Susceptibility of Plant Species in Banksia Woodlands on the Swan

Aust. J. Bot., 1996,44,433-445

Susceptibility of Plant Species in Banksia Woodlands on the Swan Coastal Plain, Western Australia, to Infection by Phytophthora cinnamomi B. L. sheareflB and M.

ill on*

*science and Information Division, Department Conservation and Land Management, 50 Hayman Road, Como, WA 6152, Australia. B~orrespondingauthor; email: [email protected]

Abstract Estimates of the susceptibility of plant species in Banksia woodland to Phytophthora cinnamomi Rands were obtained by determining the incidence of plant death and frequency of isolation of the pathogen, among species occurring in 46 disease centres on the Swan Coastal Plain south of Perth, Western Australia. In the disease centres, dicotyledons outnumbered monocotyledons. About half of all species occurring in the disease centres were from four families of dicotyledons, with the largest number of species from the Myrtaceae, Proteaceae and Papilionaceae. The greatest number of species of monocotyledons were from the Anthericaceae and Cyperaceae. No deaths were recorded for 47% of species found in three or more disease centres. These species were mainly from the Cyperaceae, Haemodoraceae, Myrtaceae and Papilionaceae. The species that tended to die frequently in disease centres were mainly from the Papilionaceae, Proteaceae, Epacridaceae, Xanthorrhoeaceae and the Zamiaceae. Phytophthora cinnamomi was isolated from 26 of the 95 species occurring in three or more disease centres. For most species, the frequency of isolation of P. cinnamomi from recently dead plants was much less than the frequency of dead plants sampled. Isolation from plants was less frequent than from adjacent soil. The pathogen was isolated from recently dead plants or soil mainly for species of the Proteaceae, Myrtaceae, Papilionaceae, Dasypogonaceae, Iridaceae and Xanthorrhoeaceae. Crosstabulation of species by incidence of plant death and isolation of P. cinnamomi from plant and soil, provided the opportunity to classify the response of plant species to infection by P. cinnamomi.

Introduction Phytophthora cinnamomi Rands is a major threat to the conservation of Banksia woodland of the Swan Coastal Plain as it significantly changes the floristic composition and the structural complexity of the vegetation (Shearer and Dillon 1996). The proportion of species dying in P. cinnamomi disease centres in Banksia woodland on the Swan Coastal Plain south of Perth varies from between 10% and 64% and is positively correlated with level of impact (Shearer and Dillon 1996). In 64-m2 quadrats, P. cinnamomi infestation has decreased the number of species by 7-15 compared with healthy woodland nearby (Shearer and Hill 1989; Shearer and Dillon 1996). Extinction of dominant overstorey and understorey species, through infection, results in substantial reduction of vegetation biomass and the disease front being delineated by a sharp boundary of dying plants (Hill et al. 1994; Shearer and Dillon 1996). Despite the high impact of P. cinnamomi in Banksia woodland, the susceptibility of component plant species to infection has rarely been documented for this community (Shearer and Hill 1989; Shearer and Dillon 1996). Podger (1968) sampled plants that showed acute chlorosis, stunting and microphylly, or that had recently died, in six disease centres of the Swan Coastal Plain. Hill (1990) listed 65 plant species of the Swan Coastal Plain north of Perth that were considered susceptible to P. cinnamomi.

B. L. Shearer and M. Dillon

The objective of this study was to determine the occurrence and mortality of plant species and the frequency of isolation of the pathogen from recently dead plants, in P,cinnamomi disease centres in Banksia woodland on the Swan Coastal Plain south of Perth. From this information, plant species were classified into susceptibility groups on the basis of the severity of P. cinnamomi infection.

Materials and Methods The vegetation was assessed and sampled in 46 disease centres infested with P. cinnamomi, in Banksia wonc!!rmrl 05 the u r z n Ceasta! ?laic seu!h ef Perth. L ~ c a t i c maxd descriptioxs ~f the vegetation, landscape elements and soils of the centres are given by Shearer and Dillon (1996). Disease centres were visited from 1989 to 1991, mainly in autumn, winter and spