blights) were observed to be causing significant losses, leaf samples were also collected. Where basal stem lesions were observed, particular effort was made to sample at the margins of the necrotic tissue. The collected plant material was surface sterilised by immersing sections in a 1% sodium hypochlorite solution for 30 sec before rinsing once with sterile deionised water. Approximately 30% of the collected plant tissue was plated onto the agar plates without being immersed in sodium hypochlorite in case the sterilisation process destroyed any pathogens present.

The sampled plant tissue was plated onto three types of agar plates, potato dextrose agar (PDA), water agar and a Phytophthora selective medium (NARPH) (Hüberli et al. 2000). The plates were incubated in the dark (24 ± 1 °C) for one to five days before being assessed for the presence of Phytophthora and other possible fungal pathogens. Phytophthora cinnamomi isolates were identified using the descriptions of Waterhouse (1963). Cultures of P. cinnamomi and other cultures of interest were sub-cultured to ensure clean colonies before being placed on PDA slopes and stored in the dark at 15 ± 1 °C. No evidence of Armillaria (mycelial sheathing, mycelia assocated with diseased roots, basidiome production) infection was observed, therefore no material was plated onto MEA.

The sampling of diseased proteas generated a large number of fungal isolates other than P. cinnamomi. To prioritize the scope of this study the isolates were sorted to determine their relative importance. The more common fungi were identified using taxonomic keys (Ellis 1971, Ellis 1976, Dennis 1978, Carmichael et al. 1980, Sutton 1980, Von Arx 1981, Domsch et al. 1993, Ellis and Ellis 1997, Barnett and Hunter 1998, Hanlin 1998, Waller et al. 2002).

• • 1. Inorganic nutrient analysis

Where appropriate, inorganic analysis (N, P Ca, Mg, Na, K, S, Zn, Cu, Fe, Mn, and B) was conducted on leaf material sampled after Maier et al. (1995). Soil samples were analysed using standard methods for Na2CO3 extractable P, NH4Cl extractable K, and DTPA extractable Zn, Cu and Mn (Rayment and Higginson 1992). SoilWorx Pty Ltd, Bibra Lake, 6160, WA, conducted all the analyses described. Foliar values published by Price et al. (1997) were used for reference.

• • 1. Pathogenicity trials

Phytophthora cinnamomi pathogenicity trial

Plant material

Four P. cinnamomi isolates were used to inoculate four different protea varieties: Leucospermum cordifolium (Salisb. Ex Knight) Fourc x tottum (L.) R.Br., Protea neriifolia R.Br. x susannae E. Philips, Leucadendron salignum x laureolum and L. salignum. The propagated plants were purchased as 75 mm tube stock from the accredited Muchea Tree Farm (Muchea, WA). The plants were maintained within an evaporative cooled glasshouse with a temperature range from 16-30 °C. The plants were watered using overhead sprinklers (2 Lhr^-1) for 5 min twice daily. After four weeks the plants were assessed for symptoms of P. cinnamomi infection and were observed to be disease free.

Phytophthora cinnamomi isolates and inoculum production

Four P. cinnamomi cultures that were isolated during the plantation visits were chosen for pathogenicity testing (Table 2.1). Prior to use, the isolates were passa¹ged¹ through a Leucadendron salignum x laureolum to ensure they had maintained their pathogenicity. Ten replicates of each isolate and plant species combination were conducted. The inoculum used in this trial consisted of sterilised (3 x 121 °C/20 min, 24 hr intervals) Miracloth (Calbiochem, USA) 5 mm discs colonised by 14-day-old cultures.

The main stem, approximately 5-10 mm thick, was inoculated using the under bark inoculation method described by O'Gara et al. (1997) (Figure 2.1). Parafilm was wrapped around the point of inoculation, followed by silver ducting tape, to protect the inoculated stem and mycelia from drying out and direct sunlight. Four replicates of each plant species were sham (fake) inoculated with sterile Miracloth discs as a control. After 21 days, the inoculated stems were harvested, stem diameter recorded and the lesion length was measured. Additionally, 20 one cm sections from above the point of inoculation were plated onto NARPH agar plates to access how far the isolates had colonised the host, beyond the lesion.

Fusarium pathogenicity trial

At one plantation a Fusarium species was consistently isolated from diseased

Leucadendrons. Dying plants had pink basal lesions that were obvious when the bark

Figure 2.1. The under bark inoculation technique used during the current study (O'Gara et al. 1997). (A) Make a shallow incision in stem; (B) add Phytophthora cinnamomi colonised Miracloth disc mycelial side towards the stem (C & D) wrap point of inoculation with parafilm and silver ducting tape to prevent evaporation. Scale bars represents one cm.

Table 2.1. The host plants from which the four Phytophthora cinnamomi isolates used in the pathogenicity tests originated.

1. Passaged refers to the inoculation and subsequent recovery of the pathogen through a susceptible host.

was removed. Plants with lesions often showed signs of chlorosis and wilting. A pathogenicity trial was conducted at the plantation in order to confirm that the Fusarium species was the cause of plant death. Three 3-year-old Leucadendron species (L. discolor E. Phillips & Hutch, L. galpinii E. Phillips & Hutch, L. laureolum), were inoculated using colonised Miracloth (5 mm) discs from one week old cultures previously isolated from the stems of dying plants at the plantation. The plants were inoculated on the lateral branches, approximately 10-20 mm thick, using the method described above (Section 2.2.4). Control plants were sham inoculated with non­colonised Miracloth. There were five clonal replicate plants per treatment per plant species. After 14 days, the inoculated stems were harvested and the visible lesions measured. Additionally, 10 one cm sections above the point of inoculation were surface sterilised as described above and then plated onto PDA plates to determine how far the Fusarium isolate had colonised beyond the lesion.

• • 1. Statistical analysis

The data collected during the pathogenicity trials were analysed using the ANOVA package of Statistica '99 Edition (Statsoft Inc, USA). Prior to analysis the data were assessed for homogeneity, variation of the mean from the variance and to determine if it was normally distributed. The means were compared by LSD (P = 0.05) and presented with the standard error of the mean.

• 1. Results
     1. Plantation surveys

Phytophthora cinnamomi

Phytophthora cinnamomi was the most frequently isolated pathogen and was recovered from recently dead or dying plants from 39% (11/28) of the plantations visited. P. cinnamomi infection was recognizable in the field as obvious dark basal lesions when the bark was removed (Figure 2.2). In some cases, surface lesions were also visible. P. cinnamomi was isolated from 12 different species/hybrids of cultivated proteas (Table 2.2). The average isolation rate using PDA and NARPH media was 47%. Infection by the pathogen had decimated some locations within plantations with losses of greater than 50% being common. P. cinnamomi was often isolated from dams and contaminated irrigation water was identified as a major cause of the pathogens spread

around the plantation. No other Phytophthora species were isolated from plant material collected during this study.

Figure 2.2. (A) Basal lesion caused by Phytophthora cinnamomi in Leucadendron salignum, (B) a high impact P. cinnamomi disease centre in Leucadendron floribunda, (C) P. cinnamomi infection often results in a complete loss of the fine root system of the host.

Other fungal pathogens

A range of other fungal pathogens, pests, nutritional imbalancestatus and physical factors (e.g. poor drainage, or inadequate watering) were observed to affect protea health (Table 2.2). Apart from P. cinnamomi, a number of other fungi were isolated from stem cankers, but only a Fusarium species and possibly a Botryosphaeria species were considered to be possible pathogens, the other fungi isolated are commonly recorded saprophytes. Overall, these fungi were not as widespread or as devastating as P. cinnamomi. The Fusarium isolate obtained from one plantation appeared particularly aggressive and was chosen for pathogenicity testing.

Eleven percent (3/28) of the plantations visited had significant leaf diseases. The king protea (Protea cynaroides L.) appeared to be particularly prone to leaf spots and blotches. An Alternaria species was frequently isolated from the margins of raised, red

blotches on the leaves. Unfortunately, due to time constraints, no pathogenicity trials were conducted. Often the presence of these leaf spots/blotches was associated with nutrient imbalances, for example, sodium toxicity or Cu deficiency (Table 2.2).

Pests

A number of pests were found to be associated with dead and dying proteas. Common pests included: root-knot nematode (a Meloidogyne sp.), longihorn borers (Coleoptera: Cerambycidae), mealy bug (Homoptera: Pseudococcidae), termites (Isoptera: Hexapoda) and scale (Coccoidea) insects. However, these pests were not widespread and were often associated with plants that were dying from other causes, for example, P. cinnamomi infection or nutritional deficiencies.

Abiotic stresses

Abiotic factors also contributed to protea decline at some plantations. These factors included: water logging, inadequate watering, physical stem damage, herbicide damage, shallow topsoil, root-bound plants and nutritional imbalances. Zinc deficiency was the most common nutritional imbalance (21% of plantations visited affected). A number of other trace elements, for example: copper, sodium and iron were found to be deficient on 14%, 11% and 7% of plantations visited, respectively. The presence of the Fusarium species was often found in association with sodium toxicity and trace element deficiencies. At one plantation the hemi-parasitic plant, Nuytsia floribunda (Labill.) R. Br., was stunting the growth of a number of Leucadendron species.

• • 1. Phytophthora cinnamomi pathogenicity trial

Lesions developed in all Protea neriifolia x susannae and Leucospermum cordifolium x tottum plants which were also more susceptible (P < 0.01) to colonisation by the P. cinnamomi isolates than the Leucadendron salignum x laureolum and Leucadendron salignum plants (Figure 2.3A). Protea neriifolia x susannae and Leucospermum cordifolium x tottum also had larger lesions (P < 0.01) than Leucadendron salignum x laureolum and Leucadendron salignum (Figure 2.3A). The four P. cinnamomi isolates showed significant (P < 0.01) intra-specific variation in the extent of colonisation and length of the lesion (Figure 2.3B).

Table 2.2. A summary of the plantations visited outlining isolation of Phytophthora cinnamomi isolates associated with dying proteas. Other identified fungi, pests and nutritional imbalances are given.

Isolates CD24, CD1 were less pathogenic, in terms of colonisation and lesion length, than isolates CD9 and CD10. There was no correlation (r = 0.15, r = 0.23) between stem diameter and the colonisation (r = 0.23) or lesion length (r = 0.15).

laureolum

Protea variety

Figure 2.3. (A) The mean colonisation (□) and lesion length (■) of four protea varieties after infection by Phytophthora cinnamomi. (B) The mean colonisation (□) and lesion length (■) after infection by the four P. cinnamomi isolates in the host stems. Letters denote significantly (α = 0.05) different colonisation (a, b, c) or lesion (x, y) lengths. Standard error bars are shown.

• • 1. Fusarium pathogenicity trial

All of the stems inoculated with Fusarium developed lesions. There was no significant (P = 0.71) difference between the three Leucadendron species in lesion length or extension of the fungus through the stem (P = 1.0) (Figure 2.4). In most of the stems, the fungus was recovered from all the sections plated, demonstrating that the fungus was aggressive. No lesions were observed in the control plants that were inoculated with non-colonised Miracloth discs.

Leucadendron species

Figure 2.4. The mean colonisation (□) and lesion length (■) for three Leucadendron species 14 days after inoculation with the Fusarium species. Standard error bars are shown. The control wounds averaged less than one cm in length.

• 1. Discussion
     1. Plantation visits

Phytophthora cinnamomi was the most frequently isolated pathogen and was recovered from recently dead or dying plants from 39% (11/28) of the plantations visited. These results are comparable to those of Boersma et al. (2000) who recovered P. cinnamomi from 34% (12/35) protea plantations visited in 1999/2000. No other Phytophthora species was isolated from plant material collected during this study. This contrasts with the findings of Boersma et al. (2000) who reported the presence of four other Phytophthora species in WA plantations. However, these

other species were isolated from soil and it is not clear what role they had, if any, in protea decline or Sudden Death.

The plantation visits uncovered a number of other potential pathogens that could be a significant threat to the WA protea industry. For example, this is the first record of a Fusarium species causing significant disease in an in Australian plantation. Other fungi (Botryosphaeria and Pestalotiopsis) isolated also have the potential to become widespread and reduce the productivity of the industry. They have been reported to be significant pathogens of cultivated proteas in other countries (Knox-Davies 1981, Knox-Davies et al. 1987, Swart et al. 1999, Crous et al. 2000, Denman et al. 2003). This is the first record of these fungi associated with disease in WA protea plantations. It is unclear if these are primary pathogens or are being driven by other problems, such as nutrient imbalances. For example, many of the leaf diseases can be controlled by correction of the nutritional status of the host and the application of commonly used fungicides. Unfortunately, in scope of the current dissertation it was not possible to investigate all these other pathogens further. It is highly recommended that future research be conducted into the threat posed by these and other possible pathogens to the WA protea industry and native plant communities.

• • 1. . Pathogenicity tests

This study demonstrated that there was variation in the colonisation (extension) by the pathogen and the length of the lesions in the four protea varieties tested. Protea neriifolia x susannae and Leucospermum cordifolium x tottum were more resistant than Leucadendron salignum and Leucadendron salignum x laureolum. Von Broembsen and Brits (1985) conducted pathogenicity tests on a number of protea species by inoculating soil with P. cinnamomi sporangia. They demonstrated members of the Protea genus were less susceptible than members of the Leucadendron genus. In the current study, the Leucospermum cordifolium x tottum was the most resistant to infection by the pathogen. Von Broembsen and Brits (1985) reported that most of the Leucospermum species they tested were very susceptible to P. cinnamomi. However, they also found that the Leucospermum hybrid, L. formosum (Andrews) Sweet x tottum, was resistant to the pathogen. This

variation in the susceptibility among different varieties suggests that some measure of control could be achieved by selecting resistant or tolerant species. Swart and Denman (2000) were able to completely control the pathogen when the metalaxyl was combined with the use of the resistant L. formosum x tottum hybrid.

The increased susceptibility of Leucadendron species was also observed during the isolation of the pathogen during the plantation visits. Over 70% of the plants infected over the 28 plantations with P. cinnamomi were Leucadendron species. Of these, 47% were Leucadendron salignum x laureolum (c.v. Safari Sunset). This in part could be due to the large percentage of proteas being grown in WA that are Leucadendron varieties, particularly Leucadendron salignum x laureolum (AgWA 2001). However, it would be wise for growers with P. cinnamomi infected plantations to consider growing a greater percentage of resistant species, such as Protea hybrids.

The current study indicates that P. cinnamomi is the major agent involved in the Sudden Death of proteas. However, a number of other fungi and pests were also contributing to protea death and decline in WA plantations. Further research is necessary to determine the extent that these pathogens and pests are contributing to protea decline in WA. Additionally, the effect of nutrition and other environmental factors on protea health warrants further investigation. Only then can appropriate control measures be formulated. The remainder of this dissertation will investigate potential control strategies that could reduce P. cinnamomi infection in protea plantations, particularly biofumigation.

Chapter 3 – Do biofumigants suppress the
vegetative growth of five Phytophthora species
in vitro?

Published:

Dunne, C.P., Dell, B. and Hardy, G.E.St J. 2003. The effect of biofumigants on the vegetative growth of five Phytophthora species in vitro. Acta Horticulture. 602: 45-52.

Conference proceedings:

Dunne, C.P., Dell, B. and Hardy, G.E.St J. The effect of biofumigants on the vegetative growth of five Phytophthora species in vitro. Proceedings of the 6^th International Protea Research Symposium, Maui, Hawaii, USA, March 2002.

Note:

Dunne conducted the research

Hardy and Dell were supervisors

• 1. Introduction

Chemicals produced during the decomposition of Brassica plants can suppress a range of fungal pathogens (Chan and Close 1987, Angus et al. 1994, Harding and Sarwar et al. 1998, Wicks 2001). The most important of these chemicals are the volatile isothiocyanates (ITCs), produced from the hydrolysis of glucosinolates (Brown and Morra 1997). Although there is little information on Phytophthora species, studies conducted on closely related Oomycetes (Aphanomyces and Pythium) have shown that biofumigants can suppress pathogen growth and reduce disease incidence (Lewis and Papavizas 1971, Manici et al. 1997, Smolinska et al. 1997, Sarwar et al. 1998, Charron and Sams 1999, Manici et al. 2000).

Different Brassica species/varieties produce different concentrations and types of ITCs during their decomposition (Kirkegaard and Sarwar 1999, Harding and Wicks 2001). The different ITCs produced during decomposition vary significantly in their toxicity to fungi (Brown and Morra 1997). Also, different fungal taxa vary in their resistance to ITCs. For example, it has been reported that Pythium irregulare Buisman. is up to 16 times more resistant to ITCs than Fusarium and Rhizoctonia (Sarwar et al. 1998). Additionally, previous authors have reported different levels of suppression in P. irregulare and P. ultimum Trow, indicating that significant intra­specific variation occurs in the Oomycetes (Manici et al. 1997, Sarwar et al. 1998, Charron and Sams 1999).

Biofumigation has the potential to become an important technique in an overall integrated management approach to controlling P. cinnamomi in horticultural crops. There is little information in the literature on how biofumigants affect Phytophthora species, particularly P. cinnamomi. The current study investigated whether biofumigants have an effect on the in vitro vegetative growth of Phytophthora species. The study aimed to measure the level of interspecific variability in Phytophthora (P. cactorum, P. cinnamomi, P. citricola, P. cryptogea, P. megasperma) and intraspecific variability in P. cinnamomi to ITCs. The sources of ITCs included a purified solution of phenylethyl ITC (PE-ITC), and the biofumigants Brassica juncea and B. napus.

• 1. Methods
     1. Experimental design

Three experiments were undertaken to investigate the effect of biofumigants on Phytophthora species. (i) The growth of the fungal isolates was measured in the presence of root or shoot tissue of the two Brassica species at two rates (0.5 g/L, 1 g/L) of biofumigant per L headspace. These two rates were chosen after a preliminary experiment (data not shown) demonstrated that 1 g/L of B. juncea completely suppressed growth of a P. cinnamomi isolate (CD10). (ii) The effect of combining the different tissues from the two Brassica species was investigated. This included the three treatments: 0.5 g/L of B. juncea root and shoot; 0.5 g/L of B. napus root and shoot; and 0.25 g/L of the root and shoot tissues from both B. juncea and B. napus. (iii) The suppressive effect of 97% PE-ITC (Sigma Chemicals, Australia) solution at two rates (0.1, 1 uL L) was investigated. All experiments included a treatment that used water in place of the biofumigant source as a control and the treatments were replicated three times. All growth tests were conducted using a randomised block design (Steel and Torrie 1986) at 24 °C in the light in a constant temperature room. All experiments were repeated twice to confirm the validity of the results.

• • 1. Experimental isolates

The ten Phytophthora isolates used included six P. cinnamomi isolates and one isolate each of P. cactorum, P. citricola, P. cryptogea and P. megasperma (Table 3.1). The cultures used were either isolated from diseased proteas in Western Australia (WA) (Chapter 2) or were from collections held at Murdoch University and the Department of Conservation and Land Management in WA. The Phytophthora isolates were maintained on Potato Dextrose Agar (PDA) plates at 24 ± 1 °C in the dark.

• • 1. Preparation of biofumigant tissues

The Brassica species used in this experiment were grown in a field plot at Murdoch University. Seeds of the two Brassica species were planted (6 g seed/10 m²) into the

Table 3.1. The isolate codes, collection and hosts of the Phytophthora isolates used for determining the affects of biofumigants on the in vitro growth of Phytophthora species.

I = Murdoch University Culture Collection

II = Conservation and Land Management Culture Collection

field plot at a depth of 2 cm using a manual seed spreader in August 2001. Four weeks after planting, NPK Blue special (Cresco fertilisers, WA) was added using a spreader at a rate of 10 g/10 m². The field plots were watered for 10 min, twice a day, using overhead irrigation. During late September, the emerging seedlings were top dressed with NPK Blue Special (Cresco) at 1 g/m². The plants were harvested by hand during mid November when 50% of the B. juncea plants were flowering. Immediately after harvesting the plants were separated into roots and shoots before being frozen at -20 °C. When needed for the growth tests, the plant material was finely chopped using a Braun 800W food processor (Gillette Australia), freeze-dried (Hetosicc CD4 freeze dryer, Foss Electric Australia) and sterilised using a gamma irradiator (Gammacell 220, Cobolt 60) for 16 hrs (2.37 Megarads).

• • 1. Determination of suppression (growth tests)

Disks of semi-permeable cellulose cellophane (Hallmark, Australia) were washed and sterilised as described by Dunne et al. (2003). The disks were placed on top of PDA (15 mL) before the colonised agar plugs (3 mm) were added to the centre of the plate. The colonised plates with lids removed were placed in 10 L plastic containers (Willow Ware, WA) along with a 100 mL glass vial containing either the Brassica tissue, ITC solution, or the 5 mL distilled water control (Figure 3.1). In the first experiment, two rates of biofumigant per L headspace (0.5 g/L, 1 g/L) were 48

investigated. The treatments in the second experiment consisted of 0.5 g/L of B. juncea root and shoot; 0.5 g/L of B. napus root and shoot; and 0.25 g/L of the root and shoot tissues from both B. juncea and B. napus. In the third experiment, two rates (0.1, 1 pL/L) of 97% PE- ITC (Sigma Chemicals, Australia) were investigated. To commence the hydrolysis reactions, 600 pL/mg tissue of sterile distilled water (pH 6.7) was added to the biofumigant tissue to commence the hydrolysis reaction. Immediately, the lids were placed on the containers and the containers were sealed with silver ducting tape. One copy of each experimental isolate was included in each of the three plastic containers.

Figure 3.1. The 10 L experimental chamber used in the current study illustrating the jar (A) containing the biofumigant tissue or phenylethyl isothiocyanate and the inoculated plates (B) with the cellophane membrane placed on top of the agar.

After 5 days incubation at 24 ± 1 °C in the dark, the cellophane disks were removed from the petri dishes and the diameter of each Phytophthora colony was measured twice (perpendicular to each other). The mycelia were scraped off the cellophane,

placed in pre-weighed 1.5 mL microfuge tubes, dried in a 60 °C oven for 72 hrs and the dry weight (mg) was determined.

• • 1. Quantification of biofumigant tissue

The ITCs were extracted from the freeze-dried Brassica tissue into ethyl acetate and analysed by gas chromatography. The ITCs were extracted by adding 10 mL 0.1 M CaCl2 and 10 mL of ethyl acetate (UniVar, Australia) to a centrifuge tube (50 mL) containing approximately 10 g of biofumigant tissue (dry weight). The tubes were shaken (G10 Gyrotory shaker, New Brunswick Scientific, USA) at 100 rpm (0.2 g) for 30 minutes before being centrifuged at 1500 (40 g) rpm for 5 min. The ethyl acetate supernatant was removed and placed at 4 °C until needed for gas chromatography. The extraction process was conducted twice producing approximately 20 mL ethyl acetate supernatant.

The concentration of ITCs in the ethyl acetate was determined by gas chromatography following the methods described by Warton et al. (2001a). Briefly, the ITCs extracted using ethyl acetate were dehydrated by passing the solution through a glass column of magnesium sulfate (AVL Chemicals, Australia). An internal standard of 100 pM methyl ITC (Sigma Chemicals, Australia) was added to all solutions tested in each ITC extraction. Samples were analysed using a Hewlett Packard 6890 GC equipped with a flame photometric detector in sulfur mode (394 nm). A 30 m x 0.32 mm internal diameter wall-coated-open-tubular (WCOT) fused silica capillary column coated with a 0.25 pm ethyl silicone stationary phase was used. The GC oven was programmed from 50-220 °C at 20 °C/min. Samples were injected using a HP 7683 auto sampler at an oven temperature of 50 °C. Helium was used as the carrier gas at a linear velocity of 19 cms^-1.

• • 1. Post treatment isolate viability

To determine if the suppression observed in all three experiments was fungistatic or fungicidal following each treatment, the colonised agar plugs were plated on to fresh PDA plates, incubated at 24 °C for 24-96 hrs and observed for growth using a BH-2 Olympus compound microscope at 200 x magnification.

• • 1. Statistical analysis

The data were analysed using the analysis of variance module of Statistica ('99 edition, Statsoft Inc., USA). The radial growth and dry weight (biomass) of the fungal isolates from the different treatments were expressed as the percent suppression compared to the mean of the controls for that treatment and isolate. Both dependant variables (dry weight and radial growth) were analysed together. To maintain an experimental error rate of 0.05, the Bonferroni correction was employed and α was set at 0.025. The growth data from the different Phytophthora species were analysed separately to the P. cinnamomi isolates. However, the P. cinnamomi isolate C10 was included with the other Phytophthora species to help in interspecific comparisons. Prior to analysis the data were assessed for homogeneity, variation of the mean from the variance and to determine if they were normally distributed.

• 1. Results
     1. Suppression by the root and shoot tissues of Brassica juncea and B. napus Inter-specific variation in Phytophthora

All five Phytophthora species showed similar responses when exposed to the different Brassica tissues (Figure 3.2). All species were highly (P < 0.01) suppressed by the root and shoot tissues of B. juncea. P. cinnamomi (C10), P. cactorum and P. citricola were more susceptible to the volatiles released from the B. juncea tissues than P. megasperma and P. cryptogea. The B. juncea shoot tissue was the most suppressive, often resulting in complete suppression. For example, even at the lower rate (0.5 g/L), the P. cinnamomi (C10) and P. cactorum isolates were 100% suppressed. In contrast, the B. juncea root tissue had no effect (P = 0.48) on the growth of the five Phytophthora species at the lower rate (0.5 g/L). However, when the B. juncea root was used at the higher rate (1 g/L) it was highly suppressive (P < 0.01) to the growth of the different species. The shoot tissues of B. napus did not suppress (P = 0.25) the growth of any of the Phytophthora species and the volatiles released from B. napus root tissues actually resulted in a significant (P < 0.01) increase in the growth of all Phytophthora species.

Radial growth

Biomass

Suppression (% of control)

P. cinnamomi C10

150

100

50

0

-50

• 1 0 0

P. megasperma

150

100

50

0

-50

• 100

P. cryptogea

150

100

50

0

-50

-100

P. citricola

150

100

50

0

-50

-100

P. cactorum

150

100

50

0

-50

-100

-150

P. cinnamomi C10

150

100

50

0

-50

-100

P. megasperma

150

100

50

0

-50

-100

P. cryptogea

150

100

50

0

-50

-100

P. citricola

150

100

50

0

-50

• 1 00

P. cactorum

150

100

50

0

-50

-100

-150

BJR BJS BNR BNS

BJR BJS BNR BNS

Brassica tissue

Figure 3.2. The suppression of growth of the five Phytophthora species in the presence of two rates (0.5 g/L = □; 1 g/L = ■) of decomposing Brassica juncea root (BJN), B. juncea shoot (BJS), B. napus root (BNR) and B. napus shoot (BNS). Error bars represent the standard error of the mean.

Intra-specific variation in P. cinnamomi

All six P. cinnamomi isolates responded similarly to the different Brassica tissues as described above for the other Phytophthora species. All isolates were completely suppressed (P < 0.01) by both rates of B. juncea shoot tissue (Figure 3.3). There was 52

Control of Sudden Death in cultivated proteas from the southwest of Western Australia

Suppression (% of control)

Biomass

P. cinnamomi C1 150 100 50

0

-50

• 1 00

P. cinnamomi C10 150 100

50

0

-50

-100

P. cinnamomi MP97-12 150 100

50

0

-50

• 1 00

Radial growth

P. cinnamomi C1

P. cinnamomi MP96 150 100

50

0

-50

• 1 00

P. cinnamomi MP97-8

.-I- –

150

100

50

0

-50

-1 00

150

100

50

0

-50

-1 00

-150

P. cinnamomi MPD087

P. cinnamomi C10 150 100 50 0 -50

• 1 00

P. cinnamomi MP97-12 150 100

50

0

-50

-100

P. cinnamomi MP96 150 100

50

0

-50

• 1 00

P. cinnamomi MP97-8 150 100

50

0

-50

-100

-150

f

150

100

50

0

-50

-100

-150

ê

4-

P. cinnamomi MPD087 150 100

50

0

-50

-100

-150

BJR BJS

BNR BNS

BJR BJS

BNR BNS

Brassica tissue

Figure 3.3. The suppression of growth of the six Phytophthora cinnamomi isolates in the presence of two rates (0.5 g/L = □; 1 g/L = ■) of decomposing Brassica juncea root (BJN), B. juncea shoot (BJS), B. napus root (BNR) and B. napus shoot (BNS). Error bars represent the standard error of the mean.

no (P = 1.0) difference in the suppression achieved by the two rates of B. juncea shoot tissue. At the lower rate (0.5 g/L) the B. juncea root tissue had no effect (P = 0.62) on the growth of the five Phytophthora species. However, B. juncea root 53

(1 g/L) was highly suppressive (P < 0.01) to the growth of the different P. cinnamomi isolates. The both rates of the B. napus shoot tissues had no effect (P = 0.53, P = 0.58) the growth of any of the P. cinnamomi isolates. Both rates of B. napus root tissue resulted in a significant (P < 0.01) increase in the growth of all six P. cinnamomi isolates.

• • 1. Suppression by combining Brassica tissues

Inter-specific variation in Phytophthora

Combining the root and shoot tissues from B. juncea achieved an increase in the level of suppression (Figure 3.4). In contrast, the volatiles released from the B. napus tissues did not (P = 0.11) suppress the growth of the different Phytophthora species. In some treatments the B. napus tissues actually increased the growth of the different Phytophthora species. The combination of the different tissues from both Brassica species resulted in an intermediate level of suppression (Figure 3.5). P. cinnamomi, P. megasperma and P. cryptogea were more susceptible to the biofumigation effect than P. cactorum and P. citricola.

Biomass

Suppression (% of control)

P. cinnamomi C10 150 100 50 0 -50 -100 -150

P. megasperma 150 100

50 0

-50 -100 -150

P. cryptogea 150 100 50 0 -50 -100 -150

P. citricola 150 100 50 0 -50 -100 -150

P. cactorum 150 100 50 0 -50 -100 -150

BN

BJ BJ+BN

Radial growth

P. cinnamomi C10 150 100

50

0

-50

-100

-1 5

⁵P⁰ . megasperma

150

100

50

0

-50

-100

-150

P. cryptogea

P. citricola

150

100

50

0

-50

-100

-150

BN

BJ BJ+BN

Brassica tissue

Figure 3.4. The suppression of growth of the five Phytophthora species in the presence of decomposing Brassica juncea (BJ; □), B. napus (BN; ■) and or a mixture of both Brassica species (BJ+BN;^). Error bars represent the standard error of the mean.

Figure 3.5. The tissues of Brassica juncea were more suppressive than B. napus and the mixture of both Brassica species, as shown above for the Phytophthora cinnamomi isolate C10.

Intra-specific variation in P. cinnamomi

The combination of the different tissue types from the same Brassica species enhanced (P < 0.01) the suppressive effect on the growth of the different P. cinnamomi isolates (Figure 3.6). The combination of the B. napus tissues increased (P < 0.01) growth of the pathogen. The combination of the different tissues from both Brassica species resulted in an intermediate level of suppression.

• • 1. Suppression using synthetic PE-ITC

Inter-specific variation in Phytophthora

The PE-ITC solution suppressed (P < 0.01) the growth of the different Phytophthora species. All of the Phytophthora species were significantly (P = 0.02) suppressed when exposed to the lower rate (0.1 µL/L) of PE-ITC. When all of the Phytophthora species were exposed to the higher rate (1 µL/L PE-ITC), 100% suppression of growth (biomass) was achieved (Figure 3.7).

Radial growth

Biomass

150

100

50

0

-50

-100

-150

P. cinnamomi C10

150

100

50

0

-50

-100

-150

P. cinnamomi C1

Suppression (% of control)

150

100

50

0

-50

-100

-150

-200

-250

P. cinnamomi MP97-12

P. cinnamomi MP96

P. cinnamomi MP97-12

150

100

50

0

-50

-100

-150

-200

-250

P. cinnamomi MP96

P. cinnamomi MP97-8

P. cinnamomi MPD087

Figure 3.6. The suppression of growth of the six Phytophthora cinnamomi isolates in the presence of decomposing Brassica juncea (BJ; □), B. napus (BN; ■) and or a mixture of both Brassica species (BJ+BN;^). Error bars represent the standard error of the mean.

Suppression (% of control)

C10

Phytophthora species

Figure 3.7. Biomass (A) and radial growth (B) suppression of the five Phytophthora species by two rates (0.1 ^L/L = □; 1 ^L/L = ■) of synthetic phenyl-ethyl isothiocyanate. Error bars represent the standard error of the means.

Intra-specific variation in P. cinnamomi

The growth of the six P. cinnamomi isolates was highly suppressed (P < 0.01) when exposed to the PE-ITC solution. At the higher rate (1 µL/L) the growth (biomass) all of the isolates were 100% suppressed (Figure 3.8).

Suppression (% of control)

Figure 3.8. Biomass (A) and radial growth (B) suppression of the six Phytophthora cinnamomi isolates by two rates (0.1 ^L/L = □; 1 ^L/L = ■) of synthetic phenyl-ethyl isothiocyanate. Error bars represent the standard error of the means.

• • 1. ITC content of the Brassica tissues

The results of the GC showed that the different ITCs were present in the root and shoot tissues of the two Brassica species (Figure 3.9). Two ITCs (2-PE ITC and 2- Propenyl ITC) were found in the shoot tissue of Brassica juncea. In contrast, the root tissue of B. juncea root tissue only contained PE-ITC, but at a higher concentration (21.2 µmoles/g plant tissue). Although the tissues of B. napus had a greater number of ITC types present, they were found at lower concentrations than the ITCs found in the B. juncea tissue. Both the roots and shoots of B. napus had the same four ITCs present with only 4-Pentenyl and 2-PE at significant concentrations.

Figure 3.9. The isothiocyanates (µmoles/ g dry tissue) present in the tissues of the two Brassica species differed in the number and type of isothiocyanates (ITCs) found.

• • 1. Viability of isolates after exposure to volatiles

In experiments 1-3, only 13.5%, 18.7% and 28% of the isolates grew, respectively. This demonstrates that in the majority of cases the suppressive biofumigant effect was permanent, i.e. fungicidal. The findings also show that the ITCs liberated from Brassica tissue were more potent than the synthetic PE-ITC with fewer isolates being viable after their growth had been suppressed. The more sensitive

Phytophthora species (P. cinnamomi and P. cactorum) were often the isolates in which the suppressive effect of biofumigants was fungicidal.

• 1. Discussion

Biofumigants can suppress the in vitro growth of Phytophthora species, including P. cinnamomi. The results indicate that the volatile compounds produced by B. juncea were superior in their suppressive ability when compared to B. napus. Therefore, B. juncea would be a superior biofumigant if used to control Phytophthora species such as P. cinnamomi. These findings compare to those of Charron and Sams (1999) who reported that Indian mustards were superior in their ability to suppress Pythium ultimum when compared to B. campestris and B. oleracea. Furthermore, Mazzola et al. (2001) reported that although B. napus seed meal was effective in suppressing Rhizoctonia it was ineffective on a Pythium species. In contrast, Smolinska et al. (1997) reported that B. napus seed meal was effective in suppressing growth of a different Oomycete, Aphanomyces euteiches f. sp. pisi (Van Hall) Synd. & Hans. However, it is important to consider that the seed of B. napus contains significantly higher levels of ITC's than the other plant tissues (Smolinska et al. 1997). The current study showed that the differences in the ITCs present and their relative concentration between plant tissues and Brassica species resulted in the variability of suppression achieved.

The current study conclusively demonstrated a marked variation in the tolerance of the P. cinnamomi isolates when exposed to the volatile chemicals from Brassica plants. Studies on members from the closely related Pythium genus have reported significant inter-specific variation with some species considered sensitive (Charron and Sams 1999, Manici et al. 2000) and others resistant (Sarwar et al. 1998). Smith and Kirkegaard (2002) reported that Aphanomyces and Phytophthora were sensitive to PE-ITC on a limited number of isolates. Further experiments need to be conducted on a larger number of isolates to fully describe the level of inter- and intra-specific variation that exists in Phytophthora species.

Breeders of new biofumigant varieties should consider this inter-specific and intra­specific variation. Growers could create seed mixes of different Brassica varieties that would broaden the scope of the suppression achieved, or target Phytophthora species/isolates where particular Brassica species/varieties are known to be more effective. If biofumigation for the control of soilborne pathogens was to become a common technique in establishing horticultural crops it would be sensible to be able to rotate a number of different Brassica species/varieties to decrease the chance of pathogen resistance developing. This study also indicated that by increasing the concentration of biofumigant the suppressive effect on the growth of Phytophthora species could be enhanced. This could have relevance to the ideal sowing density of Brassica species when used as a biofumigant compared to sowing for agricultural purposes.

Manici et al. (1997) tested the suppressive nature of ITCs on a number of plant pathogenic fungi from different taxa, including Pythium irregulare. They reported that the different pathogenic taxa could be inhibited by these ITCs, however, they responded differently to the different ITCs. In the current study, it is likely that high concentrations of PE-ITC and 2-propenyl ITC found in B. juncea were the reason for the superior suppressive ability. This hypothesis is supported by Sarwar et al. (1998) who reported the level of suppression achieved in Pythium irregulare by PE-ITC and 2-propenyl ITC was even superior to methyl ITC (a commonly used synthetic fungicide). As the suppressive ability of shoot tissue of B. juncea was greater than the root tissue it could be concluded that 2-propenyl ITC was more suppressive than PE-ITC or that a synergy exists between these two ITC types. Furthermore, it is possible that B. napus was not as effective as it did not contain 2-propenyl ITC or that the ITCs present were not at a high enough concentration. However, the ITC concentrations recorded after the hydrolysis of both Brassica species were comparable to those reported to be suppressive to Phytophthora species by Morra and Kirkegaard (2002).

In the current study, the volatiles released from decomposing B. napus tissues resulted in stimulation of growth of P. cactorum. It is unclear as to the exact cause

of this stimulation. Stephens et al. (1999) compared suppression achieved by fumigants (metham sodium, methyl bromide) with suppression of the two biofumigants, B. juncea and B. napus. They reported that the fumigants reduced the number of Pythium propagules in the soils treated with the biofumigants. However, the two Brassica species actually increased the number of Pythium propagules. It is possible that the ITCs were liberated from B. napus with poor efficiency and therefore the volatiles from B. napus were less suppressive than those from B. juncea tissues (Brown et al. 1991).

This study demonstrated that biofumigants suppress the in vitro growth of Phytophthora species. The findings show that significant differences occurred between the two Brassica species studied and in the growth responses of the different Phytophthora species. The subsequent chapters in this dissertation investigate the infective and survival mechanisms of P. cinnamomi after the incorporation of biofumigants (Chapter 4), and aim to determine if the suppressive ability of biofumigants in the laboratory will equate to suppression in soil within a glasshouse (Chapter 5) and a protea plantation (Chapters 6 & 7).

Chapter 4 – Do biofumigants affect the
sporulation and survival of
Phytophthora cinnamomi?

• 1. Introduction

All stages in the life cycle of Phytophthora cinnamomi play a role in the disease it causes (Chapter 1.4.3). However, zoospores are the major infection propagule that can increase the inoculum potential of the pathogen (Zentmyer 1980, Shearer and Tippett 1989). Sporangia are produced when conditions, such as moisture and temperature, are favorable (i.e. Spring and Autumn) (Shearer and Tippett 1989). The sporangia release zoospores that swim, or are passively transported, in flowing soil water to new host plants (Shearer 1994). Chlamydospores are often formed in diseased roots (Mircetich et al. 1968) and allow for long-term survival during periods of adverse environmental conditions (low soil moisture, absence of susceptible tissue or high microbial activity). When favourable environmental conditions return, the chlamydospores can germinate resulting in an increase in the inoculum density of the pathogen within the soil.

The majority of research on the suppressive nature of biofumigants on Oomycetes, such as Phytophthora, has been based on disease incidence assays or in vitro mycelial growth tests (Smolinska et al. 1997, Sarwar et al. 1998, Charron and Sams 1999, Manici et al. 2000). Chapter 3 demonstrated that biofumigants could suppress the in vitro growth of Phytophthora species, particularly P. cinnamomi. However, there is no information available on the effect of biofumigants on the sporangia and chlamydospores of Phytophthora cinnamomi. Lewis and Papavizas (1971) investigated the effect of biofumigants on the sporulation of the oomycete, Aphanomyces euteiches. They reported that the products of glucosinolate hydrolysis (ITCs and sulfides) significantly reduced zoospore formation, motility and zoospore germination. The aims of the current study were to determine whether the biofumigants Brassica juncea and B. napus reduce sporangia and chlamydospore production of P. cinnamomi in soil.

• 1. Methods
     1. Experimental design

To determine the effect of the biofumigants on the sporulation of P. cinnamomi in soil, colonised Miracloth discs were added to plastic containers containing 500 g of a soil conducive to the pathogen and placed in a 20 °C constant temperature room in the light. After 24 hrs, 20 g of B. napus or B. juncea tissue was incorporated into the soil (500 g). The colonised Miracloth discs (15 discs/per container, 5 mm) were removed prior to addition of the biofumigants and then after 2, 4, 6, and 8 days. Three Miracloth discs per treatment were removed from each plastic container and halved using a sterile scalpel blade. One half of each disc was mounted on a microscope slide and observed using a compound microscope to determine the number of sporangia and chlamydospores present. The other half of each disc was added to a glass jar containing 150 mL of soil extract water and baited using Pimelia ferruginea Labill. leaves to determine infectivity (number of baits infected). A treatment consisting of non-amended soil extract was included as a control. All treatments were conducted in triplicate and the experiment was repeated.

• • 1. Phytophthora cinnamomi isolate

Prior to the experiment, the P. cinnamomi isolate CD10 was passaged through a susceptible host (Leucadendron salignum x laureolum) using the method described in Chapter 2.2.4. The isolate was then used to colonise sterilised (3 x 121 °C/20 min, 24 hr intervals) Miracloth discs (5 mm). After 14 days, the colonised Miracloth discs were added to 170 x 110 x 70 mm plastic containers (Bonson Industrial Co, Auckland, New Zealand) containing 500 g of soil (1 yellow sand: 3 sphagnum peat – Soils World, Western Australia). The colonised discs were placed at a depth of 30 mm in the soil filled containers. Each plastic container had eight 10 mm drainage holes in their base. The containers were flooded with distilled water, allowed to drain for 10 min and sealed with a clear plastic film (Glad wrap®, Glad products of Australia) to prevent evaporation.

• • 1. Preparation and quantification of the biofumigant material

The two Brassica species: B. juncea (Fumus®, AgSeed Research, Victoria) and B. napus (BQmulch®, Wrightson Seeds, New South Wales) were harvested after four months from the field plot as described in Chapter 3.2.3. The plant material was finely chopped using a Braun 800W food processor, immediately frozen and stored at -20 °C until required. To characterise the ITCs in the different Brassica species, a 10 g sub-sample was de-frosted, the ITCs were extracted using ethyl acetate and gas chromatography was conducted using the methodology described in Chapter 3.2.5.

• • 1. Monitoring of infective and survival structures

Three Miracloth discs per treatment were removed from each plastic container and halved using a sterile scalpel blade. One half of each disc was mounted on a microscope slide and stained for 1 hr with 1% aniline blue. Five fields of view (500 pm) were assessed at 200 x magnification using a BH-2 Olympus compound microscope to determine the number of: (i) intact sporangia; (ii) aborted/immature sporangia; (iii) sporangia that had released zoospores; (iv) sporangia that had undergone direct germination; (v) intact chlamydospores; and (vi) aborted/immature chlamydospores (Figure 4.1). The distinction of immature and mature sporangia and chlamydospores were based on mature sporangia measuring between 40-75 pm in length and mature chlamydospores between 31-50 pm in diameter (Rands 1922, Erwin and Riberio 1996).

The other half of each disc was added to a 200 mL glass beaker containing 150 mL of soil extract (10 g soil/L) water, containing 10 Pimelia ferruginea leaves. After incubation for 72 hrs @ 20 °C the Pimelia ferruginea leaves were plated onto NARPH agar plates (Huberli et al. 2000) and incubated at 24 °C for 72 hrs to determine the infectivity (number of baits infected) using a BH-2 Olympus compound microscope.

• • 1. Statistical analysis

The data were analysed using the ANOVA package of Statistica '99 Edition (Statsoft Inc, USA). Prior to analysis the data were assessed for homogeneity, variation of the mean from the variance, to confirm they were normally distributed. The percentage of baits infected from the soil cores and soil leachate was transformed (102 = infection occurred, 101 = no infection). Direct comparisons between the two dependant variables (treatment and the day of harvest) were made using LSD tests (α = 0.05).

Figure 4.1. Stages of Phytophthora cinnamomi sporangia and chlamydospore production observed during the current study: A) sporangia undergoing direct germination; (B) coralloid hyphae; (C) sporangia after zoospore release; (D) aborted or malformed sporangia; (E) slightly collapsed chlamydospore and; (F) germinating chlamydospore. Scale bars represent 20 µm.

• 1. Results
     1. Quantification of the biofumigant tissues

The ITC compounds detected after the hydrolysis of the Brassica tissues were similar to the ITC profiles of the freeze-dried Brassica tissues described in Chapter 3.3.4. Two types of ITC were found in the tissues of Brassica juncea, 2-Phenylethyl ITC (19.2 umole g fw plant tissue) and 2-Propenyl ITC (1.4 umole g fw plant tissue) (Figure 4.2). Four different ITCs were identified in Brassica napus: 2- Phenylethyl ITC (2.7 umole g fw plant tissue); 4-Pentenyl ITC (3.4 umole g fw plant tissue); 3-Butenyl ITC (0.8 pmole/g fw plant tissue) and sec-Butyl ITC (0.1 umoleg fw plant tissue).

Propenyl sec-Butyl Butenyl Pentenyl Phenylethyl

ITCs

Figure 4.2. The isothiocyanates (ITCs) present in the Brassica juncea (|) and B. napus (□) plants used in the current study.

• • 1. The effect of biofumigants on Phytophthora cinnamomi sporangia

Brassica juncea suppressed (P < 0.01) sporangia production in the amended soil. Furthermore, there was a significant (P < 0.01) interaction between the day of harvest and treatment type on sporangial production (Figure 4.3). Maximum sporangial production was observed by day four in all treatments. Sporangial

production in the Brassica juncea amended soil was significantly (P < 0.01) reduced when compared to the other two treatments from day four onwards (Figure 4.3).

Figure 4.3. The mean sporangia produced in the Brassica juncea (—■—); and B. napus (—□—) amended soils, compared to the control ( ).

Error bars represent the standard errors of the means.

The addition of the biofumigants to the soil greatly (P < 0.01) reduced the number of intact sporangia produced by P. cinnamomi (Figure 4.4). The number of intact sporangia was lower (P < 0.01) in soils amended with B. juncea, compared to B. napus and the control treatment. The number of intact sporangia peaked at day four in all treatments (Figure 4.4).

25

0

4

Day

6

8

Figure 4.4. The mean number of intact sporangia produced in the Brassica juncea (—■—); and B. napus (—□—) amended soils, compared to the control ( ). Error bars represent the standard errors of the means.

Soils amended with the biofumigants had a higher (P = 0.02) rate of aborted sporangia or immature sporangia, compared to the control treatment (Figure 4.5). However, there was no difference (P = 0.99) in the percent of aborted/immature sporangia between the two Brassica treatments.

Figure 4.5. The percentage of aborted or immature sporangia in the Brassica juncea (—■—) and B. napus (—□—) amended soils, compared to the control ( ). Error bars represent the standard errors of the means.

The sporangia in the soils amended with the two Brassica species released zoospores earlier (P < 0.01) than the control treatment (Figure 4.6). In general, the number of sporangia that had released zoospores in all treatments was very low. There was no relationship (P = 0.82) between the different treatments and the number of sporangia that had germinated directly over the eight days of monitoring.

• • 1. The effect of biofumigants on Phytophthora cinnamomi chlamydospores Amending the soil with the two Brassica species had no suppressive effect on chlamydospore production. In all treatments, total chlamydospore production peaked (P < 0.01) at day two and day eight (Figure 4.7). A similar trend was also represented in the number of intact chlamydospores (Figure 4.8). However, there was no difference between the total number of chlamydospores (P = 0.20) or the number of intact chlamydospores (P = 0.39) in the biofumigant amended soils, compared to the control. Additionally, there was no relationship (P = 0.36) between the percentages of aborted or immature chlamydospores in the biofumigant-amended soil, compared to the control (Figure 4.9).

Figure 4.6. The mean percentage of sporangia that had undergone zoospore release in the Brassica juncea (—■—); and B. napus (—□—) amended soils; compared to the control ( ). Error bars represent the standard

errors of the means.

Figure 4.7. The number of total chlamydospores produced in Brassica juncea (—■—); and B. napus (—□—) amended soils, compared to the control ( ). Error bars represent the standard errors of the means.

Day

Figure 4.8. The mean number of intact chlamydospores produced in

Brassica juncea (—■—); and B. napus (—□—) amended soils, compared to

). Error bars represent the standard errors of the means.

the control (

Figure 4.9. The mean percentage of aborted or immature chlamydospores produced in Brassica juncea (—■—); and B. napus (—□—) amended soils, compared to the control ( ). Error bars represent the standard errors

of the means.

• • 1. The effect of biofumigants on the infectivity of Phytophthora cinnamomi The infectivity (percent baits infected) of P. cinnamomi was reduced (P < 0.01) when the soils were amended with the biofumigants. There was no difference (P = 0.25) in the reduction in the infectivity between the two Brassica species. Maximum infection (80% baits infected) occurred at day four under the control conditions and stayed high throughout the monitoring period (Figure 4.10). The infectivity within the soils amended with B. juncea tissues also peaked at day four, however, the percentage of baits infected was significantly (P < 0.01) reduced (40% baits infected) compared to the control. The soils amended with B. napus also reduced (P < 0.01) the infectivity of the pathogen compared to the control. Maximum infection in the soils amended with B. napus occurred at day two (40% baits infected) and then fluctuated between 20-40% of the baits infected for the next 6 days. There was a significant correlation (r = 0.63) between the infectivity and total sporangia numbers.

Day

Figure 4.10. The percentage of Pimelia ferruginea baits infected by Phytophthora cinnamomi zoospores in the Brassica juncea (—■—); and B. napus (—□—) amended soils, compared to the control ( ). Error

bars represent the standard errors of the means.

• 1. Discussion

The sporulation and infective ability of P. cinnamomi was significantly reduced in biofumigant-amended soils. B. juncea and B. napus significantly increased the number of aborted or immature sporangia and reduced the infective ability of the pathogen. Soils amended with B. juncea significantly reduced sporangia production. Biofumigants have previously been shown to suppress the formation of sporangia and zoospores in the oomycete, Aphanomyces euteiches (Lewis and Papavizas 1971, Smolinska et al. 1997). Lewis and Papavizas (1971) demonstrated that ITCs and sulfides derived from the hydrolysis of glucosinolates suppressed zoospore formation, motility and germination. Smolinska et al. (1997) reported significant reductions in zoospore production and disease severity using seed meal from B. napus. However, the present study is the first record that biofumigants suppress sporangia formation and subsequent plant infection by P. cinnamomi.

The current study demonstrates a complex mode of suppression by the biofumigants that disrupt the pathogen at a number of stages in its disease cycle (Chapter 1.4.3). The ability of B. juncea to decrease the infectivity of P. cinnamomi is likely to be caused by a number of factors. These include an overall reduction in the number of sporangia produced, an increased rate of sporangial abortion and a reduction in zoospore viability. In contrast, the tissues of B. napus had no significant effect on sporangia production. Therefore, the ability of B. napus to decrease the infectivity of P. cinnamomi is probably due to a reduction in zoospore formation, motility or germination. Hoitink et al. (1977) reported that leachates from hardwood compost reduced the impact of P. cinnamomi by lysing the zoospores. Furthermore, Hardy and Sivasithamparam (1991c) reported that composts that reduced sporangia production do not necessarily correspond to reduced infection rates in Telopea speciossima (Smith) R.Br.

Brassica juncea and B. napus had no significant affect on zoospore release from the sporangia, the number of intact chlamydospores or the percentage of aborted/immature chlamydospores. However, it is unclear if the chlamydospores observed were viable, ie could germinate. If the chlamydospores are not affected by

biofumigants they may germinate once the biofumigant tissue has decomposed and reactivate the infection.

The differences in the nature of suppression achieved by the two Brassica species are likely to be related to the different products of glucosinolate hydrolysis, particularly the ITCs. For example, the high concentration of PE-ITC in B. juncea could account for its suppression of sporangial production. In contrast, the reduction in zoospore viability by B. napus could be in part due to the 3-Butenyl ITC and 4-Pentenyl ITC that were present in significant concentrations. This is supported by the findings of Smolinska et al. (1997) who reported that B. napus seed meal suppressed zoospore germination in Aphanomyces euteiches. GC analysis of the seed meal found high levels of 3-Butenyl ITC and 4-Pentenyl ITC. It is unclear what role other chemical compounds, such as non-glucosinolate derived volatile compounds produced during decomposition of Brassica tissues, may be contributing to the biofumigant effect (Bending and Lincoln 1999).

Further research is needed to fully elucidate how biofumigants suppress P. cinnamomi. The current experimental design could be easily adapted to investigate the role of other factors in the biofumigation effect. For example: what is the role of matric potential, aeration, pH and other soil characteristics in the biofumigant affect; would the addition of antagonist bacteria to the soil increase suppression; are there more suppressive Brassica species than B. juncea and B. napus; would combining the tissues of different Brassica species increase suppression? Furthermore, by purifying solutions of individual ITCs, the effect of each ITC on the disease cycle could be determined.

This study demonstrated that the two biofumigants (B. juncea and B. napus) suppressed sporangial production in P. cinnamomi. B. juncea and B. napus significantly increased the number of aborted or immature sporangia and reduced the infective ability of the pathogen. Only, soils amended with B. juncea significantly reduced sporangia production. Neither Brassica species affected zoospore release or chlamydospore production. It is not yet clear whether the suppression of

P. cinnamomi observed in the current study will equate to a reduction in disease incidence when applied to field conditions. The subsequent Chapters in this dissertation investigate if biofumigants will suppress P. cinnamomi within a glasshouse (Chapter 5) or on a protea plantation in the southwest of Western Australia (Chapters 6 & 7).

Chapter 5 – Do biofumigants affect inoculum
potential, infectivity and disease incidence in
Phytophthora cinnamomi?

Conference proceedings:

Dunne, C.P., Dell, B. and Hardy, G.E.St J. Biofumigants suppress the growth, sporulation and survival of Phytophthora cinnamomi. Proceedings from the International Congress in Plant Pathology, Christchurch, New Zealand, February 2003.

Note:

Dunne conducted the research

Hardy and Dell were supervisors

• 1. Introduction

Biofumigants have been reported to reduce the inoculum potential and disease incidence in a range of pathogens, including Aphanomyces eutiches (Lewis and Papavizas 1971), Fusarium oxysporum (Blok et al. 2000), Gaeumannomyces graminis (Kirkegaard et al. 2000), Pythium ultimum (Charron and Sams 1999), Rhizoctonia solani (Blok et al. 2000) and Verticillium dahliae (Subbarao et al. 1999, Shetty et al. 2000). Previous research on the effect of biofumigants on the inoculum potential, infectivity and the incidence of disease caused by Phytophthora species has been inconclusive. Vawdrey et al. (2002) reported that the Brassica napus reduced inoculum levels and root rot severity in pawpaw caused by Phytophthora palmivora. In contrast, Utkhede and Hogue (1999) demonstrated that B. napus had no effect on the inoculum potential and disease incidence of Phytophthora crown and root rot of apple. Furthermore, Coelho et al. (1999) reported that B. oleracea had no effect on the inoculum potential of P. capsici or P. nicotianae.

Chapters 3 and 4 demonstrated that biofumigants could suppress the mycelial growth and sporulation of P. cinnamomi. There is no information available on whether biofumigants suppress root rot caused by P. cinnamomi. The current study investigated the effect of two Brassica species, B. juncea and B. napus, on the inoculum potential, infectivity and disease incidence of P. cinnamomi. The study aimed to determine if biofumigants suppress the recovery and infectivity of the pathogen from soil and if they could reduce the incidence of disease in Lupinus angustifolius.

• 1. Methods
     1. Experimental design

The effect of biofumigants on the inoculum potential, infectivity and disease incidence of P. cinnamomi was investigated under glasshouse conditions. Pots containing potting mix were arranged on benches in a randomised complete block design (Steel and Torrie 1986). Half of the pots were inoculated with P. cinnamomi colonised Japanese millet seed and the other half were left non-inoculated. The pots were then subjected to three experimental conditions: amendment with B. juncea

tissue; amendment with B. napus tissue and, no amendment (control). Therefore, the experimental treatments were:

1. Control (no biofumigant) + no P. cinnamomi inoculum
2. Control (no biofumigant) + P. cinnamomi inoculum added
3. B. juncea + no P. cinnamomi inoculum
4. B. juncea + P. cinnamomi inoculum added
5. B. napus + no P. cinnamomi inoculum
6. B. napus + P. cinnamomi inoculum added

The inoculum potential and infectivity of the pathogen was monitored by sampling the soil leachate and soil cores prior to the addition of the biofumigants and then 7, 14, 21, and 28 days after the addition of the biofumigants. The soil cores and soil leachates were processed in two ways: plated onto Phytophthora selective media to determine the inoculum potential of the P. cinnamomi in soil; and baited using Pimelia ferruginea leaves to determine the infective ability of the pathogen. After 28 days, the soil from each pot was transferred to a seedling tray and 10 seeds of Lupinus angustifolius were planted in each tray. The germination of the seeds and their subsequent growth was monitored at 14 and 28 days after planting. Seedlings with symptoms of P. cinnamomi infection (wilting, chlorosis and collar lesions) were sampled to confirm the presence of the pathogen using the methodology described in Chapter 2.2.2). All treatments were replicated four times and the experiment was repeated.

• • 1. Biological materials

Phytophthora cinnamomi isolate

P. cinnamomi isolate CD10 was passaged through a susceptible host (Leucadendron salignum x laureolum) as described in Chapter 2.2.4. The isolate was then added to 250 ml volumetric flasks containing 40 mL of DI water and 50 g sterilised Japanese millet (Panicum milaceum L.) seed that had been autoclaved on three successive days at 121 °C/20 min. The flasks containing the inoculated millet seed were incubated in the light at 20 °C for 8 weeks and shaken periodically to ensure uniform

colonisation. Prior to the experiment, 20 random millet seeds were removed from each flask and plated onto NARPH (Hüberli et al. 2000) agar plates, incubated at 20 °C for 3-5 days and assessed to ensure the millet seeds were fully colonised.

Biofumigant tissue

The biofumigant tissue used was the same as that described in Chapter 4.2.3. The two Brassica species: B. juncea (Fumus®, AgSeed Research, Victoria) and B. napus (BQmulch®, Wrightson Seeds, New South Wales) were harvested from the field plot as described in Chapter 3.2.3. The plant material was finely chopped using a Braun 800W food processor (Gillette, Australia) and immediately frozen at -20 °C. To characterise the isothiocyanates (ITCs) in the different Brassica species, a sub­sample (10 g) was later de-frosted, the ITCs were extracted using ethyl acetate and gas chromatography was conducted as described in Chapter 3.2.5.

• • 1. Monitoring of inoculum potential, infectivity and disease incidence

The colonised millet seed was mixed (1% weight: weight) with a potting mix (yellow sand 1: sphagnum peat 3 – Soil World, Western Australia) conducive to the pathogen and placed in 15 cm free-draining polyethylene pots. Half of the pots were left non-inoculated to act as a control. To commence the experiment 100 g of macerated B. juncea or B. napus tissue was mixed into the biofumigant treatment pots (Figure 5.1). All pots were watered to container capacity and allowed to drain. The pots were then watered for 10 min, twice a day with overhead sprinklers (2 Lhr^-1).

The inoculum potential and infectivity of the pathogen was monitored by sampling the soil leachate and soil cores prior to the addition of the biofumigants and then 7, 14, 21, and 28 days after the addition of the biofumigants. The pots were flooded to container capacity for 48 hrs before soil cores and soil leachates were sampled. At each harvest the pots were drained and the leachate was collect in glass petri dishes (200 mm). To determine the inoculum potential of the soil leachate, a 3 mL aliquot of the leachate was plated onto a NARPH agar plate using an automatic pipette. The plates were left agar side down for 1 hr before being incubated agar side up at 20 °C.

After 72 hrs, the plates were assessed for the number of P. cinnamomi colonies using an Olympus BH-2 microscope (200 x). To determine the infectivity of the pathogen in the soil leachate, a 10 mL aliquot of the leachate was baited in plastic containers (170 x 110 x 70 mm) filled with 300 mL distilled water and with 10 rose petal discs (8 mm) floating on the surface (Figure 5.1). After 48 hrs, the petals were plated onto NARPH plates, incubated at 20 °C for 24-72 hrs before the percentage of the baits infected was determined.

To determine the inoculum potential and infectivity of the pathogen, four soil cores were collected per pot using an 18 mm diameter x 200 mm long rigid plastic tube. The soil cores from each pot were mixed and three sub-samples (2.5 g soil) were spread across the surface of three NARPH agar plates. The plates were incubated for 24 hrs at 20 °C before the soil was removed by gently washing with DI water. The plates were further incubated for 24-72 hrs before the total number of P. cinnamomi colonies per plate was determined. Another sub-sample (2.5 g) of the soil cores was baited using the methodology describe above (Section 5.2.3) for baiting the soil leachate.

At day 28, the remaining soil in the pots was transferred to seedling trays. To determine if the biofumigants would suppress disease incidence in a susceptible host, 10 L. angustifolius seeds were added to each tray. The seedlings were watered for 10 min, twice a day with overhead sprinklers (2 L/hr). The trays were assessed after 14 days to determine the percentage germination of L. angustifolius and after 28 days to determine the disease incidence of the pathogen in the biofumigant amended soil. Any seedlings that showed signs of P. cinnamomi infection (root rot, chlorosis and wilting) were plated onto NARPH agar plates to confirm the presence of the pathogen.

• • 1. Statistical analysis

The data were analysed using the ANOVA package of Statistica '99 Edition (Statsoft Inc, USA). Prior to analysis the data were assessed for homogeneity, variation of the mean from the variance to confirm they were normally distributed.

The percentage of baits infected from the soil cores and soil leachate was transformed (102 = infection occurred, 101 = no infection). Direct comparisons

between the three treatments were made using LSD tests (α = 0.05).

Figure 5.1. Elements of the experimental design used in the current study: (a) the Brassica tissue was chopped using a food processor; (B) mixed with soil in plastic containers; (C) added to pots containing P. cinnamomi colonised millet seed; (D) after 28 days the soil was transferred to seedling punnets and Lupinus angustifolius seeds were planted in the punnets. Scale bar represents two cm.

• 1. Results
     1. Inoculum potential and infectivity in soil cores

The two biofumigants, B. juncea and B. napus, significantly (P < 0.01) reduced the inoculum potential (number of P. cinnamomi colonies) within the soil cores compared to the non-amended soil (Figure 5.2). There was no difference (P = 0.23) in the reduction of inoculum potential between the two Brassica species. Both Brassica species had an immediate suppressive effect on the inoculum potential in

the soil cores (Figure 5.3). For example, at day 7 the soils amended with B. juncea tissue completely suppressed the recovery of the pathogen. After this initial suppression there was a gradual increase in the inoculum potential in the soils amended with the biofumigants over the remaining period of monitoring, indicating fungistasis.

Figure 5.2. The biofumigant treatments (Brassica juncea and B. napus) reduced the number of Phytophthora cinnamomi colonies after 28 days, compared to the control. Black dots demarcate the P. cinnamomi colonies. Scale bar represents two cm.

Day

Figure 5.3. The number of Phytophthora cinnamomi colonies in the soil cores from the non-amended soil ( ); Brassica napus (—□—); and

B. juncea (—■—) amended soils. Error bars represent standard errors of the means.

The infectivity of P. cinnamomi in the soil cores was significantly (P < 0.01) reduced in the soils amended with the biofumigants (Figure 5.4). The greatest suppression was observed at 7 days after the incorporation of the biofumigants. After day 21 the reduction in the infectivity of the pathogen was negligible. There was no difference (P = 0.48) between the suppression in infectivity of the pathogen in the soils amended with B. juncea or B. napus.

Day

Figure 5.4. The infectivity (baits infected) of Phytophthora cinnamomi in the soil cores from the non-amended soil ( ); Brassica juncea (—■—); and

B. napus (—□—) amended soil. Error bars represent standard errors of the means.

• • 1. Inoculum potential and infectivity in soil leachate

The biofumigants treatments reduced (P = 0.03) the number of P. cinnamomi colonies in the soil leachate. Brassica juncea and B. napus reduced the number of P. cinnamomi colonies in the soil leachate compared to the control, P = 0.01 and P = 0.04, respectively (Figure 5.5). At 7 days after the incorporation of the biofumigants, the recovery of the pathogen was reduced to 76% of the control in B. napus amended soils and was completely suppressed in B. juncea amended soils. The treatments had no effect (P = 0.36) on the infectivity of the pathogen in the soil leachate. The infectivity of the pathogen varied significantly (P < 0.01) on the different days of harvest, however, there was no clear trend in the data.

Figure 5.5. The number of Phytophthora cinnamomi colonies in the soil leachate from the non-amended soil ( ); Brassica juncea (—■—); and

B. napus (—□—) amended soil. Error bars represent standard errors of the means.

• • 1. Disease incidence in Lupinus angustifolius

There was no difference (P = 0.12) between the treatments in the incidence of disease caused by P. cinnamomi in L. angustifolius after 14 days (Figure 5.6). However, after 28 days there was a significant (P < 0.01) difference between the three treatments. Soils amended with B. juncea resulted in reduced (P = 0.01) disease incidence (65% of lupin seedlings dead), compared to the control (90% of lupin seedlings dead) (Figure 5.7). Soils amended with B. napus had no effect (P = 0.79) on the incidence of disease after 28 days. The pathogen was recovered from 94%, 90% and 69% of plants that died in the control, B. napus and B. juncea treatments, respectively.

Figure 5.6. Phytophthora cinnamomi plant deaths in Lupinus angustifolius in different soil treatments over 28 days: control ( ); Brassica juncea

(—■—); and B. napus (—□—). Error bars represent standard errors of the means.

Figure 5.7. Lupinus angustifolius seedlings 28 days after planting illustrating the suppressive effect of amending soil with Brassica juncea, compared to B. napus and the control. Scale bar represents 10 cm.

• 1. Discussion

The two biofumigants suppressed the inoculum potential and infectivity of P. cinnamomi in soil. Both B. juncea and B. napus greatly suppressed the inoculum potential of the pathogen from the soil cores and soil leachate. Seven days after the incorporation of the pathogen, the B. juncea and B. napus treatments caused 100% and 76% reduction of inoculum potential of P. cinnamomi, respectively. This is comparable to Vawdrey et al. (2002) who reported that a mixture of B. napus and B. campestris caused 100% reduction in the inoculum potential of P. palmivora (Bulter) Bulter. Of the two Brassica species, only B. juncea reduced the incidence of root rot in L. angustifolius. Utkhede and Hogue (1999) reported that B. napus had no effect on disease incidence caused by Phytophthora species.

Although both B. juncea and B. napus caused significant reductions in the inoculum potential of P. cinnamomi, it was clear that they differ in how they achieve this reduction. B. juncea was highly suppressive and has been shown to suppress the mycelial growth (Chapter 3.3), sporangia production (Chapter 4.3), inoculum potential and disease incidence of P. cinnamomi (Chapter 5.3). B. napus is less suppressive, it only had a minimal effect on sporangia production, and although it reduced the inoculum potential it did not suppress disease incidence. Previous studies have reported that a reduction in the inoculum potential or sporangia production of P. cinnamomi did not always equate to a reduction in disease incidence (Hardy and Sivasithamparam 1991b). Vawdrey et al. (2002) reported that the suppression of P. palmivora did not correspond to a reduction in the incidence of disease in pawpaw. In contrast, Spencer and Benson (1982) reported that a reduction in chlamydospore and sporangia production in P. cinnamomi could be correlated to reduced disease incidence in L. angustifolius. In the current study, the difference in the response of the pathogen to the two Brassica species is likely to be related to the glucosinolate content of each species. It would be of interest to screen individual ITCs, and other glucosinolate and non-glucosinolate derived chemicals to determine their relative suppressiveness to P. cinnamomi. It is possible that the different components of the life cycle of the pathogen are affected by different ITCs. Further experiments should be conducted to screen a number of Brassica species to

obtain the most efficient ITC combination for the suppression of all the life stages of P. cinnamomi.

The suppression of P. cinnamomi described in the current study was only short lived and the pathogen was able to recover to significant levels by day 28. This indicates that the biofumigants were disease suppressive rather than pathogen suppressive. Kreutzer (1965) reported soilborne pathogens are often unrecoverable after soil treatments are applied, but often reappear later in greater quantities. Kreutzer defined this as the vacuum effect and hypothesised it was caused by the killing or inhibition of antagonists which permits the reinvading pathogen to grow through the soil without biological opposition. Biofumigants suppress the growth, sporulation and germination of the pathogen, but not the survival. Vawdrey et al. (2002) also reported that biofumigants initially suppressed the recovery of P. palmivora by 100%, however, after the initial suppression the pathogen was able to gradually increase its inoculum potential. I hypothesize that chlamydospores may have a role in this post-treatment increase in inoculum potential. The chlamydospores of P. cinnamomi are resistant to biofumigant effects (Chapter 4.3.3) and can germinate once the biofumigant tissue has decomposed and ITCs have disappeared or been sufficiently diluted. The chlamydospores then germinate to produce vegetative mycelia capable of further increasing the inoculum level of the pathogen by the production of sporangia and zoospores. Further experiments are necessary to determine if other Brassica species can reduce the chlamydospore production or viability in Phytophthora species.

The two biofumigants, B. juncea and B. napus, reduced the inoculum potential of P. cinnamomi in soil. However, after the immediate suppression the pathogen was able to gradually increase in inoculum potential and infectivity. Only the suppression of B. juncea resulted in reduced root rot in L. angustifolius. The research described in the next two chapters describes two field trials conducted on a protea plantation in the southwest of WA. The first (Chapter 6) compares biofumigation with the use of two other soil treatments (soil solarisation, fumigation) in an integrated management approach to control P. cinnamomi. The

second trial (Chapter 7) compares the effectiveness of the B. juncea and B. napus in reducing the impact of P. cinnamomi infection in when applied on a field scale.

Chapter 6 – Can soil solarisation, fumigation and
biofumigation reduce Phytophthora cinnamomi
infection of Leucadendron Safari Sunset?

Conference proceedings:

Dunne, C.P., Dell, B. and Hardy, G.E.St J. Control of Phytophthora cinnamomi cultivated Proteas. Proceedings from the International Congress in Plant Pathology, Christchurch, New Zealand, February 2003.

Note:

Dunne conducted the research

Hardy and Dell were supervisors

• 1. Introduction

Phytophthora cinnamomi is the major cause of sudden death in cultivated proteas in Western Australia (WA). A survey of protea plantations in the southwest of WA during 1999 to 2001 confirmed the presence of P. cinnamomi at 11 of the 28 plantations visited (Chapter 2). Plant losses caused by P. cinnamomi within diseased plantations can be as severe as 50% or greater. When growers re-plant on high impact sites, losses are immediate and devastating. Furthermore, the cost associated with establishing a new plantation is expensive. Therefore, growers need to consider methods of controlling the pathogen.

No single control treatment can eradicate P. cinnamomi from the soil in an infected site (Irwin et al. 1995). The use of phosphite and sterilisation of irrigation water can be used in an Integrated Management (IM) approach to help reduce the impact of the pathogen. The biological suppression of P. cinnamomi by mulches/composts has also been well documented (Hoitink et al. 1977, Hoitink and Fahy 1986, Hardy and Sivasithamparam 1991a). However, information is limited concerning the potential of pre-planting techniques such as biofumigation, fumigation and soil solarisation when used in an IM approach to control P. cinnamomi. The current study aimed to compare the effectiveness of biofumigation, fumigation and soil solarisation for suppressing P. cinnamomi infection of Leucadendron salignum x laureolum (Safari Sunset) under field conditions.

• 1. Methods
     1. Field trial design

The field trial was conducted on a protea plantation in the southwest of WA. Three soil treatments (biofumigation, fumigation, solarisation) were combined with the use of a hardwood mulch and compost. Two control treatments were included: a positive control (no soil treatment with mulch/compost); and a negative control (no soil treatment, compost or mulch). Therefore, the experimental treatments were:

1. Soil solarisation + compost/mulch
2. Fumigation (metham sodium) + compost/mulch
3. Biofumigation (Brassica juncea) + compost/mulch
4. Positive control (compost/mulch only)
5. Negative control (no treatment)

The five treatments were replicated in four plots in a completely randomised design (Steel and Torrie 1986) (Figure 6.1). Each treatment plot contained between 18-22 plants. Half the plants in each treatment plot were inoculated by placing colonised Banksia stem pieces in the soil surrounding the plants. The remaining half of the plants were sham inoculated by placing non-colonised Banksia stem pieces in the soil surrounding the plants.

• • 1. Field trial history

The trial was conducted at a plantation 30 km NNE of Margaret River (155° 34°) in the southwest of WA during September 1999 – December 2002. The trial was conducted on a site with a history of severe P. cinnamomi disease (Chapter 2, Figure 2.3B). The soil on the soil site consists of a gray sandy loam with a distinct B- horizon. The site is adjacent to a paddock that can flood during the winter months. The trial site was cleared of diseased Leucadendron floribunda seven months prior to the commencement of the experimental treatments.

• • 1. Biological materials

Plant material

The Leucadendron hybrid, Safari Sunset (L. salignum x laureolum), was used as the host plant within the trial. Cuttings of approximately 15 cm were harvested at a disease-free plantation near the trial site. The bottom two-thirds of the leaves were removed from the cuttings before being sterilised with 70% ethanol for 15 sec. The cuttings were dipped in 3 g/L indole butyric acid (Richgrow Garden Products, Australia), placed into pots containing a propagation potting mix (sand [4]: perlite [4]: peat [1]; v:v:v) and placed in a misting chamber with hot beds (25 ± 1 °C) to

promote rapid root development. Within the chamber the cuttings were watered overhead using mist sprinklers for 15 sec, every 20 min. During this time the cuttings were treated alternatively with 0.5% Mancozeb (Barmac Chemicals, Australia), 0.2% Scala (AroEvo, Australia) and 0.1% Rovral (Rhone Poulend, Australia) once every 14 days according to the manufacturers instructions. These non-systemic chemicals were used for the control of anthracnose, grey mould, downy mildew and scale insects.

Figure 6.1. The experimental design used to compare the five soil treatments (1 = metham sodium, 2 = solarisation, 3 = control + mulch, 4 = control (with no mulch, 5 = Brassica juncea) included a split-plot (+P = inoculated, -P = non-inoculated) randomised design with four replicate plots for each treatment.

After two months in the misting chamber, the rooted cuttings were transferred to 10 cm black plastic tubes containing potting mix (sand [1]: peat [3]; v:v) and 5 g per L mix of slow release fertilizer (3-4 month Osmocote fertilizer, Scotts, USA) and placed in a shade house where they were watered with overhead sprinklers (5 Lhr^-1) twice daily for 5 min for 10 months before being transplanted to the field.

The Leucadendrons were planted in the trial in November 2000 (Figure 6.2). The plants were fertilized around their base using a slow release fertilizer (~5 g per plant) (NPK Blue Special – Whitford fertilizers, WA) and watered three times a week for 20 min over the first summer (December – March) using 2 Lhr^-1 inline drippers.

Figure 6.2. The Leucadendron salignum x laureolum (Safari Sunset) plants were planted in the treatment plots with the assistance of a number of protea growers from nearby plantations.

The water used for the irrigation in the trial was sourced from an open dam with a clay base (Figure 6.3). Prior to the trial, P. cinnamomi was found in the dam using the baiting technique described by McIntosh (1964). Therefore, a chlorine injection system was established in the irrigation system in order to sterlise the water before

use in the trial. The system chosen was a Venturi® (Watering Concepts, WA) that injected 12.5% sodium hypochlorite (APS Chemicals, Australia) into the mainline at a rate of 5-10 ppm (pg/g) chlorine.

Phytophthora isolate

The P. cinnamomi isolate (C10) used in this study was isolated from a diseased Leucadendron florida at the field site in August 1999. The P. cinnamomi isolate was used to colonise sterilised Banksia grandis stem pieces using the methodology described by D'Souza (2001). Briefly, B. grandis stems were cut into pieces (2 cm length) using a bandsaw. The plugs were rinsed and then soaked in distilled water for 24 hrs. The plugs were transferred to 2 L conical flasks containing 1 cm of distilled water in the bottom of the flask. Approximately, 150 Banksia stem pieces were added to each flask. The flasks were plugged with cotton wool and autoclaved (121 °C; 20 min) on three consecutive days. One-week-old cultures of P. cinnamomi C10 on potato dextrose agar were cut into 1 cm² squares and aseptically transferred to the flasks containing the autoclaved plugs. The inoculated flasks were incubated at 24 °C in constant light for 8-12 weeks before use. The Banksia inoculum produced was either used to monitor the survival of the pathogen over the duration of the trial (Section 6.2.5) and to inoculate half the plants in the trial (Section 6.2.6).

Figure 6.3. (A) An open-cut seasonal dam was used to irrigate the plants within the trial site; (B) chlorine was injected into the dam water prior to use in the irrigation system.

• • 1. Soil treatments

Two weeks prior to the treatments being established, the soil in the plots were rotary-hoed using a rotorvator (12 cm) to 10 cm deep into the soil to remove any clumps and the treatment areas were sprayed with herbicide, 0.5% Fusilade (Westfarmers, WA). After the three soil treatments had been conducted, composted mulch (Malatesta Green Waste Recycling, Bunbury, WA) were added at the time of

planting in 1 m wide strips down the dripper lines (Figure 6.4). The compost and mulch used in the trial were derived from various hardwood tree species native in WA, including a number of Banksia and Eucalyptus species. Approximately, 4 mm of compost (Custom Composts, Mandurah, WA) was added to the soil surface within the treatment areas, and then overlain to a depth of 100 mm with mulch.

Figure 6.5. Application of hardwood compost (Custom Composts) and hardwood composted mulch (Malatesta Green Waste Recyclers) to the treatment plots at the field trial site, November 2000.

Solarisation

The soil solarisation was conducted during the summer months of December 1999 –

March 2000. Treatment areas were rotorvated as described above and watered with

Control of Sudden Death in cultivated proteas from the southwest of Western Australia sprinkler irrigation for two hours before the topsoil was covered with 100 µm clear plastic (Jaylon, WA) and fastened on the sides with soil (Figure 6.6). Air space between the plastic sheeting and soil was minimised to allow for more efficient heating.

Figure 6.6. The soil solarisation was conducted over the summer months between 1999 – 2000 using 100 µm plastic sheeting.

Biofumigant

The biofumigant used in this trial is a registered variety of Indian mustard (B. juncea Fumus®, AgSeed Research, Australia). The biofumigant seeds (8 kg/ha) and NPK Blue special (12 kg/ha) were sown using a seed drill to a depth of 2 cm in late April 2000. The emerging Brassica seedlings were treated two weeks after planting, with 100 mL/ha Le Mat (Bayer, Australia) to prevent damage by the red-legged earth mite.

In August 2000, the Brassica plants were top dressed using a mixture of urea and NPK Blue Special fertilizer (Cresco Fertilisers, WA). When 50% of B. juncea plants were in flower (mid-September) the biofumigants were sprayed with 0.5% Fusilade. At this time, the biomass of Brassica per m² was determined in all

treatment plots. Briefly, the Brassica plants within three randomly chosen quadrants (1 m²) from each treatment plot were removed and fresh weight (mg) was determined. The sampled plants were then dried at 60 °C before the dry weight (mg) was determined. Two weeks after being sprayed with the herbicide, the biofumigants were slashed and incorporated into the topsoil using a rotorvator (Figure 6.6).

Metham sodium

The fumigant, metham sodium (Vapam, Cropcare Australiasia) was applied at the same time that the biofumigant was incorporated at 1 L active ingredient/10 m² to the treatment plots using a blade plow with a boom spray in front of the plow (Figure 6.7).

Figure 6.6. (A) Slashing Brassica juncea plants on the treatment plants on the treatment plot 2 weeks prior to planting trials. (B) Incorporation B. juncea plants immediately after slashing.

Figure 6.7. The metham sodium treatment was applied to the treatment areas at 1 L active ingredient/10 m².

• • 1. Soil analysis

To monitor the experimental treatments during their application and post-application a number of different soil analyses were conducted. These included: (i) monitoring of P. cinnamomi inoculum survival and infective ability (ii) soil temperature during the soil solarisation; (iii) ITC concentration in the soil after the addition of the fumigant and after the incorporation of the biofumigant; (iv) soil microbial activity; and (v) soil pH and electrical conductivity (EC). The timing of these different analyses is outlined in the time line presented in Table 6.1.

Inoculum monitoring

To determine the effect of the five experimental treatments on the survival of P. cinnamomi within the soil, colonised Banksia stem pieces (6.2.2) were placed at two depths (10, 20 cm) in the soil profile. Before burial, one colonised stem piece was placed in 5 mm ultra-violet (UV) resistant netting (Q.E. Marine, WA), tied closed with UV twine and a UV resistant identification tag attached with a length of this UV resistant twine (Figure 6.8). Three bags per treatment plot. The plugs within

Table 6.1. A timeline showing the timing of the different soil analyses conducted the experimental treatments during their application and post-application.

J

F

M

2002

2003

A

M

J

J

A

S

O

N

D

J

F

M

A

M

J

J

A

S

O

N

D

the netting were placed at the two soil depths with the UV tag placed at the soil surface (Figure 6.8). The stem pieces remained in the soil for a minimum period of 42 days before being removed at 6, 13, 23, 31, 52, 78 and 104 weeks after the treatments were conducted. Three replicate stem pieces were removed at both depths at each sample time.

Figure 6.8. (A) The bags made of netting used to recover the colonised Banksia stem pieces. (B) Addition of colonised Banksia stem pieces for monitoring survival and infective ability of Phytophthora cinnamomi.

The colonised Banksia stem pieces were harvested and the survival of the P. cinnamomi isolate was determined. To compare the initial effectiveness of each treatment, the survival of the pathogen was compared six weeks after the treatments were conducted. The harvested stem sections were cut longitudinally in two. Then one half was cut into

10 pieces and plated onto NARPH agar plates. The plated stem pieces were incubated at 24 °C in the dark for 64-120 hrs before they were assessed microscopically for the presence of P. cinnamomi.

To determine the infective ability of the pathogen, the second half of the Banksia stem section was placed into plastic containers (5 x 10 x 20 cm) and immersed in distilled water. After 24 hrs the water was replaced to remove any potential suppressive chemical compounds that may had leached from the stem piece, before ten healthy Pimelia ferruginea leaves were placed on the water surface to act as bait for potential zoospore infection. After 7 days incubation at 24 °C in the light, the leaves were removed from the plastic containers and plated onto NARPH agar plates. After 64-120 hrs incubation at 24 °C incubation in the light the plates containing the leaves were visually assessed for the presence of P. cinnamomi using a BH-2 Olympus microscope.

Temperature

The soil temperature was recorded during the soil solarisation treatment. The temperature was measured in an untreated area as a control. Temperature was measured in a solarisation plot and the ambient air temperature every 30 min using a Dataflow datalogger (Dataflow systems, Australia). Ambient temperature was measured by placing a probe in the shade created by a small tree adjacent to the soalrisation plot. Soil temperatures within the untreated area and solarisation plot were measured at three soil depths (10, 20 and 40 cm).

Isothiocyanate analysis

The concentration of ITCs after the application of metham sodium and after the incorporation of the Indian mustard plants was determined using gas chromatography. Soil samples were collected from a depth of 20 cm within the treatment plots: at the time of treatment; then after 1, 24 and 72 hrs. Soil was collected from an adjacent untreated area for a control. The extraction of the ITCs was conducted as described in Chapter 3.2.5.

Microbial activity

Soil microbial activity was measured at the time intervals specified in Figure 6.7 over the duration of the trial. The microbial activity was measured at three random points across the treatment areas at two soil depths (10, 20 cm). Microbial respiration was determined using the fluoroscein diacetate (FDA) (Sigma Chemicals, Australia) method described by Schnurer and Rosswall (1982). Briefly, 5 g of soil was added to centrifuge tubes containing 20 mL of sterile 60 mM potassium phosphate buffer (pH 7.6). The assay commenced by adding 0.2 mL of FDA (2 mg/mL) to the buffer-soil mix. The tubes were shaken (2 g) at 25 °C for 20 minutes on a G10 Gyrotory shaker (New Brunswick Scientific, New Jersey, USA) before adding 20 mL acetone to stop the reaction. The soil mixes were centrifuged at 10 g for 10 min before being filtered through Whatman (No 1) filter paper under vacuum. The filtrate was then transferred to 3 mL plastic curvettes and read in a Hitachi U-1100 (Tokyo, Japan) spectrophotometer at an optical density of 490 nm.

At each sample time, standard curves were prepared to quantify the FDA solution. Five screw cap tubes containing 0, 100, 200, 300 and 400 pg of FDA in 5 mL of phosphate buffer were heated in boiling water for 60 min. The hydrolysed FDA was transferred to centrifuge tubes containing 5 g of soil and 15 mL of phosphate buffer. The tubes were placed onto a rotary shaker at 2 g for 20 min at 25 °C before 20 mL of acetone was added. The samples were centrifuged and filtered as described above before the absorbance of each sample was read at 490 nm.

Physical soil characteristics

The pH and EC of the soil collected at both depths (10, 20 cm) for the determination of microbial activity were measured using the methods of Raymond and Higgins (1982). A 5 g sample of soil was added to 0.05 M CaCl2 and shaken at 2 g for 20 min on a rotary shaker before the pH and EC was measured using an Activon meter (Model 190 pH, mV).

• • 1. Inoculation

The plants were acclimatised for 15 weeks before half the plants within each of the treatment plots were inoculated using the colonised Banksia stem pieces (Section 6.2.2). Four holes were made using a metal rod approximately 40 cm from the base of each plant. Two of the holes were made to a depth of 20 cm and two to a depth of 40 cm. One P. cinnamomi colonised Banksia stem piece was added before the holes were re-filled with soil. The soil surrounding the non-inoculated plants was disturbed in a similar manner except no inoculum was added.

• • 1. Monitoring of plant death

After inoculation, symptoms of P. cinnamomi infection were monitored on a weekly basis for the duration of the trial. When plants showed symptoms of P. cinnamomi infection (chlorosis, wilting, root rot) or died asymptomatically, they were removed from the soil and isolations were conducted to confirm the presence of the pathogen. When there was a lesion at the base of the plant, the plant tissue at the active lesion front was preferentially sampled. The tissue was plated onto NARPH media and incubated at 24 °C for 64-120 hrs before being microscopically assessed for the presence of P. cinnamomi. If P. cinnamomi was not recovered, then root and shoot tissue or if the dying plant was asymptomatic the stem tissue and root tissue was baited using the methodology described in Section 6.2.4.

• • 1. Rainfall and temperature data

Meteorological data (rainfall, minimum and maximum temperature) were obtained from a record site at Busselton Regional airport, 38 km E/NE for the field site.

• • 1. Statistical analysis

All data collected during the current study were analysed using the ANOVA module of Statistica '99 edition (Statsoft, USA). Prior to analysis the data were assessed for homogeneity, variation of the mean from the variance to confirm they were normally distributed. Daily rainfall and temperature data were obtained from the Bureau of Meteorology, WA. The data were averaged and presented as total monthly rainfall, mean monthly maximum temperature and mean monthly minimum temperature (Figure 6.9).

The temperature data collected during the solarisation treatment was averaged and presented as mean weekly maximum and minimum temperature (Figure 10).

ANOVA's were conducted on the survival, infective ability, microbial activity data at the first harvest after the treatments had been conducted to determine the initial effect of the treatments of the P. cinnamomi and microbial populations within the soil. Direct comparisons were made between dependent variables in all ANOVAs using LSD tests where a = 0.05. The data collected (percent baits infected) during the assessment of the infective ability of the pathogen in the colonised Banksia plugs (6.2.5) were transformed prior to statistical analysis. The data were transformed as: 101 = zero percent of baits infected; 102 10-100% baits infected. The number of plant deaths in each replicate plot in all five treatments was averaged and presented as mean plant death (Figure 6.21).

• 1. Results

The mean monthly temperature and total monthly rainfall from a site near the protea plantation are shown in Figure 6.9. A total of 2,570 mm of rain was recorded between November 1999 and October 2003. The mean annual rainfall over the duration of the trial was 732.8 mm. The majority of rain fell from late autumn (May) to early spring (September) each year. There was no significant differences in minimum and maximum temperatures over the 3 year trial period. However, there was significantly lower winter rainfall (June, July, August) in the second year of the field trial.

• • 1. The soil treatments

Soil solarisation

The soil solarisation treatment significantly (P < 0.01) increased the soil temperatures compared to an adjacent untreated plot (Figure 6.10). Daily maximum soil temperatures often exceeded 50 °C at a soil depth of 10 cm and 30 °C at a soil depth of 40 cm. Mean weekly temperatures in the solarised soils ranged from 37-48, 28-39 and 25-32 °C at a soil depth of 10 cm, 20 cm and 40 cm, respectively. Mean weekly temperatures in the untreated soil ranged from 22-25, 23-28 and 24-39 °C at a soil depth of 10 cm, 20 cm and 40 cm, respectively. The mean daily ambient temperature during the application of the solarisation treatment ranged from 13-27 °C.

1. Mean monthly maximum & minimum temperature

1. Total monthly rainfall

Figure 6.9. The mean monthly air temperature (A) and total monthly rainfall (B) from a meteorological record station near the field trial site. Standard error bars are shown.

Biofumigation

The mean biomass of Brassica juncea in the biofumigant treatment plots at time of incorporation was 1093 g fresh weight/m² and 230 g dry weight/m² (Figure 6.11). The biomass from the different biofumigant treatment plots varied significantly (P = 0.02). The lowest recorded biomass was measured from the B. juncea plants in one of the four plots (mean biomass = 315 g fresh weight/m² and 85 g dry weight/m²).

Control

C ₅₀

Q

°ï 40 L.

CD 30

O

E 20

o

10

Solarisation

012345678

Week

..»..Depth 1

..>..Depth 2

. . _À. . . Depth 3

Ambient Temp

B

O 50

V 40

u_

Π30

0)

E 20

12345678

0

Week

Week

Maximum

Figure 6.10. The weekly average soil temperature during the eight weeks the soil solarisation was conducted during summer 1999/2000. The soil temperature was monitored at three depths (10 cm, 20 cm and 40 cm) and ambient temperature was also recorded. (A) Control minimum temperature, (B) control maximum temperature, (C) solarisation minimum temperature and (D) solarisation maximum temperature. Standard error bars are shown.

1. Fresh weight

F J L R Mean

1. Dry weight

Figure 6.11. The biomass of the Brassica juncea determined at the time of incorporation or the four different plots (F, J, L, R). Fresh weight (A) and dry weight (B) are shown. Standard error bars are shown. Letters that are not the same are significantly different (α = 0.05).

Six ITCs were detected in the soil from the B. juncea treatment areas, including: propyl­ITC, propenyl-ITC, butenyl-ITC, benzyl-ITC, methiobutyl-ITC and phenylethyl-ITC (Figure 6.12A). Maximum concentration (P < 0.01) of the six dominant ITCs was

recorded at 1 hr after incorporation, although significant (P = 0.02) concentrations were still present 24 hrs after the incorporation of the biofumigant. Phenylethyl and Propenyl- ITCs were the most concentrated of the ITCs detected in the B. juncea amended soil, reaching mean concentrations of 3.30 nmole ITC/g soil and 2.48 nmole ITC/g soil, respectively. No ITC was detected in the untreated areas, prior to the incorporation of the Brassica plants or at 72 hrs after incorporation in the biofumigant treatment areas.

Metham sodium

The results of the GC illustrated that the metham sodium was rapidly (P < 0.01) converted into methyl-ITC (MITC) once incorporated into the soil within the treatment areas. The concentration of MITC was highest at 1 hr after the application of the fumigant at 71 nmole MITC/g soil (Figure 6.12B). Substantial levels of MITC were still recorded at 24 hrs (21.5 nmole MITC/g soil) and 72 hrs (8.85 nmole MITC/g soil) after the fumigant had been applied. No MITC was recorded prior to the application or in the untreated areas during application.

Propyl

M M M Propenyl

—■ Butenyl

– ■- Benzyl

MeThiobutyl

■■■•■■ Phenylethyl

(B) Metham sodium

Figure 6.12. (A) The mean isothiocyanates (ITC) detected within the soil from the Brassica juncea treatment plots. (B) The mean methyl isothiocyanate detected within soil from the metham sodium treatment plots. Error bars represent the standard error of the means.

• • 1. Post-treatment analysis

Survival of Phytophthora cinnamomi

To compare the initial effectiveness of each treatment, the survival of the pathogen was compared six weeks after the treatments were conducted (Figure 6.13). The survival of the pathogen was significantly (P < 0.01) reduced by the metham sodium and biofumigation treatments. Metham sodium was superior in its suppression of P. cinnamomi, compared to the biofumigant treatment. Solarisation had no effect (P = 0.64) on the survival of the pathogen when compared to the control treatments. There was no difference in the recovery of the pathogen from the negative control (untreated) and positive control (plus mulch) treatments. The survival of the pathogen was significantly (P = 0.02) lower when the pathogen was recovered from a soil depth of 10 cm for the metham sodium (plus mulch) and Brassica juncea (plus mulch) treatments, compared to the recovery from 20 cm.

All three-soil treatments significantly (P < 0.01) reduced the survival (% recovery from the Banksia stem pieces) of P. cinnamomi over the three years of monitoring (Figure 6.14). The metham sodium treatment was more suppressive than the biofumigation (P = 0.02) and solarisation (P < 0.01) treatments. A seasonal trend was observed for all treatments: during the summer months recovery of the pathogen was reduced to almost nil; while in the subsequent spring months the recovery of the pathogen increased. The pathogen was not recovered in any of the treatment plots from 110 weeks (December 2002) after the treatments had been applied. The survival of the pathogen was significantly (P < 0.01) lower from a soil depth of 10 cm for all treatments, compared to the recovery from 20 cm deep.

(A) Soil depth 10 cm

(B) Soil depth 20 cm

Figure 6.13. The survival of Phytophthora cinnamomi from the colonised Banksia stem pieces at two soil depths (A = 10 cm; B = 20 cm) six weeks after the treatments had been established. Standard error bars are shown. Letters that are not the same are significantly different (α = 0.05).

(B) Soil depth 20 cm

Negative Control – mulch

■ ■ ■ ■ Positive Control + mulch Brassica juncea + mulch

– -■- – Metham Sodium + mulch Solarisation + mulch

Week

Figure 6.14. The survival of Phytophthora cinnamomi from the colonised Banksia stem pieces at two soil depths (A = 10 cm; B = 20 cm) over the trial period. Error bars represent standard error of the means.

Infective ability

The three soil treatments all reduced the infective ability (% baits infected) of P. cinnamomi within six weeks of the treatments being conducted (Figure 6.15). The metham sodium treatment was the most suppressive (P = 0.01) with a mean infective rate of 13% (soil depth 10 cm) and 32% (soil depth 20 cm). The infective ability of the pathogen in the control (untreated) plots was 83% (soil depth 10 cm) and 93% (soil depth 20 cm). The infective ability of the pathogen was significantly (P < 0.01) lower from a soil depth of 10 cm for all treatments, compared to the infective ability of the pathogen from soil depth of 20 cm.

The infective ability of P. cinnamomi over the three years of monitoring showed similar trends to that observed in the survival of the pathogen (Figure 6.16). There was a strong correlation (r = 0.75) between the infective ability and survival data. There was a marked seasonal fluctuation in the infective ability of the pathogen in all treatments i.e. the summer months coincided with very low rates of infection compared to the cooler months. All the soil treatments significantly (P < 0.01) reduced the infective ability of the pathogen. The infective ability of the pathogen had ceased in all treatments by week 110 (December 2002). The infective ability of the pathogen was significantly (P < 0.01) lower from a soil depth of 10 cm for all treatments, compared to the recovery from 20 cm deep.

■Negative Control – mulch

□ Positive Control + mulch

■ Brassica juncea + mulch

n Metham sodium + mulch

□Solarisation

+ mulch

(B) Soil depth 20 cm

Figure 6.15. The infective ability (% baits infected) of Phytophthora cinnamomi from the colonised Banksia stem pieces at two soil depths (A = 10 cm; B = 20 cm) six weeks after the treatments were conducted. Standard error bars are shown. Letters that are not the same are significantly different (α = 0.05).

Week

Figure 6.16. The infective ability of Phytophthora cinnamomi from the colonised Banksia stem pieces at two soil depths (A = 10 cm; B = 20 cm) over the three year monitoring period. Standard error bars are shown.

Soil microbial activity

The soil microbial activity varied significantly (P = 0.03) between the different treatments plots at 15 weeks after treatments were applied (Figure 6.17). Only the soils from the biofumigant (P = 0.02) treatment had significantly elevated microbial activity. The control (untreated) soils had the lowest microbial activity of all treatments. Although not significant the microbial activity of the control plus mulch and solarisation treatments was intermediate to the biofumigant and control treatments. There was no difference (P = 0.37) in the soil microbial activity between the two soil depths. However, the Brassica treatment did record a higher level of microbial activity at a soil depth of 10 cm compared to the soil depth of 20 cm.

The soil microbial activity in the soils from the different treatments varied significantly (P = 0.04) over the three years of monitoring (Figure 6.18). Only the biofumigation (P = 0.02) and control (plus compost/mulch) (P < 0.01) treatments significantly increased the soil microbial activity. There was no difference (P = 0.53) between the soil microbial activities at the two soil depths investigated.

₇₀₀ (A) Soil depth 10 cm

Treatment

Figure 6.17. The soil microbial activities within the treatment areas (A = soil depth 10 cm; B = soil depth 20 cm) 15 weeks after the treatments were conducted. Soil microbial activity is shown in relative units of fluorescein diacetate (FDA). Standard error bars are shown. Letters that are not the same indicate significant differences between treatments (α = 0.05).

• Negative Control – mulch
• Positive Control + mulch
• Brassica juncea + mulch
• Metham Sodium

+ mulch

• Solarisation

+ mulch

(B) Soil depth 20 cm 1000

Week

0

0 20 40 60 80 100 120 140 160

Week

Figure 6.18. The soil microbial activities within the treatment areas (A = soil depth 10 cm; B = soil depth 20 cm) over the three years of monitoring the field trial. Soil microbial activity is shown in relative units of fluorescein diacetate (FDA). Standard error bars are shown.

Soil pH and electrical conductivity

The soil pH varied significantly (P < 0.01) between the different treatments over the three years of monitoring (Figure 6.19). The soil pH in all the treatment plots, except the solarisation (P = 0.23) plots, differed significantly (P < 0.01) from the control (untreated) areas. The soil pH increased in all treatments over the first 50 weeks and then fluctuated around pH 5.5 for the remaining 100 weeks of monitoring. The soil pH at a soil depth of 20 cm was higher (P = 0.02) than the soil pH from a soil depth of 10 cm.

The soil electrical conductivity (EC) varied significantly (P < 0.01) between the different treatments over the three years of monitoring (Figure 6.20). There was a strong correlation (r =0.89) between the soil pH and soil EC. The soil EC decreased in all treatments over the first 50 weeks and then fluctuated around 50 mV for the remaining 100 weeks of monitoring. There was no difference (P = 0.07) between the soils EC at the two soil depths investigated.

Negative Control – mulch

■ ■ ■ ■ Positive Control

+ mulch

5.5

—■— Brassica juncea + mulch

– -■- – Metham Sodium + mulch

—Solarisation

4.5

+ mulch

0 20 40 60 80 100 120 140 160

Week

Week

Figure 6.19. The soil pHCaCl2 within the different treatment plots (A = soil depth 10 cm; B = soil depth 20 cm) over the three years of monitoring. Standard error bars are shown.

(A) Soil depth 10 cm 6.5

(A) Soil depth 10 cm

0 20 40 60 80 100 120 140 160

Week

(B) Soil depth 20 cm

Week

Figure 6.20. The soil electrical conductivity within the different treatment plots (A = soil depth 10 cm; B = soil depth 20 cm) over the three years of monitoring. Standard error bars are shown.

Leucadendron deaths

The plating out of dying plant tissue confirmed that metham sodium was the only treatment to significantly (P = 0.05) reduce P. cinnamomi infection in Leucadendron Safari Sunset, compared to the controls (Figure 6.21). The mean death rate in Leucadendron was reduced in the metham sodium treatment plots to 26% of the death rate in the control (untreated) areas. There was no significant (P = 1.0) difference in the number of Leucadendron deaths in the inoculated plants compared to the non-inoculated plants.

mulch mulch mulch

Figure 6.21. The mean plant deaths from the different soil treatments conducted on the field trial (■ = Phytophthora cinnamomi inoculated; □ = non-inoculated). Standard error bars are shown. Letters that are the same indicate significant differences (α = 0.05).

6.4. Discussion

All the soil treatments (B. juncea, metham sodium and solarisation) suppressed P. cinnamomi populations in the soil. However, only the suppression of metham sodium resulted in reduced infection in Leucadendron Safari Sunset. The manner in which suppression is achieved by each treatment appears diverse and complex. The findings of the current study indicate that the treatments utilised different elements of physical, chemical and biological control. For example, the metham sodium caused an immediate

and long-term suppression of the survival and infective ability of the pathogen. The cause of this effect is hypothesized to be largely chemical, caused by the production of toxic MITC in the topsoil (Utkhede and Smith 2000). Since metham sodium had no effect on the soil microbial activity it is further proof the mode of suppression is largely chemical. However, further investigations are required to determine the effects of metham sodium of other soil micro-organisims, in particular on niche manipulations and population shifts. The minimal effect of fumigants on soil microbial activity has been previously reported (Stephens et al. 1999). It is possible increased suppression could be achieved by combining the use of metham sodium with a biological amendment or composts/mulch (Utkhede and Smith 2000).

Although, B. juncea caused an immediate and sustained suppression of the recovery and infective ability of the pathogen, it did not significantly decrease the incidence of disease in Leucadendron Safari Sunset. The two biofumigant plots with lower biomass (Figure 6.11) did not significantly affect the number of Leucadendron deaths. Problems with the agronomy of the Brassica variety contributed to the varying levels of Brassica biomass in the different biofumigant replicate plots. However, the levels of ITC detected in soil were comparable to those reported to be suppressive to P. cinnamomi in vitro (Morra and Kirkegaard 2002). This nature of suppression is consistent with the findings in the in vitro experiments described in Chapters 4 and 5. However, in the current study B. juncea did not reduce disease incidence as illustrated in Chapter 5.3.3. Vawdrey et al. (2002) were also able to gain control of Phytophthora using biofumigants during a pot experiment, but were unable to replicate the effect when the biofumigation was conducted in the field. The dominance of propenyl-ITC and phenylethyl-ITC in the soil after the incorporation of the B. juncea is consistent with the ITC profiles of B. juncea described previously in this dissertation (Chapters 3.3.1, 4.3.5) and other published studies (Bending and Lincoln 1999, Kirkegaard and Sarwar 1999). The suppressive nature of B. juncea is hypothesised to be a result of a number of chemical (e.g. decomposing Brassica tissues released significant levels of toxic ITCs into the soil) and biological factors (e.g. the treatment increased the microbial activity in the soil, potentially increasing levels of antagonistic bacteria and fungi). The possible

mechanisms of control achieved by biofumigants will be discussed in detail later (Chapter 8.4.1).

In the current study, the solarisation treatment was only mildly suppressive to P. cinnamomi. This is in contrast to previous studies that reported solarisation to be effective in controlling P. cinnamomi (Barbercheck and Von Broembsen 1986) and a range of other Phytophthora species (Coelho et al. 1999, Coelho et al. 2001, Pinkerton et al. 2002). The soil temperatures achieved during the application of the solarisation treatment are comparable to other published studies (Hardy 1983, Coelho et al. 1999, McGovern et al. 2000). Nesbitt et al. (1979) reported soil temperatures >36 °C cause complete hyphal lysis in P. cinnamomi. Coelho et al. (1999) reported solarisation was as effective as fumigation in reducing populations of P. nicotianae and P. capsici. It is possible that the limited success of the solarisation treatment in the current study was compromised by the lag (10 months) between the treatment being conducted (Summer 1999/2000) and the addition of the compost/mulch (November 2000) (Figure 6.8). If solarisation is to be used a control technique for reducing the impact of P. cinnamomi it recommended that the mulch/compost is added and the proteas planted immediately after the solarisation is conducted. Although the current study was only able to demonstrate that solarisation is mildly suppressive, further investigations are required before soil solarisation is disregarded as a potential pre-planting technique.

Organic matter can increase microbial populations in soil and thereby contribute to the biological control of Phytophthora diseases (Weste and Vithanage 1979). Furthermore, hardwood composts and mulches have been previously shown to suppress sporangia and chlamydospore production in P. cinnamomi (Hoitink et al. 1977, Hardy and Sivasithamparam 1991c). In the current study, the positive control (plus mulch/compost) treatment had no effect on the survival, infective ability of the pathogen and disease incidence in Leucadendron Safari Sunset. It is possible the use of compost and mulch alone is not sufficient to reduce the impact of the pathogen. However, the use of composts in combination with soil treatments could reduce the impact of the pathogen. For example, Chapter 4 identified that biofumigants have no effect on chlamydospore production in P. cinnamomi. Hardy and Sivasithamparam (1991c) demonstrate that

compost can suppress chlamydospore production by the pathogen. Therefore, the combination of compost with biofumigation may create a synergy and result in increased suppression of P. cinnamomi.

While monitoring the survival and infective ability of pathogen a seasonal pattern developed. During the dry summer months the pathogen was only recovered at low rates, however, the recoveries increased during the following autumn and spring. Collins et al. (2001) reported a similar seasonal pattern in the recovery of the pathogen during an investigation of P. cinnamomi survival in the Eucalyptus marginata forest in southwest WA. They compared the survival of the pathogen in colonised Banksia stems in soils of a rehabilitated bauxite mine pit and an adjacent undisturbed forest site. They found the detection of P. cinnamomi would drop to 0% during summer, but then detection would occur in the autumn and winter harvest. It is well known that the isolation of P. cinnamomi is significantly reduced in soil with low soil moisture content (Hwang and Ko 1978, Old et al. 1984, Hansen and Hamm 1996). It is possible that during these times the pathogen becomes dormant in the form of chlamydospores. When favourable environmental conditions return, the chlamydospores can germinate to re-initiate the infection. To reduce the impact of dormancy affecting the survival and infective ability data, the harvested Banksia stem pieces were soaked for 24 hrs prior to being plated and baited. Hüberli et al. (2000) reported false-negative isolations when trying to recover the pathogen. They noted that repeated washing and re-plating of the colonised plant tissue resulted in increased recovery of the pathogen. They suggested that the pathogen is present in some dormant, but viable, unit such as chlamydospores. Further investigations are needed to describe this nature of dormancy in P. cinnamomi and determine methods for over coming dormancy.

The yearly rainfall (732 mm) at the field trial site was substantially lower than the average yearly rainfall (~1200 mm) (Bureau of Meterorology 2004). P. cinnamomi is a water mould and requires moist warm conditions to sporulate and cause disease (Halsall and Williams 1984). Therefore, in the current study the environmental conditions were not optimal for disease development. It is likely that this may have contributed to the poor suppressive ability of the different soil treatments when conducted in the field.

It is unclear what caused the increase in soil pH and decrease in soil EC immediately after the treatments had been conducted. Porter et al. (1998) also reported that the use of fumigants increased soil pH by up to 18%. It is possible the physical disturbance of the soil profile using the rotary-hoe could have been responsible for the shift in soil pH and EC. Regardless, soil pH from all treatments was between pH 5–6, which has been reported to be optimal for sporangia production, zoospore formation and motility in P. cinnamomi (Blaker and MacDonald 1983).

In the current experiment only half of the Leucadendrons were inoculated with the colonized Banksia stem pieces. During the design of the experiment this was considered as a compromise considering the distribution of the pathogen in the soil across the field site was unknown. Therefore, it is unclear as to the effect of the resisdent inoculum or the functional redundance of the added inoculum. It is possible that the efficacy of the treatments could have been masked by an excess of inoculum. Any further investigations should consider these possible effects when designing the methodology.

If the current field trial is to be considered a success it is essential that monitoring of Leucadendron deaths are continued. Cultivated proteas generally reach their maximum productivity in their 3^rd to 5^th year. For the trial to be considered a success, at least one of the soil treatments needs to have a significantly lower incidence of disease by the end of the 5^th year. Further investigations are also recommended to determine if the treatments affect the number of saleable stems per plant (Stephens et al. 1999).

The current study shows the potential of fumigation, biofumigation and solarisation to become common tools in an integrated approach to controlling P. cinnamomi. Metham sodium was the most effective of all the treatment trialed, however, it also was the most expensive. Biofumigation and solarisation were only shown to affect the survival and infective ability of P. cinnamomi and had no effect on the number of Leucadendron deaths. It is likely, with further refinement, the biofumigation and solarisation techniques could improve the level of suppression achieved (Chapter 8.4.2). For example, future research could investigate the effect of combining these two soil treatments. Coelho et

al. (2000) combined soil solarisation with biofumigation and reported that the Brassica amendment reduced the time required for the inactivation of P. nicotianae chlamydospores and increased the efficiency of inactivation. Growers considering these soil treatments for an integrated management plan should also include: sterilisation of irrigation water; fungicides (Cho 1981, Dixon et al. 1990, Marks and Smith 1992, Turnbull et al. 1995); plant a greater proportion of resistant varieties (McCredie et al. 1985); and the regular application of composted hardwood mulch (Hardy and Sivasithamparam 1991a). An integrated management control plan for P. cinnamomi in cultivated proteas will be discussed in detail later (Chapter 8.5).

All three soil treatments, B. juncea, metham sodium and solarisation were able to suppress the survival or infective processes of P. cinnamomi. However, within the three years of monitoring, metham sodium was the only treatment that actually reduced P. cinnamomi infection in Leucadendron Safari Sunset. The field trial described in the proceeding chapter compares the effectiveness of two biofumigant species, B. juncea and B. napus, to reduce P. cinnamomi infection in Leucadendron Safari Sunset. No organic amendments (mulches/compost) were included, as the trial aimed to ascertain the effect of the biofumigants alone.

Chapter 7 – Can Brassica juncea or B. napus reduce
Phytophthora cinnamomi infection of Leucadendron
Safari Sunset?

• 1. Introduction

Chemicals released from decomposing Brassica tissues suppress the growth of Phytophthora species, including P. cinnamomi (Chapter 3). Furthermore, Brassica species can reduce sporangia production, inoculum potential, infective ability and disease incidence of P. cinnamomi (Chapter 4 & 5). It is unclear if the suppression of the pathogen by the biofumigants will be achieved when applied in the field. The current study aimed to reduce the mortality of Leucadendron Safari Sunset planted within an active P. cinnamomi disease centre by using biofumigants. The study compared the effectiveness of two Brassica species, B. juncea and B. napus.

• 1. Methods
     1. Experimental design

The field trial was conducted on a protea plantation in the southwest of WA (Chapter 6.3.2). The trial consisted of three experimental treatments: (i) biofumigation using B. juncea; (ii) biofumigation using B. napus; (iii) and a control treatment in which no biofumigant was planted. The treatments were replicated three times in a completely randomised design and each replicate plot contained 30 Leucadendron Safari Sunset (Figure 7.1). Half the plants (15) in each treatment plot were inoculated with P. cinnamomi by placing colonised Banksia stem pieces in the soil immediately surrounding the plants roots. The remaining plants (15) were left non-inoculated.

• • 1. Biological materials

Host plants

The Leucadendrons were propagated as described previously (Chapter 6.2.3). The Leucadendrons were planted in the trial in November 2000. Upon planting the plants were fertilized using NPK Blue Special (Whitfords fertilizer, WA) and watered 2-3 times a week for 1 hr over the first summer (December-March) using 2 L/hr inline drippers.

The water used in the trial was sourced from an open-cut dam with a clay base and injected with 12.5% sodium hypochlorite into the mainline at a rate of 5-10 µg/g chlorine before use.

Figure 7.1. The experimental design used to compare the two Brassica species (Brassica juncea and B. napus) compared to an untreated area (control). The design included a split-plot (P+ = inoculated, P- = non-inoculated) randomised design with three replicate plots for each of the three treatments (1 = control, 2 = Brassica napus, 3 = Brassica juncea).

Phytophthora cinnamomi isolate

The P. cinnamomi C10 isolate used in this trial was isolated from Leucadendron florida (Chapter 2). Banksia grandis stem pieces (2 cm length) were colonised using the P. cinnamomi isolate using the methodology described by D'Souza (2001). The

inoculum produced was used to monitor the pathogens survival in soil over the duration of the trial (Section 7.2.3) and to inoculate half the plants in the trial (Section 7.2.4).

Biofumigants

Two weeks prior to the treatments being established, the soil in the plots were rotary- hoed with a rotorvator (12 cm) to 10 cm deep into the soil to remove any clumps and the treatment areas were sprayed with the herbicide, 0.5% Fusilade (Westfarmers, WA). The biofumigants used in this trial consisted of two registered varieties Brassica, including Brassica juncea (Fumus®, AgSeed Research, Australia) and B. napus (BQMulch®, Wrightson Seeds, Australia). The biofumigant seeds (8 kg/ha) and NPK Blue special (Cresco fertilisers, WA) (12 kg/ha) were sown using a seed drill to a depth of 2 cm in late April 2000. To prevent damage by the red-legged earth mite, the emerging Brassica seedlings were treated with 100 mL/ha Le Mat (Bayer, Australia) at 2 weeks after planting.

In August 2000, the Brassica plants were top dressed using a mixture of urea and NPK Blue Special (~10 g/m²). By mid-September 2000, 50% of the Brassica juncea were flowering. At this time, the biofumigants were sprayed with 0.5% Fusilade to kill the plants and allow for rapid decay once incorporated. At this time, the biomass (mg fresh weight/m²) of Brassica was determined in all treatment plots. Briefly, the Brassica plants within three randomly chosen quadrants (1 m²) from each treatment plot were removed and fresh weight (mg) was determined. The sampled plants were then dried at 60 °C before the dry weight (mg) was determined. Two weeks after being sprayed with the herbicide, the biofumigants were slashed and rotary hoed into the topsoil.

• • 1. Soil analysis

To monitor the experimental treatments during their application and post-application soil analyses were conducted as follows: (i) monitoring of P. cinnamomi inoculum survival and ability to cause disease; (ii) ITC concentration in the soil after incorporation of the biofumigants; (iii) soil microbial activity; (iv) soil pH and electrical conductivity (EC). The timing of the different soil analyses pre- and post incorporation of the biofumigants is shown in Table 7.1.

Table 7.1. The timing of the soil analyses conducted pre- and post-treatments being conducted.

Year

1999

2000

N

D

J

F

M

A

2001

M

J

J

A

S

O

N

D

J

F

M

A

2002

M

J

J

A

S

O

N

D

J

F

M

2003

AM

J

J

A

S

O

N

D

Inoculum monitoring

To determine the effect of the biofumigant treatments on the survival of P. cinnamomi within the soil, colonised Banksia stem pieces (Section 7.2.2) were placed at two depths (10, 20 cm) in the soil profile. Before burial the colonised stem pieces were placed in 5 mm ultra-violet (UV) resistant netting (Q.E. Marine, WA), tied closed with UV twine and a UV resistant identification tag attached with a length of this UV resistant twine. The plugs within the netting were placed at the two soil depths with the UV tag placed at the soil surface. The stem pieces remained in the soil before being removed at 9, 15, 28, 52, 76, 89, 126 and 154 weeks after the treatments were conducted. Three replicate stem pieces were removed at both depths at each sample time from each treatment. To determine the survival and infectivity of the pathogen the harvested stem pieces were plated onto Phytophthora selective agar and baited as described in Chapter 6.2.5.

Isothiocynate analysis

The concentration of ITCs after the incorporation of the two biofumigants was determined by gas chromatography as described in Chapter 6.2.5. Soil samples were collected from a depth of 20 cm within the treatment plots: at the time of treatment and; then after 1, 24, 72 and 168 hrs. Soil was collected from an adjacent untreated area to act as a control. The ITCs were extracted in ethyl acetate and placed at 4 °C until needed for gas chromatography. The extraction process was conducted twice producing approximately 20 mL ethyl acetate supernatant. The concentration of the ethyl acetate was determined by gas chromatography following the methods described by Warton et al. (2001a).

Microbial activity

Soil microbial activity was measured at the time intervals over the duration of the trial. The microbial activity was measured at three random points across the treatment areas at two soil depths (10, 20 cm). The microbial activity was determined by a spectroscopic assay using fluorescein diacetate (Sigma Chemicals, Australia) described in Chapter 6.2.5 (Schnurer and Rosswall 1982).

Physical soil characteristics

The pH and EC of the soil collected at both depths (10, 20 cm) for the determination of microbial activity were measured using the methods of Raymond and Higgins (1982). A 5 g sample of soil was added to 0.05 M CaCl2 and shaken at 2 g for 20 min on a rotary shaker before the pH and EC was measured using an Activon meter (Model 190 pH, mV).

• • 1. Inoculation

The plants were established for 15 weeks before half the plants within each treatment plot were inoculated using the colonised Banksia stem pieces (Section 7.2.2). Four holes were made using a metal rod (2 cm diameter) approximately 40 cm from the base of each plant. Two of the holes were made to a depth of 20 cm and two to a depth of 40 cm. One P. cinnamomi colonised Banksia stem piece was added before the holes were re-filled with soil. The soil surrounding the non-inoculated plants was disturbed in a similar manner, however no inoculum was added.

• • 1. Monitoring of disease incidence

After the inoculation, symptoms of P. cinnamomi infection were monitored on a weekly basis for the duration of the trial. When plants showed symptoms of P. cinnamomi infection (chlorosis, wilting, root rot) or died asymptomatically the plants were removed and harvested to confirm the presence/absence of the pathogen as described previously (Chapter 6.2.4).

• • 1. Statistical analysis

All data collected during the current study were analysed using the ANOVA module of Statistica '99 edition (Statsoft, USA). Prior to analysis the data were assessed for homogeneity, variation of the mean from the variance to confirm it was normally distributed. ANOVAs were conducted on the survival, infective ability, microbial activity data at the first harvest after the treatments had been conducted to determine the initial effect of the treatments of the P. cinnamomi and microbial populations within the soil. Direct comparisons were made between dependant variables in all ANOVA's using LSD tests (α = 0.05). The data collected (percent baits infected) during the

assessment of the infective ability of the pathogen in the colonised Banksia plugs (Section 7.2.5) were transformed prior to statistical analysis. The data were transformed as: 101 = zero percent of baits infected; 102 = 10-100% baits infected.

• 1. Results
     1. Treatment analysis

Isothiocyanate analysis

The two Brassica species produced distinct ITC profiles when incorporated into the soil (Figure 7.2). The maximum concentration of the ITCs detected from both Brassica species occurred at 1 hr after incorporation. ITCs were not detected in the B. juncea and B. napus amended soils at 72 hrs and 168 hrs after incorporation, respectively. The decomposition of the B. juncea tissues produced elevated levels of phenylethyl-ITC (3.3 nmoles/g soil) and propenyl-ITC (2.48 nmoles/g soil) compared to tissues of B. napus. In comparison, the B. napus amended soils had two additional ITCs (pentenyl-ITC, methiopenyl-ITC) not detected in the B. juncea amended soils and Propyl-ITC was the most concentrated (2.4 nmoles/g soil). No ITCs were detected in the control (untreated) areas, or prior to incorporation in the B. juncea treatment areas. However, four ITCs (propyl, propenyl, pentenyl and methiobutyl) were identified in the soils where B. napus was grown, immediately prior to incorporation.

Biofumigant biomass

The mean biomass of the two Brassica species was calculated to be 69 g per m² dry weight for the B. juncea treatment plots and 44 g per m² dry weight for the B. napus treatment plots (Figure 7.3). There was no difference (P = 0.40) in the biomass of the two Brassica species at the time of incorporation.

(A) Brassica juncea ^Propyl

Hour(s) after application

Figure 7.2. The isothiocyanates (ITCs) detected in soil from the Brassica juncea (A) and B. napus (B) treatment areas for 168 hrs after their incorporation. Standard error bars are shown.

(A) Brassica juncea

Dry weight

Figure 7.3. The biomass (fresh weight and dry weight) of the Brassica juncea (A) and B. napus (B) determined at the time of incorporation in the three replicate treatment plots (A, D, F). Standard error bars are shown.

• • 1. Post-treatment analyses

Both B. juncea and B. napus significantly (P < 0.01) reduced the survival (percent stem pieces colonized by P. cinnamomi) and infective ability of P. cinnamomi at nine weeks after their incorporation as biofumigants (Figure 7.4, Figure 7.5). The survival and infective ability of P. cinnamomi was affected when compared over the three years of monitoring. The survival of the pathogen was significantly (P < 0.01) reduced by the incorporation of the biofumigants (Figure 7.6). The survival of the pathogen was greater (P = 0.02) at a soil depth of 20 cm for all treatments, compared to a soil depth of 10 cm. A seasonal pattern in recovery (survival) was observed in all treatments, where the recovery during the summer months would drop off to almost nil and in spring the recovery would increase again. The pathogen was not isolated from any of the treatment areas at 152 weeks after the incorporation of the biofumigants.

There was a moderate correlation (r = 0.63) between the infective ability and survival data. For example, the infective ability of the pathogen was significantly (P = 0.03) reduced by the incorporation of the biofumigants and a similar seasonal pattern was observed (Figure 7.7). There was no difference (P = 0.86) in the suppression achieved by B. juncea and B. napus. Again, the seasonal pattern in infective ability of the pathogen was observed. The infective ability of the pathogen was significantly lower (P = 0.01) at a soil depth of 10 cm, compared to a soil depth of 20 cm.

Control

B. napus B. juncea

Figure 7.4. The survival of Phytophthora cinnamomi from the colonised Banksia stem pieces (A = soil depth 10 cm; B = soil depth 20 cm) nine weeks after the biofumigants were incorporated. Standard error bars are shown.

Control Brassica napus Brassica juncea

Figure 7.5. The infective ability of Phytophthora cinnamomi from the colonised Banksia stem pieces (A = soil depth 10 cm; B = soil depth 20 cm) nine weeks after the biofumigants were incorporated. Standard error bars are shown.

0 20 40 60 80 100 120 140 160

Week

0 20 40 60 80 100 120 140 160

Week

Figure 7.6. The survival of Phytophthora cinnamomi in the colonised Banksia stem pieces (A = soil depth 10 cm; B = soil depth 20 cm) over the three years of monitoring. Standard error bars are shown.

0 20 40 60 80 100 120 140 160

Week

Figure 7.7. The infective ability of Phytophthora cinnamomi recovered from the colonised Banksia stem pieces (A = soil depth 10 cm; B = soil depth 20 cm) over the three years of monitoring. Standard error bars are shown

The incorporation of the biofumigants had no immediate effect (P = 0.27) on the soil microbial activity at 15 weeks after the incorporation of the biofumigants (Figure 7.8). However, the soil microbial activity varied significantly (P = 0.01) over the three years of monitoring (Figure 7.9). All treatments demonstrated a seasonal pattern in microbial activity, where the soil microbial activity was higher in autumn and spring, compared to winter and summer. When compared over the three years of monitoring, there was no difference (P = 0.09) in the soil microbial activity at the two soil depths investigated.

The soil pH was not affected (P = 0.65) by the different biofumigant treatments. In contrast, the biofumigant treatments did affect (P = 0.01) the soil EC. The soil pH in all the treatments increased (P < 0.01) over the first 50 weeks after the biofumigants were incorporated (Figure 7.10). This was matched with an obvious decrease in soil EC, particularly in the soils amended with either Brassica species (Figure 7.11). After this time, the soil pH and EC leveled off and only fluctuated mildly for the remaining monitoring period. There was no difference between the soil pH (P = 0.49) and EC (P = 0.14) at the two soil depths investigated.

Neither biofumigant (B. juncea or B. napus) affected (P = 0.90) the number of deaths in the Leucadendron Safari Sunsets (Figure 7.12). There was no difference (P = 0.64) in the number of Leucadendron deaths from the plants that were inoculated, compared to the non-inoculated plants. Furthermore there was no apparent relationship of the lower biomass of the biofumigants in different replicate plots and the number of Leucadendron deaths.

Figure 7.8. The microbial activity (µg FDA) within the soil from the different treatment plots (A = soil depth 10 cm; B = soil depth 20 cm) 15 weeks after the biofumigants had been incorporated. Soil microbial activity is shown in relative units of flurescein diacetate (FDA). Standard error bars are shown.

(A) Soil depth 10 cm

600

Q 500

CD

^ 400

CD CD

E 300

o

^ 200

b o

o 100

0

Control

■ ■ ■ ■ Brassica juncea

Brassica napus

0 20 40 60 80 100 120 140 160

Week

(B) Soil depth 20 cm 800

700

600

CD CD CÜ

E o

500

400

s £2 O

O

300

200

100

0

0 20 40 60 80 100 120 140 160

Week

Figure 7.9. The microbial activity (µg FDA) within the soil from the different treatment plots (A= soil depth 10 cm; B = soil depth 20 cm) over the three years of monitoring. Soil microbial activity is shown in relative units of flurescein diacetate (FDA). Standard error bars are shown.

Control

■ ■ ■ ■ Brassica napus

Brassica juncea

0 20 40 60 80 100 120 140 160

Week

Figure 7.10. The soil pH within the treatment plots (A= soil depth 10 cm; B = soil depth 20 cm) over the three years of monitoring. Standard error bars are shown.

(A) Soil depth 10 cm

0 20 40 60 80 100 120 140 160

Week

Figure 7.11. The soil electrical conductivity (ms^-1) within the treatment plots (A = soil depth 10 cm; B = soil depth 20 cm) over the three years of monitoring. Standard error bars are shown.

Figure 7.12. The mean plant deaths from the different soil treatments conducted on the field trial (■ = Phytophthora cinnamomi inoculated; □ = non – inoculated). Standard error bars are shown.

• 1. Discussion

The current study demonstrated that B. juncea and B. napus mildly suppressed the survival and infective ability of P. cinnamomi. The suppressive nature of the biofumigants was similar to that described previously in Chapters 4 and 5. However, unlike the suppression described previously that resulted in reduced disease incidence in L. angustifolius (Chapter 5.3.3), neither Brassica species reduced the incidence of disease in Leucadendron Safari Sunset. Unfortunately, a very slow death rate in Leucadendron Safari Sunset did not generate any statistical differences in the deaths over the three years of monitoring. Further monitoring is required if the biofumigant treatments can be concluded to have no effect on the deaths caused by P. cinnamomi. Vawdrey et al. (2002) also reported difficulty in converting the suppression of Phytophthora from the glasshouse to the field. In addition, Hardy and Sivasithamparam (1991c) have reported that suppression of sporangial production in P. cinnamomi did not reduce infection of Telopea speciossima. Shetty et al. (2000) reported that Brassica species could reduce disease incidence by Verticillium. However, they showed that even in the absence of wilt symptoms the pathogen was still colonizing the host's roots. Therefore, it is possible that biofumigants may not reduce root infection by

P. cinnamomi, rather the biofumigants could be reducing the viability of the infective propagules (Chapter 4) used to colonise the host and thereby reduce the incidence of disease.

Both biofumigants caused a gradual reduction on the survival and infective ability of the pathogen. The suppressive ability of B. juncea was far greater when applied with an organic amendment (Chapter 6.3.2), compared to the suppression achieved in the current study. The two biofumigants differed in the types and amounts of ITCs liberated from the decomposing Brassica tissues. Furthermore, B. napus differed from B. juncea in that significant levels of ITC were released during growth of the biofumigant crop. Gardiner et al. (1999) also reported that significant levels of glucosinolate hydrolysis products were released into the soil during the growth of B. napus and theorised that these products could be contributing to the biofumigants effect. Moreover, Vawdrey et al. (2002) reported that some reduction in Phytophthora inoculum density could be achieved by the release of ITCs during the growth of the B. napus.

The ITCs detected in soil from the incorporation of B. napus and B. juncea is consistent with the previously described ITC profiles (Chapters 3.3.1, 4.3.1) and in the studies of Brown and Morra (1997), Bending and Lincoln (1999), and Gardiner et al. (1999). Unlike earlier studies described in this dissertation there was no difference in the suppression achieved by the two Brassica species, B. juncea and B. napus. ITCs released from both biofumigants peaked within 1 and 24 hrs and were no longer detected in soil from 72 and 168 hrs after incorporation. This is comparable to the findings of Gardiner et al. (1999) who reported that the maximum concentration of ITCs (~1 nmole/g soil) from B. napus was reached within 30 hrs from incorporation. However, their study found that ITCs were still detected in soil 20 days after incorporation. It is unclear why the retention of the ITCs was significantly lower in the current study. It is possible that differences in the watering regime of the treatments areas, the method of incorporation, soil temperature, age of the plant, soil type or other factors may have contributed to the reduced ITC retention time.

The poor suppression of P. cinnamomi achieved in the current study is likely to be related to the poor quality of the biofumigant crops. For example, the mean biomass (69 g dry weight/m²) for the B. juncea treatment plots in this field trial was substantially lower than the field trial (230 g dry weight/m²) described in Chapter 6 and previously published biomass levels achieved in the field (Kirkegaard and Sarwar 1998). For the biofumigant effect to be maximised it is essential that the optimal Brassica crop is grown. It was interesting to note that there was no apparent relationship of the lower biomass of the biofumigants in different replicate plots and the number of Leucadendron deaths.

Factors such as: plant age; soil nutrition; soil type; moisture; plant spacing; and day length affect the success of the biofumigant crop (Brown and Morra 1997, Leonard 2000). These factors determine the quality of the resulting Brassica crop and the efficiency of ITC production during incorporation. It is recommended that the B. juncea be incorporated when ~50% of the plants are flowering. At this stage in development the glucosinolates levels in the plants are highest and therefore will release the highest concentrations of glucosinolate hydrolysis products into the soil. During the current study, the B. napus were incorporated at the time >50% of the plants were flowering (Kirkegaard and Sarwar 1998). However, Kirkegaard and Sarwar (1998) reported that B. napus might require up to 50 more days to reach mid-flowering, than B. juncea. Future experiments need to allow for B. napus to reach maturity if the biofumigant effect it to be maximised.

Bending and Lincoln (1999) reported that low ITC concentrations in B. juncea amended soils were often related to the higher soil microbial respiration. They hypothesised that the higher microbial respiration resulted in rapid degradation of the glucosinolate hydrolysis products. In the current study the incorporation of the biofumigants was also related to an increase in soil microbial activity. Recent investigations have identified that the continual use of fumigants can result in decreased effectiveness due to a phenomenon called ‘enhanced biodegradation' (Matthiessen pers. comm.). Enhanced

biodegradation refers to the situation in which continual use of the same fumigant results in a shift in the soil micro-flora to populations that can break down the fumigant so rapidly that the fumigant is not available to destroy the target organism (Warton et al. 2001b). The increase in the microbial activity observed in the biofumigant treatment plots during the current study may be evidence of a biofumigant-degrading soil micro­flora and therefore this soil type may be sensitive to the enhanced biodegradation. Enhanced biodegradation and its relevance to protea cultivation are discussed later (Chapter 8.4.4).

As in the field trial described in Chapter 6.4, there was an immediate increase in soil pH and decrease in soil EC after the treatments had been conducted. It is unclear what is the exact cause of this shift in the soil pH and EC, however, it is likely to be partly caused by the physical disturbance of the soil using the rotary-hoe prior to treatments being conducted. It is unclear as to whether this shift has any impact on the diseases caused by P. cinnamomi. Additionally, it has been shown that the formation and disappearance of the glucosinolate hydrolysis products is dependant on the soil pH. Therefore, it is likely this shift in soil pH could have a significant effect on efficiency of the biofumigant process.

The current study indicated that B. juncea and B. napus as biofumigants were only mildly suppressive to P. cinnamomi. This suppression did not equate to a reduction in disease incidence in the field. It is possible that with further monitoring, statistical differences in the number of Leucadendron deaths from the different treatment plots may develop. Poor quality of the Brassica crops may have contributed to the reduced suppressive nature of the biofumigants. The mechanisms of suppression by both Brassica species and issues concerning their agronomy will be discussed later (Chapter 8.4).

Chapter 8 – General discussion

• 1. Plantation visits

This study clearly showed that P. cinnamomi is a devastating and widespread pathogen in protea plantations throughout the south-west of Western Australia (WA) and is the primary cause of ‘Sudden Death' of proteas. P. cinnamomi was isolated from dying plants at 39% (11 out of 28) of the plantations visited. These results are comparable to those of Boersma et al. (2000) who recovered P. cinnamomi from 34% (12 out of the 35) protea plantations visited in 1997/98. Infection by the pathogen totally decimated some plantations, resulting in large areas within plantations being abandoned. It is likely the pathogen is often introduced into disease-free plantations by infested soil or infected nursery stock (Hardy and Sivasithamparam 1988, McLennan 1993). Alternatively, it is possible that new infections may have arisen from the movement of soil, plant material, soil water or stream water from surrounding properties and native plant communities.

The current study also identified a range of other potentially significant fungal plant pathogens (Fusarium, Botryosphaeria, Pestalotiopsis, Alternaria) and pests (nematodes, mealy bug, scale) associated with protea death and decline in WA plantations. It is unclear if these are primary pathogens or secondary pathogens being driven by other problems, such as nutrient imbalances. For example, correcting nutritional imbalances and the application of common fungicides could control many of the leaf diseases described. Many of these pathogens, including Fusarium (Swart et al. 1999) and Botryosphaeria (Crous et al. 2000) have only been described recently as significant pathogens of cultivated protea worldwide. The distribution and impact of these pathogens in Australian plantations are still unclear. Crous et al. (2000) hypothesized that many of these pathogens were introduced into Australia through contaminated plant material and may constitute a threat to our native Proteaceae. Further research is needed, particularly into the origin and host specificity of these pathogens and what control measures can be used to reduce their impact. Furthermore, molecular analysis

would be useful in determining if these pathogens are endemic or if they have been introduced.

• 1. Effect of biofumigants on Phytophthora cinnamomi

The current study comprehensively demonstrated that biofumigants (B. juncea and B. napus) suppress P. cinnamomi in vitro and in vivo. By conducting a series of investigative experiments the biofumigants were found to suppress the growth, sporulation, survival and infective ability of the pathogen (Table 8.1). In the first experiment, the suppressive nature of B. juncea and B. napus was investigated in vitro on the vegetative growth of five Phytophthora species (P. cinnamomi, P. cactorum, P. citricola, P. cryptogea and P. megasperma). The volatiles released from B. juncea were highly suppressive to all the five Phytophthora species. In contrast, the volatiles released from B. napus did not suppress the growth of any of the five Phytophthora species and in some cases they increased vegetative growth. For both Brassica species, the shoot tissues were more suppressive to all five Phytophthora species than the root tissues. The study also identified reasonable inter-specific variation in the response of the different Phytophthora species with P. cinnamomi and P. cactorum considerably more sensitive than the other three species. Smith and Kirkegaard (2002) reported significant variation in vitro in the response of Phytophthora species and other Oomycetes to isothiocyanates (ITCs). When 100% suppression of the Phytophthora species growth was achieved, the effect was often fungicidal.

In the second experiment, the effect of soil amended with B. juncea and B. napus on P. cinnamomi sporangia and chlamydospores production was investigated in vitro. P. cinnamomi colonised Miracloth discs were added to soil amended with the two Brassica species, before being removed every two days for the determination of sporangia and chlamydospore production and infective ability. Only the soils amended with B. juncea significantly reduced sporangia production in P. cinnamomi. This is the first record that biofumigants suppress sporangia formation in P. cinnamomi. However, both Brassica species increased the percentage of aborted or immature sporangia and reduced the infective ability of the pathogen. Neither Brassica species had any effect on zoospore release or chlamydospore production in P. cinnamomi.

The final investigative experiment examined the effect of amending soils with the two Brassica species on the inoculum potential, infective ability and disease incidence caused by the pathogen under glasshouse conditions. Both Brassica species had an immediate suppressive effect on the inoculum potential and infective ability of P. cinnamomi. However, after this initial suppression there was a gradual increase in the recovery of the pathogen over the four week monitoring period. Only the soils amended with B. juncea reduced the incidence of root rot in Lupinus angustifolius. This is the first record that biofumigants can reduce disease incidence caused by P. cinnamomi. B. napus had no effect on disease incidence in L. angustifolius. Utkhede and Hogue (1999) also reported that B. napus had no effect of disease incidence of Phytophthora in apple.

• 1. Field trials

The investigative experiments demonstrated the suppressive nature of biofumigants on P. cinnamomi. Two field trials were conducted to determine if this suppressive nature could also reduce disease incidence under field conditions. The first field trial compared biofumigation (B. juncea) to chemical fumigation (metham sodium) and soil solarisation. The three soil treatments were used in an integrated management approach to control P. cinnamomi that included the use of a hardwood compost, mulch and water sterilisation. All three-soil treatments significantly reduced the recovery of the pathogen and its infective ability to a soil depth of 20 cm. Metham sodium and soil solarisation were the most and least suppressive treatment, respectively. Only the metham sodium treatment significantly reduced the incidence of root rot in Leucadendron Safari Sunset over the three years of monitoring. The superior suppressive ability of metham sodium compared to biofumigation using B. juncea and B. napus has been previously reported by Stephens et al. (1999).

Table 8.1. A comparison of the suppressive nature of the two Brassica species used as biofumigants from the investigative experiments conducted in the current study.

The second field trial compared the effectiveness of B. juncea and B. napus, without the use of other control strategies, to reduce the incidence of P. cinnamomi infection of Leucadendron Safari Sunset. Although both Brassica species reduced the recovery and infective ability of the pathogen, neither biofumigant reduced the incidence of root rot in Leucadendron Safari Sunset. It is unclear if the lack of disease suppression in the Leucadendron hybrid is due to the environmental conditions during the trial being non- conducive to the pathogen (Chapter 6.4). Alternatively, the limited suppressive ability of the two Brassica species could be related to the poor quality of the biofumigant crops

(Chapter 7.4). It is recommended that future studies investigate the effectiveness of biofumigation in the field. It is vital that these studies quantify the Brassica plants prior to incorporation and ensure the best biofumigant crop is grown.

• 1. Biofumigation in protea cultivation
     1. Mechanisms of suppression

1. juncea was superior to B. napus in its ability to suppress P. cinnamomi. The poor suppressive ability of B. napus to Phytophthora species has been previously reported by Vawdrey et al. (2002) and Utkhede and Hogue (1999). Regardless, the ‘biofumigant' effect can be largely attributed to a chemical suppression due to the release of toxic chemicals during the decomposition of the Brassica tissues. These chemicals are likely to include glucosinolate and non-glucosinolate hydrolysis products (Bending and Lincoln 1999, Mazzola et al. 2001).

Smith and Kirkegaard (2002) found that there was significant variation in the response of different bacteria and fungi to phenylethyl-ITC. In the current study, biofumigation resulted in an increase in the soil microbial activity in the two field trials (Chapter 6 & 7). It is possible that biofumigants may increase bacteria and fungi that could be suppressive, if only fungistatically, to P. cinnamomi. Mazzola et al. (2001) reported that populations of fluorescent Pseudomonas were enhanced when soils were amended with biofumigant tissues. Moreover, Smith and Kirkegaard (2002) reported that Trichoderma species used as a biological control agent were tolerant to phenylethyl­ITC. These studies contradict those of Thompson et al. (2001) who reported biofumigants can decrease vesicular arbuscular mycorrhiza. Arbuscular mycorrhizal fungi have been reported to be important in the biocontrol Phytophthora parasitica Dastur (Vigo et al. 2000). Further experiments are needed to fully elucidate the effect of biofumigants on the soil micro-flora, endophytes and mycorrhizal fungi.

• • 1. Optimising biofumigation

For the biofumigant effect to be maximised it is essential that the optimal Brassica crop is grown. Factors such as genetics of the Brassica species/variety used, timing of

planting, sowing density, soil nutrition, soil type and day length affect the success of the biofumigant crop (Leonard 2000). These factors determine the quality of the resulting Brassica crop and the efficiency of ITC production during incorporation. The current study demonstrated that the two Brassica species varied significantly in their ability to suppress P. cinnamomi. Further experiments are needed to screen a greater number of Brassica species/varieties to determine those that are most appropriate for the control P. cinnamomi in WA soils. Furthermore, growers need to ensure the Brassica's are incorporated into the soil at the time of maximum glucosinolate concentration and that soil moisture is optimal to ensure maximum ITC retention and efficacy (Morra and Kirkegaard 2002).

• • 1. Incorporation methods

A recent study (Morra and Kirkegaard 2002) has shown that the amount of tissue disruption has marked effects on the concentration of ITCs released from B. juncea and B. napus tissues. Only about 1% of available ITCs are released when tissue is roughly chopped with a rotary hoe. Their study found tissue disruption at the cellular level afforded by freezing and thawing the tissues, resulted in ITC release efficiencies of 14% (B. juncea) and 26% (B. napus). Morra and Kirkegaard (2002) concluded that the method of incorporation was the most significant factor in maximizing glucosinolate hydrolysis and ITC release, compared to the Brassica variety or soil moisture. Future experiments are recommended that try different incorporation methods in the field (e.g. mulching) and to determine how much of the ITCs may have been lost during freeze drying.

• • 1. Enhanced biodegradation

Recent investigations have shown that the continual use of fumigants (e.g. metham sodium) can result in decreased effectiveness of the fumigant due to a phenomenon termed enhanced biodegradation. Enhanced biodegradation refers to the continual use of the same fumigant which results in a shift in the soil-microflora to populations that can break down the fumigant compounds so rapidly that they are not available to destroy the target organisms (Chapman and Harris 1990). Warton et al. (2003) reported

enhanced biodegradation of MITC on a site in the southwest of WA with a history of continuous metham sodium use. As the active ingredient in metham sodium is related to the chemicals formed from decomposing Brassica plants, it is likely that biofumigation could be susceptible to the enhanced biodegradation effect when incorporated into soil that has been previously fumigated with metham sodium and/or biofumigants. However, since protea growers would only be using biofumigation as a pre-plating tool approximately every 10 years, biodegradation is less likely to become an issue in protea cultivation. Growers could try to prevent enhanced biodegradation by alternating different Brassica species and varieties.

• 1. Integrated management of Phytophthora cinnamomi

Integrated management (IM) holds the greatest promise in the control of diseases caused by soil-borne pathogens such as P. cinnamomi (Sivasithamparam 1991). Currently, there is no single control treatment for P. cinnamomi within diseased protea plantations. P. cinnamomi is a difficult pathogen to control as it can: invade the soil profile to considerable depths; spread rapidly in poorly drained or waterlogged soils; and form chlamydospores that can persist in harsh environmental conditions (Chapter 1.4.4). Vawdrey et al. (2002) demonstrated that an IM approach to controlling P. palmivora could reduce disease severity in a pawpaw plantation from Queensland. They reported the combination of organic soil amendments, inorganic soil amendments and biological antagonists could reduce the impact of the pathogen. Dixon et al. (1990) successfully tested a similar approach using organic mulch, the fungicide (prothiocarb) and herbicide to reduce P. cinnamomi root rot of Banksia. However, their experiment was conducted in a glasshouse and it was unclear if this would equate to disease control in the field. An IM approach to controlling P. cinnamomi would have the added advantage of controlling other soil-borne pathogens and pests.

Once P. cinnamomi is introduced to disease-free areas it is almost impossible to eradicate (Ribeiro 1978). Therefore, growers with disease-free plantations need to ensure that the pathogen is not introduced and growers that have the pathogen need to ensure the pathogen is contained and not spread further around their plantation. Growers can reduce the spread of the pathogen around the plantation by correcting

drainage problems. Growers should be aware of high risk activities (e.g. movement of infested soil and) that can spread rapidly spread the pathogen around the plantation. If these high-risk activities are essential it is recommended that they be conducted during the hot, dry summer months when the chance of spreading the pathogen is minimal (Hardy et al. 2001). When viewing perspective land for a protea plantation in WA it is important to employ qualified plant pathologists that can determine if P. cinnamomi is present. If the pathogen is found, it is essential that risk analysis is conducted and the area infected by the pathogen is mapped. Hardy and Sivasithamparam (1988) identified WA production nurseries as a likely source of new Phytophthora infections in WA plantations. They reported that all 14 nurseries surveyed in WA were infested with at least one Phytophthora species. Therefore, any new plant material brought onto a grower's property should always be screened prior to planting in the plantation (Sivasithamparam and Goss 1980).

Prior to planting or re-planting on a P. cinnamomi infested site, it is important to remove all infected plant material that may harbor the pathogen. It is recommended that high-impact P. cinnamomi sites are left fallow or a resistant rotation crop is planted to allow for the inoculum potential of the pathogen to drop. Vawdrey (2001) showed repeated plantings of Papaw crops increased the population density of Phytophthora palmivora within the soil. They demonstrated that rotation crops or fallows reduced inoculum levels and the incidence of root rot.

P. cinnamomi can multiply in open water supplies, such as dams, streams and open tanks (McLennan 1993). It is therefore essential to sterilise any water that is to be used for irrigation. During the plantation visits the pathogen was found in a number of dams used for irrigation. In these plantations the P. cinnamomi infested irrigation water was identified as the likely mode of spread of the pathogen around the plantation. In the current study, a Venturi chlorine injection system was used to treat the dam water used to irrigate the trial (Chapter 6, Figure 6.4). The system was cheap and relatively easy to use. The injection system had the added bonus of being used to distribute liquid fertilisers.

Selection and breeding for resistance has been used to control Phytophthora diseases in cultivated proteas (Sedgley et al. 1997) and other horticultural crops, including avocadoes and Chamelaucium (Irwin et al. 1995, O'Sullivan et al. 1999). Swart and Denman (2000) demonstrated that the combined use of resistant plants and fungicides could completely suppress symptom development caused by P. cinnamomi. Wilcox et al. (1999) tested various proposed components of an IM plan for P. fragariae and concluded the resistance of the cultivar was the most important factor in reducing disease incidence and severity. Over 70% of the proteas grown in WA plantations are susceptible Leucadendron and Leucospermum varieties (AgWA 2001). Growers that have P. cinnamomi within their plantation should consider growing a greater percentage of resistant genera (e.g. Protea) or ensuring the Leucadendron or Leucospermum varieties grown are resistant. McCredie et al. (1985) reported that survival rates of 95% could be achieved after P. cinnamomi infection by using resistant Banksia species. Previous studies can be consulted to identify resistant varieties (Cho 1981, Cho 1983, McCredie et al. 1985, Von Broembsen and Brits 1985, Tynan et al. 1995, Denman et al. 2000, Denman and Sadie 2001, Nakao et al. 2003). Resistant varieties can be used preferentially for breeding new varieties or as a source for rootstock development (McCredie et al. 1985, Denman and Sadie 2001). When choosing what resistant varieties to grow it is important to consider the profitability of each variety, as some are more profitable than others (Skipworth and Bus 1991).

The use of composts and mulch is under utilised in the WA protea industry. In the current study, combining the use of a Eucalyptus hardwood compost/mulch with other soil treatments resulted in suppression of the pathogen. This suppressive effect of hardwood composts has been previously shown to be due to a combination of chemical and biological factors (Hoitink et al. 1977, Hardy and Sivasithamparam 1988, 1991b). Hardy and Sivasithamparam (1991b) illustrated that this suppression was more fungistatic that fungicidal. Therefore, to ensure continual suppression of the pathogen it is important that the mulch/compost is applied regularly (i.e. annually). Although not observed in the current study (Chapter 6), composts have been shown to reduce disease incidence of P. cinnamomi in cultivated Proteaceae (Telopea speciossima and Banksia occidentalis R.Br.) (Hardy and Sivasithamparam 1991c).

Although P. cinnamomi penetrates the soil profile to considerable depths (> 2 m) pre­planting techniques can reduce the inoculum density of the pathogen within the root zone area of the host plant (Kinal et al. 1993). The current study compared the efficacy of three soil treatments (biofumigation, fumigation and soil solarisation) when applied in the field (Chapter 6). Although all treatments had some suppressive effect on the inoculum potential and infective ability of the pathogen, only metham sodium actually resulted in a reduced disease incidence in Leucadendron Safari Sunset. Future studies should investigate the effect of combining two of these treatments. For example, solarisation combined with metham sodium or biofumigation has been shown to increase the effectiveness of the treatments in controlling soilborne pathogens such as Phytophthora species (Duniway 2002). By combining treatments, potential weaknesses of the individual treatments can be overcome. The current study demonstrated that the two Brassica species had no effect on chlamydospore production in P. cinnamomi (Chapter 4) that could allow for the re-activation of infection once the biofumigants had decomposed. Benson (1978) demonstrated that solarisation could be effective in inactivating chlamydospores. Therefore, the combination of solarisation and biofumigation could lead to increased control of the pathogen. Coelho et al. (1997, 2000, 2001) reported increased suppression of P. nicotianae by combining solarisation and biofumigation. They demonstrated that by adding biofumigants to the soil to be solarised, the time taken to inactivate chlamydospores of P. nicotianae was reduced.

A range of fungicides has been used to control P. cinnamomi in horticultural settings (Pegg et al. 1987, Marks and Smith 1990). Phosphite is arguably the most effective fungicide for widespread use against the pathogen. Turnbull et al. (1995) demonstrated that 3-6 weekly spraying of phosphite at 1.2% phosphite when disease pressure is high (spring, autumn) could reduce the impact of P. cinnamomi in proteas. Ali et al. (2000) reported that the use of phosphite and the systemic resistance inducing chemical, Bion, could achieve greater control. Protection by trunk infection of phosphite has been reported to last as long as 2-5 years in native Banksia populations (Shearer and Fairman 1997). Trunk injection of phosphite has also been widely used in the avocado industry

(Leonardi et al. 1999). It is unclear if trunk infection of phosphite could be used in cultivated proteas due to the limited trunk size.

Other fungicides, for example metalaxyl, may have a role as a soil drench in the eradication of the pathogen in spot infections (Cho 1981, Pegg et al. 1987, Marks and Smith 1990, 1992, Hill et al. 1995).

• 1. Limitations of the current study

The following limitations of the current study were identified:

1. Due to time constraints it was not possible to investigate the full range of leaf, stem and root diseases within WA plantations.
2. The two field trials were conducted in years with well below average rainfall (Chapter 6.4). P. cinnamomi is a water mould and requires moist warm conditions to sporulate and cause disease. Therefore, in the current study the environmental conditions were not optimal for disease development. It is likely that this may have contributed to the poor suppressive ability of the different soil treatments in the field.
3. There were a number of agronomy issues in the field trials that lead to the reduced quality of the Brassica crops. For example, both Brassica species were grown for an equal period of time before their incorporation. Kirkegaard and Sarwar (1998) reported that glucosinolate concentrations in Brassica plants reached their maximum when 50% of the plants were flowering (Chapter 7.4). However, only B. juncea was in flower at the time of incorporation. Therefore, it is likely that suppression achieved by B. napus was only a fraction of its potential. Other agronomy issues that may have led to a reduced quality of the biofumigant crops included: the timing of sowing; contamination of the Brassica seeds and; the poor soil nutrition.
4. The soil solarisation was conducted in the summer prior to the other two soil treatments being conducted. As such the ground was left fallow for 8 months. It is likely this would have reduced the suppressive ability of the solarisation treatment (Chapter 6.4). However, given the long fallow period, the solarisation treatment still reduced the level of P. cinnamomi inoculum in the soil.
   1. Future research directions

Future research should involve the following:

1. Investigate the extent that other pathogens and pests, particularly Fusarium and Botryosphaeria, are contributing to plant death and decline in WA protea plantations.
2. Determine whether other pathogens pose a risk to native plant communities in WA. Conduct molecular analysis on the common plant pathogens to determine if they are introduced or endemic.
3. Develop management techniques for these other important pathogen and pests.
4. Further elucidate the mechanisms of how the two biofumigants suppress P. cinnamomi and account for the difference of the two Brassica species. This could be achieved by purifying ITCs or other glucosinolate derived chemicals, for the use in screening against P. cinnamomi isolates.
5. Screen more Brassica species, varieties and cultivars for their suppressive effect on P. cinnamomi to identify the best Brassica for use in controlling P. cinnamomi.
6. Screen more isolates of P. cinnamomi and other Phytophthora species to describe the intra- specific variation in P. cinnamomi and inter-specific variation in Phytophthora species.
7. The methodologies described in the current dissertation contain novel techniques that could be used to determine how fungal plant pathogens respond to fumigants and organic amendments. For example, determining if the suppressive nature of Acacia pulcella is chemical or biological (Malajczuk et al. 1983).
8. Continue monitoring of the existing field trials. For the trials to be considered a success the plants have to survive their productive years (approx. year 3-5). It is possible by year five statistical differences in the plant deaths from the different treatments may become evident.
9. Conduct a field trial that combines solarisation with biofumigation or metham sodium. Attention should be made to growing a good biofumigant crop and

changing the incorporation method to one that more effectively breaks up the biofumigant tissues (Morra and Kirkegaard 2002).

1. Conduct a field trial that combines all the suggested treatments in an IM approach. This trial should include the use of resistant and susceptible varieties and spraying with phosphite.
   1. Conclusions

P. cinnamomi is the most common and devastating pathogen in WA protea plantations. The current study demonstrated that P. cinnamomi is sensitive to the suppressive nature of biofumigants. Biofumigants can suppress the growth, sporulation, infective ability of P. cinnamomi and reduce the incidence of the disease caused by the pathogen. Of the two Brassica species investigated, B. juncea was superior in its ability to control P. cinnamomi compared to B. napus. When applied in the field, biofumigation using B. juncea was found to be more suppressive that soil solarisation, but not as effective as metham sodium. Neither Brassica species resulted in reduced incidence of root rot of Leucadendron Safari Sunset when the biofumigants were used in a P. cinnamomi infested protea plantation. Biofumigation has the potential to reduce the impact of the pathogen as a component in an integrated management approach to control P. cinnamomi. The findings described in this dissertation have the potential to reduce the impact of P. cinnamomi in cultivated proteas, other floriculture industries (e.g. Boronia and Chamelaucium), nurseries and horticultural industries where P. cinnamomi is a common pathogen.

References

AgWA 2001. Australian native and South African species cultivated for the export cutflower industry of Western Australia – planting and production results of the 1999 survey. Department of Agriculture. Miscellaneous Publication 28/2001, ISSN 1326-4168.

Ali, Z., Smith, I. and Guest, D.I. 2000. Combinations of potassium phosphonate and Bion (acibenzolar-S-methyl) reduce root infection and dieback of Pinus radiata, Banksia integrifolia and Isopogon cuneatus caused by Phytophthora cinnamomi. Australasian Plant Pathology. 29: 59-63.

Anderson, R.D. and Guest, D.I. 1990. The control of black pod, canker and seedling blight of cocoa, caused by Phytophthora palmivora, with potassium phosphonate. Australasian Plant Pathology. 19: 127-129.

Angus, J.F., Gardner, P.A., Kirkegaard, J.A. and Desmarchelier, J.M. 1994. Biofumigation: isothiocyanates released from Brassica roots inhibit growth of the take- all fungus. Plant and Soil. 162: 107-112.

Barbercheck, M.E. and Von Broembsen, S.L. 1986. Effects of soil solarisation on plant parasitic nematodes and Phytophthora cinnamomi in South Africa. Plant Disease. 70: 945-950.

Barlow, S. and Haigh, A.M. 1997. A guide to the nutritional needs of some Proteaceae. RIRDC Research Paper Series. No 97/38.

Barnett, H.L. and Hunter, B.B. 1998. Illustrated genera of the imperfect fungi. 4^th Ed. APS Press, St Paul, Minnesota, USA.

Barrett, S.R., Shearer B.L. and Hardy, G.E.StJ. 2003. The efficacy of phosphite applied after inoculation on the colonisation of Banksia brownii stems by Phytophthora cinnamomi. Australasian Plant Pathology. 32: 1-7.

Bartnicki-Garcia, S. and Wang, M.C. 1983. Biochemical aspects of morphogenesis in Phytophthora. Phytophthora. Its biology, taxonomy, ecology and pathology (Eds. Erwin, D.C., Bartnicki-Garcia, S. and Tsao, P.H.). The American Phytopathological Society, St Paul, Minnesota, USA. pp 121-137.

Barrett, S.R., Shearer, B.L. and Hardy, G.E.StJ. 2003. The efficacy of phosphite applied after inoculation on the colonisation of Banksia brownii stems by Phytophthora cinnamomi. Australasian Plant Pathology. 32: 1-7.

Bending, G.D. and Lincoln, S.D. 1999. Characteristics of volatile sulphur- containing compounds produced during decomposition of Brassica juncea. Soil Biology and Biochemistry. 31: 695-703.

Benic, L.M. and Knox-Davies, P.S. 1983. Anthracnose of Protea compacta, caused by Colletotrichum gloeosporioides. Phytophylactica. 15: 109-119.

Benson, 1978. Thermal inactivation of Phytophthora cinnamomi for control of Fraser Fir root rot. Phytopathology. 68: 1373-1376.

Blaker, N.S. and MacDonald, J.D. 1983. Influence of container medium pH on sporangium formation, zoospore release and infection of Rhododendron by Phytophthora cinnamomi. Plant Disease 67: 259-263.

Blok, W.J., Lamers, J.G., Termorshuizen, A.J and Bollen, G.J. 2000. Control of soilborne plant pathogens by incorporating fresh organic amendments followed by tarping. Phytopathology. 90: 253-259.

Blowes W.M. 1980. A comparison of the occurrence, sporulation and survival of Phytophthora cinnamomi Rands in soils supporting native forests in south-eastern New South Wales and South-Western Australia. PhD Thesis, Australian National University, Australian Capital Territory, Australia.

Boersma, J.G., Cooke, D.E.L. and Sivasithamparam, K. 2000. A survey of wildflower farms in the south-west of Western Australia for Phytophthora spp. associated with root-rots. Australian Journal of Experimental Agriculture. 40: 1011­1019.

Bold, H.C., Alexopoulos, C.J. and Delevoryas, T. 1987. Morphology of plants and fungi. 4^th Ed. Harper and Row, New York.

Bones, A.M. and Rossiter, J.T. 1996. The myrosinase-glucosinolate system: organisation and biochemistry. Plant Physiology. 97: 184-208.

Broadbent, P. and Baker, K.F. 1974. Behaviour of Phytophthora cinnamomi in soils suppressive and conducive to root rot. Australian Journal of Agricultural Research. 25: 121-137.

Brown, P.D., Morra, M.J., McCaffrey, J.P. and Morra, M.J. 1991. Alleochemicals produced daring glucosinolate degradation in soil. Journal of Chemical Ecology. 17: 2021-2034.

Brown, P.D. and Morra, M.J. 1997. Control of soil-borne plant pests using glucosinolate-containing plants. Advances in Agronomy. 61: 167-231.

Cahill, D. 1994. Review of Phytophthora diseases in Australia. Rural Industries Research Development Corporation, Research Paper Series, No 93/4.

Cahill, D.M., Legge, N., Grant, B. and Weste, G. 1989. Cellular and histological changes induced by Phytophthora cinnamomi in a group of plant species ranging from fully susceptible to fully resistant. Phytopathology. 79: 417-424.

Carmichael, J.W., Kendrich, W.B., Conners, I.L. and Sigler, L. 1980. Genera of the Hyphomycetes. University of Alberta Press, Edmonton, Canada.

Chapman, R.A. and Harris, C.R. 1990. Enhanced degradation of insecticides in soil. Enhanced biodegradation of pesticides in the environment (Eds. Racke, K.D. and Coats, J.R.). American Chemical Society, Washington, DC. pp 82-96.

Chan, M.K.Y. and Close, R.C. 1987. Aphanomyces root rot of peas. Control by the use of cruciferous amendments. New Zealand Journal of Agricultural Research. 30: 225-233.

Charron, C.S. and Sams, C.E. 1999. Inhibition of Pythium ultimum and Rhizoctonia solani by shredded leaves of Brassica species. Journal of the American Society of Horticultural Science. 124: 462-467.

Cho, J.J. 1981. Phytophthora root rot of Banksia: Host range and chemical control. Plant Disease. 65: 830-833.

Cho, J.J. 1983. Variability in susceptibility of some Banksia species to Phytophthora cinnamomi and their distribution in Australia. Plant Disease. 67: 869­871.

Coelho, L. 1997. Reduction of populations of Phytophthora spp. with soil solarisation under field conditions and thermal inactivation of Phytophthora nicotianae. PhD Thesis, University of Florida. pp 136.

Coelho, L., Chellemi, D.O. and Mitchell, D.J. 1999. Efficacy of solarisation and cabbage amendment of the control of Phytophthora spp. in North Florida. Plant Disease. 83: 293-299.

Coelho, L., Mitchell, D.J., and Chellemi, D.O. 2000. Thermal inactivation of Phytophthora nicotianae. Phytopathology. 90: 1089-1097.

Coelho, L., Mitchell, D.J. and Chellemi, D.O. 2001. The effect of soil moisture and cabbage amendment on the thermoinactivation of Phytophthora nicotianae. European Journal of Plant Pathology. 107: 883-894.

Cohen, Y. and Coffey, M.D. 1986. Systemic fungicides and the control of Oomycetes. Annual Review of Phytopathology. 24: 311-338.

Collin, S., Shearer, B., McComb, J., Colquhoun, I. and Hardy, G. 2001. Long term survival of Phytophthora cinnamomi in Banksia grandis tissue buried in jarrah forest and rehabilitated bauxite mine soils. Proceedings of the 13^th Biennial Plant Pathology Conference, Cairns, Australia. pp 290.

Colquhoun, I. and Hardy, G.E.St.J. 2000. Managing the risks of Phytophthora root and collar rot during bauxite mining in the Eucalyptus marginata (Jarrah) forest of Western Australia. Plant-Disease. 84: 116-127.

Cook, R.J. and Baker, K.F. 1983. The nature and practice of biological control of plant pathogens. The American Phytopathological Society, St Paul, Minnesota.

Crous, P.W., Summerell, B.A., Taylor, J.E. and Bullock, S. 2000. Fungi occurring in Australia: selected foliicolous species. Australasian Plant Pathology 29: 267-278.

de Boer, R.F. and Greenhalgh, F.C. 1990. Efficacy of potassium phosphonate in controlling Phytophthora root rot of subterranean clover and ornamental plants in Victoria. Australasian Plant Pathology. 19: 124-125.

Denman, S., Crous, P.W. and Wingfield, M.J. 1999. A taxonomic reassessment of Phyllachora proteae, a leaf pathogen of Proteaceae. Mycologia. 91: 510-516.

Denman, S., Sadie, A. and Criley, R.A. 2000. Evaluation of a stem inoculation technique for assessing resistance to Phytophthora cinnamomi in Leucospermum cultivars. Acta-Horticulturae. 545: 281-285.

Denman, S. and Sadie, A. 2001. Evaluation of a stem inoculation technique for assessing resistance to Phytophthora cinnamomi in Leucospermum cultivars. Australasian Plant Pathology. 30: 11-16.

Denman, S., Crous, P.W., Groenewald, J.Z., Slippers, B., Wingfield, B.D. and Wingfield, M.J. 2003. Circumscription of Botryosphaeria species associated with Proteaceae based on morphology and DNA sequence data. Mycologia. 95: 294-307.

Dennis, R.W.G. 1978. British Ascomycetes. J. Cramer, Lehre.

Dixon, K.W., Thinlay, Sivasithamparam, K. 1984. Technique for rapid assessment of tolerance of Banksia spp. to root rot caused by Phytophthora cinnamomi. Plant Disease. 68: 1077-1080.

Dixon, K.W., Frost, K. and Sivasithamparam, K. 1990. The effect of amendment of soil with organic matter, a herbicide and a fungicide on the mortality of seedlings of two species of Banksia inoculated with Phytophthora cinnamomi. Acta- Horticulturae. 264: 123-131.

Dolan, T.E. and Coffey, M.D. 1988. Correlative in vitro and in vivo behaviour of mutant strains of Phytophthora palmivora expressing different resistances to phosphorous acid and Fosetyl-Na. Phytopathology. 78: 974-978.

Domsch, K.H., Gams, W. and Anderson, T.H. 1993. Compendium of soil fungi. Volume 1. IHW-Verlag.

D'Souza, N.K. 2001. Biological control of Phytophthora cinnamomi: the potential of Western Australian native legumes to protect susceptible plant species. Honours Thesis. Murdoch University, Western Australia.

Duncan, L.W. 1998. How to Tissue Sample for Phytophthora in Proteaceae – Sudden Death in Proteaceae Workshops. Department of Agriculture, Western Australia.

Duniway, J.M. 1975. Formation of sporangia by Phytophthora drechsleri in soil at high matric potentials. Canadian Journal of Botany. 53: 1270-1275.

Duniway, J.M. 2002. Status of chemical alternatives to methyl-bromide for pre­plant fumigation of soil. Phytopathology. 92: 1337-1343.

Dunne, C.P., Dell, B. and Hardy, G.E.StJ. 2003. The effect of biofumigants on the vegetative growth of five Phytophthora species in vitro. Acta Horticulture. 602: 45­52.

Eberlein, C.V., Morra, M.J., Guttieri, M.J., Brown, P.D. and Brown, J. 1998. Glucosinolate production by five field-grown Brassica napus cultivars used as green manures. Weed Technology. 12: 712-718.

El-Tarabily, K.A., Sykes, M.L., Kurtboke, I.D., Hardy, G.E.St.J., Barbosa, A.M. and Dekker, R.F.H. 1996. Synergistic effects of a cellulase-producing Micromonospora carbonacea and an antibiotic producing Streptomyces violascens on the suppression of Phytophthora cinnamomi root rot of Banksia grandis. Canadian Journal of Botany. 74: 618-624.

Ellis, M.B. 1971. Dematiaceous hyphomycetes. Commonwealth Agricultural Bureau, United Kingdom.

Ellis, M.B. 1976. More dematiaceous hyphomycetes. Commonwealth Agricultural Bureau, United Kingdom.

Ellis, M.B. and Ellis, J.P. 1997. Microfungi on land plants-an identification handbook. The Richmond Publishing Company Limited, England.

Erwin, D.C. and Ribeiro, O.K. 1996. Phytophthora diseases worldwide. The American Phytopathological Society Press. Minnesota, USA. pp 562.

Falk, S.P. and Parbery, D.G. 1995. Armillaria luteobubalina population structure in horticultural plantings in Victoria, Australia. Mycological Research. 99: 216-220.

Fairbanks, M.M., Hardy, G.E.StJ. and McComb, J.A. 2002. Mitosis and meiosis in plants are affected by the fungicide phosphite. Australasian Plant Pathology. 31: 281-297.

Fenn, M.E. and Coffey, M.D. 1984. Studies on the in vitro and in vivo antifungal activity of Fosetyl-Al and phosphorous acid. Phytopathology. 74: 606-611.

Flett, S.P., Ashcroft, W., Lim, T. and Jerie, P. 1990. Evaluation of phosphonic (phosphorous) acid for the control of Phytophthora root rot in processing tomatoes. Australasian Plant Pathology. 19: 131-132.

Forsberg, L. 1993. Protea diseases and their control. Department of Primary Industries. Queensland, Australian. pp 1.

Finlay, A.R. and McCraken, A.R. 1991. Microbial suppression of Phytophthora cinnamomi. Phytophthora (Eds. J.A. Lucas, R.C. Shattock, D.S. Shaw & L.R. Cooke), Cambridge University Press, pp. 383-398.

Galindo, J. and Zentmyer, G.A. 1964. Mating types in Phytophthora cinnamomi. Phytopathology. 54: 238-239.

Gardiner, J.B., Morra, M.J., Eberlein, C.V., Brown, P.D. and Borek, V. 1999. Alleochemicals released in soil following incorporation of rapeseed (Brassica napus) green manures. Journal of Agricultural and Food Chemistry. 47: 3837-3842.

Gerriste, R.G., Adeney, J.A., Baird, G. and Colquhoun, I. 1992. The reaction of copper ions and hypochlorite with minesite soils in relation to fungicidal activity. Australian Journal of Soil Research. 30: 723-735.

Gerstl, Z. Mingelgrin, U. and Yaron, B. 1977. Behaviour of Vapam (metham- sodium) and methylisothiocyanate [MITC] in soils. Soil Science Society of America Journal. 41: 545-548.

Gisi, U., Zentmyer, G.A. and Klure, L.J. 1980. Production of sporangia by Phytophthora cinnamomi and P. palmivora. Phytopathology. 70: 301-306.

Grant, B.R., Dunstan, R.H., Griffith, J.M., Niere, J.O. and Smillie, R.H. 1990. The mechanisms of phosphonic (phosphorous) acid action in Phytophthora. Australasian Plant Pathology. 19: 115-261.

Grinstein, A., Orion, D., Greenberger, A. and Katan, J. 1979. Solar heating of the soil for the control of Verticillium dahliae and Pratylenchus thornei in potatoes. Soil Borne Plant Pathogens (Eds. Schippers, B. and Gams, W.). Academic Press, New York, pp 431-438.

Guest, D. and Bompeix, G. 1990. The complex mode of action of phosphonates. Australasian Plant Pathology. 19: 113-115.

Guest, D. and Grant, B. 1991. The complex action of phosphonates as antifungal agents. Biological Reviews. 66: 159-187.

Halsall, D.M. 1982. A forest soil suppressive to Phytophthora cinnamomi and conducive to Phytophthora cryptogea. I Survival, germination and infectivity of mycelium and chlamydospores. Australian Journal of Botany. 30: 11-25.

Halsall, D.M. and Williams, J.D. 1984. Effect of root temperature on the development of Phytophthora cinnamomi root rot in Eucalyptus seedlings. Australian Journal of Botany. 32: 521-528.

Hanlin, R.T. 1998. Illustrated genera of the Ascomycetes. APS Press, St Paul, Minnesota, USA.

Hansen, E.M. and Hamm, P.B. 1996. Survival of Phytophthora lateralis in infected roots of port-orford cedar. Plant Disease 80: 1075-1078.

Hardham, A.R. and Gubler, F. 1990. Polarity of attachment of zoospores of a root pathogen and pre-alignment of the emerging germ tube. Cell Biology International Reports. 14: 947-956.

Harding, R.B. and Wicks, T.J. 2001. Effects of incorporating Brassica and cereal cover crop residues on soil populations of Verticillium dahliae. Proceedings of the 2^nd Soil-borne Diseases Conference, Lorne, Victoria. pp 148-149.

Hardy, G.E.StJ. 1983. Soil solarisation: effects on Fusarium wilt or carnations and on Verticillium wilt of eggplants. Honours Thesis. University of Western Australia.

Hardy, G.E.StJ. and Sivasithamparam, K. 1988. Phytophthora spp. associated with container-grown plants in nurseries in Western Australia. Plant Disease 72: 435­437.

Hardy, G.E.StJ. and Sivasithamparam, K. 1991. Effects of sterile and non-sterile leachates extracted from composted Eucalyptus bark and pine-bark container media on Phytophthora spp. Soil Biology and Biochemistry. 23: 25-30.

Hardy, G.E.StJ. and Sivasithamparam, K. 1991. How container media and matric potential affect the production of sporangia, oospores and chlamydospores by three Phytophthora species. Soil Biology and Biochemistry. 23: 31-39.

Hardy, G.E.StJ. and Sivasithamparam, K. 1991. Sporangial responses do not reflect microbial suppression of Phytophthora drechsleri in composted eucalyptus bark mix. Soil Biology and Biochemistry. 23: 757-765.

Hardy, G.E.StJ., Barrett, S. and Shearer, B.L. 2001. The future of phosphite as a fungicide to control the soilborne plant pathogen Phytophthora cinnamomi in natural ecosystems. Australasian Plant Pathology. 30: 133-139.

Hill, T.C.J., Tippett, J.T. and Shearer, B.L. 1994. Invasion of the Bassendean dune Banksia woodland by Phytophthora cinnamomi. Australian Journal of Botany. 42: 725-738.

Hill, T.C.J., Tippett, J.T. and Shearer, B.L. 1995. Evaluation of three treatments for eradication of Phytophthora cinnamomi from deep, leached sands in southwest Australia. Plant Disease. 79: 122-127.

Ho, H.H. and Zentmyer, G.A. 1977. Morphology of Phytophthora cinnamomi. Mycologia. 69: 701-713.

Hoitink, H.A.J., Van Doren, D.M.Jr and Schmitthenner, A.F. 1977. Suppression of Phytophthora cinnamomi in a composted hardwood bark mix. Phytopathology. 67: 561-565.

Hoitink, H.A.J. and Fahy, P.C. 1986. Basis for the control of soilborne plant pathogens with composts. Annual Review of Phytopathology. 24: 93-114.

Horsman, C. 2000. Phytophthora – One of the ‘Bad Bugs'. Journal of the Australian Flora and Protea Growers Association 12: 26-27.

Hüberli, D., Tommerup, I.C. and Hardy G.E.St.J. 2000. False-negative isolations or absence of lesions may cause mis-diagnosis of diseased plants infected with Phytophthora cinnamomi. Australasian Plant Pathology. 29: 164-169

Hwang, S.C. and Ko, W.H. 1978. Biology of chlamydospores, sporangia and zoospores of Phytophthora cinnamomi in soil. Phytopathology. 68: 726-731.

Irwin, J.A.G., Cahill. D.M. and Drenth. A. 1995. Phytophthora in Australia. Australian Journal of Agricultural Research. 46: 1311-1337.

Kaewruang, W., Sivasithamparam, K. and Hardy, G.E. 1989. Effect of solarization of soil within plastic bags on root rot of gerbera (Gerbera jamesonii L.). Plant and Soil. 120: 303-306.

Katan, J., Grennberger, A., Grinstein, A. and Alon, H. 1976. Solar heating by polyethylene mulching for the control of diseases caused by soilborne pathogens. Phytopathology. 66: 683-688.

Katan, J. 1980. Solar pasteurization of soil for disease control: status and prospects. Plant Disease. 64: 450-454.

Keast, D., Tonkin, C. and Sanfelieu, L. 1985. Effects of copper salts on growth and survival of P. cinnamomi in vitro and on the antifungal activity of Actinomycete populations from the roots of Eucalyptus marginata and Banksia grandis. Australian Journal of Botany. 33: 115-130.

Kennedy, B.W. and Erwin, D.C. 1961. Some factors influencing sporangia production of a Phytophthora species isolated from Lucerne in certain salt solutions. Transactions of the British Mycological Society. 44: 291-297.

Kinal, J., Shearer, B.L. and Fairman, R.G. 1993. Dispersal of Phytophthora cinnamomi through lateritic soil of laterally flowing subsurface water. Plant Disease. 77: 1085-1090.

Kirkegaard, J.A. and Sarwar, M. 1999. Glucosinolate profiles of Australian Canola (Brassica napus annua L.) and Indian Mustard (Brassica juncea L.) cultivars: implications for biofumigation. Australian Journal of Agricultural Research. 50: 315­324.

Kirkegaard, J.A., Sarwar, M., Wong, P.T.W., Mead, A, Howe, G. and Newell, M. 2000. Field studies on the biofumigation of take-all by Brassica break crops. Australian Journal of Agricultural Research. 51: 445-456.

Koike, S.T., Nelson, J.H., Bishop, T.B. and Perry, D.K. 1991. Verticillium wilt of protea (Leucospermum cordifolium) caused by Verticillium dahliae. Plant Disease. 75: 1074.

Knox-Davies, P.S. 1981. Comments on fungus diseases of plants indigenous to the South-Western Cape. Veld & Flora. 67: 88-91.

Knox-Davies, P.S., Van Wyk, P.S. and Marasas, W.F.O. 1987. Diseases of Protea, Leucospermum and Leucadendron recorded in South Africa. Phytophylactica. 19: 327-337.

Kreutzer, W.A. 1963. Selective toxicity of chemicals to microorgansims. Annual Review of Phytopathology. 1: 101-126.

Kreutzer, W.A. 1965. The reinfestation of treated soil. Ecology of soil-borne plant pathogens (Eds. Baker, K.F. and Snyder, W.C.), John Murray, London. pp 495­508.

Lake, J. 2000. Counting the benefits of treating water. Australian Nursery Manager. pp 4-6.

Larsen, P.O. 1981. Glucosinolates. The biochemistry of plants – secondary plant products Volume 7 (Ed. Conn, E.E.). Academic Press, New York, pp 501-525.

Leonard, E. 2000. Biofumigation – fact or fiction. Australian Grain. 10: 18-19.

Leonardi, J., Whiley, A.W., Langdon, P.W., Pegg, K.G., Cheyne, J. 1999. Progress on the use of foliar applications of phosphonate for the control of Phytophthora root rot in avocados. Talking Avocados Vol. 10. Orf Bartrup, Devonport, Tasmania. pp 18-21.

Lewis, J.A. and Papavizas, G.C. 1971. Effect of sulphur containing volatile compounds and vapors from cabbage decomposition on Aphanomyces euteiches. Phytopathology. 61: 208-214.

Maher, Y. 1979. Prediction of soil temperatures of a soil mulched with transparent polyethylene. Journal of Applied Meteorology. 18: 1263-1267.

Maher, Y. and Katan, J. 1981. Spatial soil temperatures under transparent polyethylene mulch – numerical and experimental studies. Soil Science. 131: 82-87.

Maier, N.A., Barth, G.E., Cecil, J.S., Chvyl, W.L. and Bartetzko, M.N. 1995. Effect of sampling time and leaf position on leaf nutrient composition of Protea ‘pink ice'. Australian Journal of Experimental Agriculture. 35: 275-283.

Malajczuk, N., McComb, A.J. and Parker, C.A. 1977. Infection by Phytophthora cinnamomi Rands of roots of Eucalyptus calophylla R. Br. and Eucalyptus marginata Donn. ex Sm. Australian Journal of Botany. 25: 483-500.

Malajczuk, N., Sanfelieu, C.L. and Hossen, S. 1983. Production and survival of Phytophthora cinnamomi zoospores in suppressive and conducive soils. Transactions of the British Mycological Society. 80: 305-312.

Manici, L.M., Lazzeri, L. and Palmieri, S. 1997. In vitro fungitoxic activity of some glucosinolates and their enzyme-derived products toward plant pathogenic fungi. Journal of Agricultural and Food Chemistry. 45: 2768-2773.

Manici, L.M, Lazzeri, L., Baruzzi, G., Leoni, O., Galletti, S. and Palmieri, S. 2000. Suppressive activity of some glucosinolate enzyme degradation products on Pythium irregulare and Rhizoctonia solani in sterile soil. Pest Management Science. 56: 921-926.

Marks, G.C. and Smith, I.W. 1988. Phytophthora cinnamomi disease control in proteas: a summary of investigations carried out in 1987/88. Report to the 1988 Annual Conference of the Australian Protea Growers Association, Robertson, NSW.

Marks, G.C. and Smith, I.W. 1990. Control of experimental Phytophthora cinnamomi stem infections of Rhododendron, Leucadendron and Eucalyptus by dimethomorph, Fosetyl-Al and metalaxyl. Australian Journal of Experimental Agriculture. 30: 139-143.

Marks, G.C. and Smith, I.W. 1992. Metalaxyl and Phosphonate as prophylactic and curative agents against stem infection of Leucadendron caused by Phytophthora cinnamomi. Australian Journal of Experimental Agriculture. 32: 255-259.

Marks, G.C., Smith, I.W. and Kassaby, F.Y. 1981. Trunk infection of Eucalyptus species by Phytophthora cinnamomi Rands: a preliminary report. Australian Journal of Botany. 20: 141-154.

Mathews, A. and Mathews, D. 1997. Proteas – an Australian cut flower growers guide. Proteaflora Nursery Pty Ltd. Monbulk, Australia. pp 4.

Matthiessen, J.N., Kirkegaard, J.A., Morra, M.J. and Warton, B. 2001. Taking biofumigation research beyond the empirical – towards linking isothiocyanate release and presence ins oil to pesticidal efficacy. Proceedings of the Second Soil-borne Diseases Conference, Lorne, Victoria. pp 144-145.

Mazzola, M., Granatstein, D.M., Elfving, D.C. and Mullinix, K. 2001. Suppression of specific apple root pathogens by Brassica napus seed meal amendment regardless of Glucosinolate content. Phytopathology. 91: 673-679.

McComb, J., Bennett, I., Stukely, M. and Crane, C. 1991. Selection and propagation of jarrah for dieback resistance: a progress report. The International Plant Propagators' Society – Combined Proceedings. 40: 86-90.

McCredie, T.A., Dixon, K.W. and Sivasithamparam, K. 1985. Grafting Banksias to avoid root rot. Australian Horticulture. 83: 75-79.

McCredie, T.A., Dixon, K.W. and Sivasithamparam, K. 1985. Variability in the resistance of Banksia L. f. species to Phytophthora cinnamomi Rands. Australian Journal of Botany. 33: 629-637.

McGovern, R.J., McSorley, R. and Urs, R.R. 2000. Reduction of Phytophthora blight of Madagascar Periwinkle in Florida by soil solarisation in autumn. Plant Disease. 84: 185-191.

McIntosh, D.L. 1964. Phytophthora spp. in soils of the Okanagan and Similkameen Valleys of British Columbia. Canadian Journal of Botany. 42: 1411-1415.

McLennan, R. 1993. Growing proteas. Kangaroo Press, Sydney, Australia.

Mirecetich, S.M., Zentmyer, G.A. and Kendrick, J.B. 1968. Physiology of germination of chlamydospores of Phytophthora cinnamomi. Phytopathology. 58: 666­671.

Mizubuti, E.S.G., Aylor, D.E. and Fry, W.E. 2000. Survival of Phytophthora infestans sporangia exposed to solar radiation. Phytopathology. 90: 78-84..

Morra, M.J. and Kirkegaard, J.A. 2002. Isothiocyanate release from soil- incorporated Brassica tissues. Soil Biology and Biochemistry. 34 : 1683-1690.

Munnecke, D.E. 1972. Factors affecting the efficacy of fungicides in soil. Annual Review of Phytopathology. 10: 375-398.

Munnecke, D.E. 1984. Establishment of micro-organisms in fumigated avocado soil to attempt to prevent reinvasion of soils by Phytophthora cinnamomi. Transactions of the British Mycological Society. 83: 287-294.

Nakao, P., Leonhardt, K., Nagata, T. and Shingaki, P. 2003. Screening Leucospermum hybrids for tolerance to Phytophthora cinnamomi. Acta Horticulture. 602: 113-117.

Nesbitt, H.J., Malajczuk, N. and Glenn, A.R. 1979. Effect of soil moisture and temperature on the survival of Phytophthora cinnamomi Rands in soil. Soil Biology and Biochemistry. 11: 137-140.

O'Gara, E., McComb, J.A., Colquhoun, I.J. and Hardy, G.E.St.J. 1997. The infection of non-wounded and wounded periderms tissue at the lower stem of Eucalyptus marginata by zoospores of Phytophthora cinnamomi, in a rehabilitated bauxite mine. Australasian Plant Pathology. 26: 135-141.

Old, K.M., Oros, J.M. and Malafant, K.M. 1984. Survival of Phytophthora cinnamomi in root fragments in Australian forest soils. Transactions of the British Mycological Society. 82: 605-613.

Orffer, S. and Knox-Davies, P.S. 1989. A canker and die-back of Protea repens. Phytophylactica. 21: 189-194.

O'Sullivan, G.J., Galea, V.J., Gordon, J.I. and Flethcher, R.J. 1999. Relative resistance to Phytophthora spp. amongst six rootstocks used for waxflower production. Proceedings of the 12^th Biennial Australasian Plant Pathology Conference, Canberra, pp 265.

Oudemans, P.V. 1999. Phytophthora species associated with cranberry root rot and surface irrigation water in New Jersey. Plant Disease. 83: 251-258.

Ouimette, D.G. and Coffey, M.D. 1989. Phosphonate levels in avocado (Persea americana) seedlings and soil following treatment with Fosetyl-Al or potassium phosphonate. Plant Disease. 73: 212-215.

Ouimette, D.G. and Coffey, M.D. 1990. Symplastic entry and phloem translocation of phosphonate. Pesticide Biochemistry and Physiology. 38: 18-25.

Pegg, K.G., Whiley, A.W., Langdon, P.W. and Saranah, J.B. 1987. Comparison of phosetyl-Al, phosphorous acid and metalaxyl for the long term control of Phytophthora root rot of avocado. Australian Journal of Experimental Agriculture. 27: 471-474.

Pegg, K.G., Whiley, J.B. and Hargreaves, P.A. 1990. Phosphonic (phosphorous) acid treatments control Phytophthora diseases in avocado and pineapple. Australasian Plant Pathology. 19: 122-124.

Pilbeam, R.A., Colquhoun, I.J., Shearer, B. and Hardy, G.E.St.J. 2000. Phosphite concentration: its effect on phytotoxicity symptoms and colonisation by Phytophthora cinnamomi in three understorey species of Eucalyptus marginata forest. Australasian Plant Pathology. 29: 86-95.

Pinkerton, J.N., Ivors, K.L., Miller, M.L. and Moore, L.W. 2000. Effect of soil solarisation and cover crops on populations of selected soilborne plant pathogens in western Oregon. Plant Disease. 84: 952-960.

Pinkerton, J.N., Ivors, K.L., Reeser, P.W., Bristow, P.R. and Windom, G.E. 2002. The use of soil solarisation for the management of soilborne plant pathogens in strawberry and red raspberry production. Plant Disease. 86: 645-651.

Pratt, B.H., Sedgley, J.H., Heather, W.A. and Shepherd, C.J. 1972. Oospore production in Phytophthora cinnamomi in the presence of Trichoderma koningii. Australian Journal of Biological Sciences. 25: 861-863.

Podger, F.D., Doepel, R.F. and Zentmyer, G.A. 1965. Association of Phytophthora cinnamomi with a disease of Eucalyptus marginata forest in Western Australia. Plant Disease Reporter. 49: 943-947.

Podger, F.D. 1972. Phytophthora cinnamomi, a cause of lethal disease in indigenous plant communities in Western Australia. Phytopathology. 62: 972-981.

Porter, I., Brett, R. and Wiseman, B. 1998. Alternatives to soil fumigation with methyl bromide in the strawberry and ornamentals industries. Horticultural Research and Development Corporation Final Report. Project No. HG95015.

Porter, I., Brett, R. and Wiseman, B. 1999. Alternatives to methyl bromide: chemical fumigants or integrated pest management systems. Australasian Plant Pathology. 28: 65-71.

Price, G.H., Cresswell, G.C. and Handreck, K.A. 1997. Ornamentals. Plant Analysis – an interpretation manual. 2nd Ed. (Eds. Reuter, D.J. and Robinson, J.B.). CSIRO Publishing, Australia.

Rabie, M. 2000. Protea production around the world. Journal of the International Protea Association. 40: 7.

Rands, R.D. 1922. Streepkanker van Kaneel, veroorzaakt door Phytophthora cinnamomi n. sp. (stipe canker of cinnamon caused by Phytophthora cinnamomi n. sp.) Meded. Inst. Plantenziekten 54: 41.

Rayment, G.E. and Higginson, F.R. 1992. Australian laboratory handbook of soil and water chemical methods. Inkata Press, Melbourne, Australia.

Ribeiro, O.K. 1978. A sourcebook of the genus Phytophthora. (Ed. Cramer, J.). Germany.

RIRDC 1997. The Australian wildflower industry. A review (Second Edition).

Rural Industry Research and Development Corporation Research Paper. No.97/64..

Sarwar, M., Kirkegaard, J.A., Wong, P.T.W. and Desmarchelier, J.M 1998. Biofumigation potential of Brassica's III. In vitro toxicity of isothiocyanates to soil- borne fungal pathogens. Plant and Soil. 201: 103-112.

Schnurer, J. and Rosswall, A. 1982. Fluoroscein diacetate hydrolysis as a measure of microbial activity in soil and litter. Applied Environmental Microbiology. 43: 1256-1261.

Serfontein, S. and Knox-Davies, P.S. 1990. A flower-head blight of Leucadendron discolor caused by Botrytis cinerea. Phytophylactica. 22: 119-120.

Serfontein, S. and Knox-Davies, P.S. 1990. Leaf spot of Protea magnifica and copper leaf of Leucospermum cordifolium associated with Coleroa senniana. Phytophylactica. 22: 103-107.

Shea, S.R., Gillen, K.J. and Leppard, W.I. 1980. Seasonal variation in population levels of Phytophthora cinnamomi Rands in soil in diseased, freely drained Eucalyptus marginata Sm sites in the northern jarrah forest of south-western Australia. Protection Ecology. 2: 135-156.

Shea, S.R. 1979. Phytophthora cinnamomi Rands – a collar rot pathogen of Banksia grandis Wild. Australasian Plant Pathology. 8: 32-34.

Shearer, B.L. 1994. The major plant pathogens occurring in native ecosystems of south-western Australia. Journal of the Royal Society of Western Australia. 77: 113­122.

Shearer, B.L. and Dillon, M. 1995. Susceptibility of plant species in Eucalyptus marginata forest to infection by Phytophthora cinnamomi. Australian Journal of Botany. 43: 113-134.

Shearer, B.L and Fairman, R. 1997. Application of phosphite delays and reduces the rate of mortality of three Banksia species in communities infected with Phytophthora cinnamomi. Proceedings of the 11^th Biennial Australasian Plant Pathology conference, Perth, pp 180.

Shearer, B.L. and Tippett, J.T. 1989. Jarrah Dieback: The dynamics and management of Phytophthora cinnamomi in the Jarrah (Eucalyptus marginata) forest of south-western Australia. Research Bulletin No. 3. Department of Conservation and Land Management, Perth.

Shetty, K.G., Subbarao, K.V., Huisman, O.C. and Hubbard, J.C. 2000. Mechanisms of broccoli-mediated Verticillium wilt reduction in cauliflower. Phytopathology. 90: 305-310.

Shivas, R.G. 1989. Fungal and bacterial diseases of plants in Western Australia. Journal of the Royal Society of Western Australia. 72: 1-62.

Silber, A., Ganmore-Neumann, R. and Ben-Jaacov, J.B. 1998. Effects of nutrient addition on growth and rhizosphere pH of Leucadendron ‘Safari Sunset'. Plant and Soil. 199: 205-211.

Sivasithamparam, K. and Gos, O.M. 1980. Jarrah dieback – a threat to horticulture. Journal of Agriculture – Western Australia.

Sivasithamparam, K. 1991. Management of root rot. Proceedings of the 6^th Biennial International Protea Association Conference, Perth, Western Australia. pp 283­287.

Skipworth, E. and Bus, B. 1991. A budget assessment of protea growing in Western Australia. Proceedings of the 6^th Biennial International Protea Association Conference, Perth, Western Australia. pp 81-88.

Smith, B.J. and Kirkegaard, J.A. 2002. In vitro inhibition of soil microorganisms by 2-phenylethyl isothiocyanate. Plant Pathology. 51: 585-593.

Smith, B.J., Shearer, B.L. and Sivasithamparam, K. 1997. Compartmentalisation of Phytophthora cinnamomi in stems of highly susceptible Banksia brownii treated with phosphonate. Mycological Research. 101: 1101-1107.

Smolinska, U., Knudsen, G.R., Morra, M.J. and Borek, V. 1997. Inhibition of Aphanomyces euteiches f. sp. pisi by volatiles produced by hydrolysis of Brassica napus seed meal. Plant Disease. 31: 288-292.

Spencer, S. and Benson, D.M. 1982. Pine bark, hardwood compost and peat amendment effects on development of Phytophthora spp. and Lupine Root Rot. Phytopathology. 72: 346-351.

Steel, R.G.D. and Torrie, J.H. 1986. Principles and procedures of statistics. A biometrical approach (2^nd Ed). McGraw-Hill, USA.

Stephens, P.M., Davoren, C.W. and Wicks, T. 1999. Effect of methyl bromide, metham sodium and the biofumigants Indian mustard and canola on the incidence of soilborne fungal pathogens and growth of grapevine nursery stock. Australasian Plant Pathology. 28: 187-196.

Subbarao, K.V., Hubbard, J.C. and Koike, S.T. 1999. Evaluation of broccoli residue incorporation into field soil for Verticillium wilt control in cauliflower. Plant Disease. 83: 124-129.

Sutton, B.C. 1980. The Coelomycetes. Commonwealth Mycological Institute, Kew, United Kingdom.

Swart, L., Denman, S., Lamprecht, S.C. and Crous, P.W. 1999. Fusarium wilt: a new disease of cultivated protea in southern Africa. Australasian Plant Pathology. 28: 156-161.

Swart, L. and Denman, S. 2000. Chemical control of Phytophthora cinnamomi in potted Leucospermum plants. Australasian Plant Pathology. 29: 230-239.

Taylor, L.E. and Crous, P.W. 2000. Fungi occurring on Proteaceae. New anamorphs for Teratosphaeria, Mycospherella and Lembosia, and other fungi associated with leaf spots and cankers on Proteaceous hosts. Mycological Research. 104: 618-636.

Tesoriero, L., Lidbetter, J., M. Terras and Drenth, A. 2001. Phytophthora species associated with root and crown rots of Boronia in cut-flower production. Proceedings of the 13^th Biennial Plant Pathology Conference, Cairns, Australia. pp 290.

Thompson, J.P., Owen, K.J. and Clewett, T.G. 2001. Pre-cropping with canola decreases vesicular-arbuscular mycorrhiza and growth of wheat in a low phosphorous soil. Proceedings of the Second Soil-borne Diseases Conference, Lorne, Victoria.

Tippett, J.T. and O'Brien, T.P. 1976. The structure of eucalypt roots. Australian Journal of Botany. 24: 619-632.

Tippett, J.T. and Hill, T.C. 1984. Role of periderm in resistance of Eucalyptus marginata roots against Phytophthora cinnamomi. European Journal of Forest Pathology. 14: 431-439.

Turnbull, L.V. 1991. Production of Phytophthora tolerant rootstocks: I. Screening proteas for resistance to Phytophthora cinnamomi. Proceedings of the 6^th Biennial International Protea Association Conference, Perth, Western Australia. pp 275­280.

Turnbull, L.V., Crees, L.R., Brits, G.J. and Wright, M.G. 1995. Field studies on the effectiveness of phosphonate suppression of Phytophthora root rot in proteas. Acta Horticulturae. 387: 141-151.

Turnbull, L.V., Ogle, H.J. and Dart, P.J. 1989. Biological control of Phytophthora cinnamomi in proteas. Proceedings of the Western Australian Department

of Agriculture Conference on the Production and Marketing of Australian Flora, Perth, Western Australia.

Turnbull, L.V., Ogle, H.J., Stirling, A.M. and Dart, P.J. 1992. Preliminary investigations into the influence of Pseudomonas cepacia on infection and survival of proteas in Phytophthora infected potting mix. Scientia Horticulture. 52: 257-263.

Tynan, K.M., Scoot, E.S., Sedgley, M., Dixon, K., Sivasithamparam, K., Brits, G.J. and Wright, M.G. 1995. Phytophthora dieback in banksias: screening for resistance, Acta Horticulturae. 387: 159-162.

Tynan, K.M., Wilkinson, C.J., Holmes, J.M., Dell, B., Colquhoun, I.J., McComb, J.A. and Hardy, G.E.StJ. 2001. The long-term ability of phosphite to control Phytophthora cinnamomi in two native plant communities of Western Australia. Australian Journal of Botany. 49: 761-770.

Utkhede, R.S. and Hogue, E.J. 1999. Influence of ground cover on development of Phytophthora crown and root rot of apple trees. Canadian Journal of Plant Pathology. 21: 106-109.

Utkhede, R.S. and Smith, E.M. 2000. Impact of chemical, biological and cultural treatments on the growth and yield of apple in replant-disease soil. Australasian Plant Pathology. 29: 129-136.

Vawdrey, L.L. 2001. Quantification of inoculum density of Phytophthora palmivora in soil and its relation to disease incidence in papaw in far northern Queensland. Australasian plant Pathology. 30: 199-204.

Vawdrey, L.L., Martin, M. and De Faveri, J. 2002. The potential of organic and inorganic soil amendments, and a biological control agent (Trichoderma sp.) for the management of Phytophthora root rot of papaw in far northern Queensland. Australasian Plant Pathology. 31: 391-399.

Vigo, C., Norman, J.R. and Kooker, J.E. 2000. Biocontrol of the pathogen Phytophthora parasitica by arbuscular mycorrhizal fungi is a consequence of effects on infection loci. Plant Pathology. 49: 509-514.

Von Arx, J.A. 1981. The genera of fungi sporulating in pure culture. (Ed. Carmer, J.). Freie Universitat, Berlin.

Von Broembsen, S.L. 1984. Occurrence of Phytophthora cinnamomi on indigenous and exotic hosts in South Africa, with special reference to the south-western Cape province. Phytophylactica. 16: 221-225.

Von Broembsen, S.L. and Brits, B. 1985. Phytophthora Root Rot of Commercially Cultivated Proteas in South Africa. Plant Disease. 69: 211-213.

Von Broembsen, S.L. and Kruger, F.J. 1985 Phytophthora cinnamomi associated with mortality of native vegetation in South Africa. Plant Disease. 69: 715­717.

Von Broembsen, S.L. 1986. Blight of pincushions (Leucospermum spp.) caused by Drechslera dematioidea. Plant Disease. 70: 211-213.

Waller, J.M., Lenne, J.M. and Waller, S.J. 2002. Plant pathologists pocketbook. 3^rd Ed. CABI Publishing, London.

Warton, B., Matthiessen, J.N. and Shackleton, M.A. 2001. Glucosinolate content and isothiocyanate evolution – two measures of the biofumigation potential of plants. Journal of Agricultural and Food Chemistry. 49: 5244-5250.

Warton, B., Matthiessen, J.N. and Roper, M.M. 2001. The soil organisms responsible for the enhanced biodegradation of metham sodium. Biology and Fertility of Soils. 34; 264-269.

Waterhouse, G.M. 1963. Key to the species of Phytophthora de Barry. Mycological Papers. No 122.

Webb, M. 1997. Growing Proteas. Farmnote No 79/96. Agriculture Western Australia Publication.

http://www.agric.wa.gov.au/agency/pubns/farmnote/1996/f07996.htm

Weste, G. 1983. Population dynamics and survival of Phytophthora. Phytophthora: its biology, taxonomy, ecology and pathology (Eds. Erwin, D.C., Bartnicki-Garcia, S. and Tsao, P.H.). The American Phytopathological Society, Minnesota, USA. pp 237-257.

Weste, G. 1994. Impact of Phytophthora species on native vegetation of Australian and Papua New Guinea. Australasian Plant Pathology. 23: 190-209.

Weste, G. and Marks, G.C. 1987. The biology of Phytophthora cinnamomi in Australasian forests. Annual Review of Phytopathology. 25: 207-229.

Weste, G. and Ruppin, P. 1977. Phytophthora cinnamomi: population densities in forest soils. Australian Journal of Botany. 25: 461-475.

Weste, G. and Taylor, P. 1971. The invasion of native forest by Phytophthora cinnamomi I. Brisbane Ranges, Victoria. Australian Journal of Botany. 19: 281-294.

Weste, G. and Vithanage, 1978. Seasonal variation in numbers of chlamydospores in Victorian forest soils infected with Phytophthora cinnamomi. Australian Journal of Botany. 26: 657-662.

Weste, G. and Vithanage, K. 1979. Production of sporangia by Phytophthora cinnamomi in forest soils. Australian Journal of Botany. 27: 693-701.

Whiley, A.W., Hargreaves, P.A., Pegg, K.G., Doogan, V.J., Ruddle, L.J., Saranah, J.B. and Langdon, P.W. 1995. Changing sink strengths influence translocation of phosphonate in avocado (Persea americana Mill.) trees. Australian Journal of Agricultural Research. 46: 1079-1090.

Wongwathanarat, P. and Sivasithamparam, K. 1991. Effect of phosphonate on the rhizosphere microflora and the development of root rot of avocado and peppercorn tree seedlings caused by Phytophthora cinnamomi. Biology and Fertility of Soils. 11: 13-17.

Wicks, T.J. and Hall, B. 1988. Preliminary evaluation of phosphorous acid, Fosetyl-Al and metalaxyl for controlling Phytophthora cambivora on almond and cherry. Crop Protection. 7: 314-318.

Wilcox, W.F., Pritts, M.P. and Kelly, M.J. 1999. Integrated control of Phytophthora root rot of red raspberry. Plant Disease. 83: 1149-1154.

Wilkinson, C.J., Holmes, J.M., Tynan, K.M., Colquhoun, I.J., McComb, J.A., Hardy, G.E.StJ. and Dell, B. 2001. Ability of phosphite applied in a glasshouse trial to control Phytophthora cinnamomi in five plant species native to Western Australia. Australasian Plant Pathology. 30: 343-351.

Wills, R.T. 1992. The ecological impact of Phytophthora cinnamomi in the Stirling Range National Park, Western Australia. Australian Journal of Ecology. 18:145-159.

Wilton, J.H., Basham, L.M., Hoitink, H.A.J. and Doskotch, R.W. 1983. Isolation of inhibitors of Phytophthora cinnamomi from composted hardwood bark. Phytopathology. pp 1010.

Xie, J., Cardenas, E.S., Sammis, T.W., Wall, M.M., Lindsey, D.L. and Murray, L.W. 1999. Effects of irrigation method on chile pepper yield and Phytophthora root rot incidence. Agricultural Water Management. 42: 127-142.

Yip, H.Y. 1989. Five new species of Phyllosticta on Australian native plants. Mycological Research. 93: 489-496.

Zentmyer, G.A. 1980. Phytophthora cinnamomi and the diseases it causes.

Monograph 10, American Phytopathological Society, St. Paul, USA.

Zentmyer, G.A and Marshall, L.A. 1959. Factors affecting sporangia production by Phytophthora cinnamomi. Phytopathology. 49: 445.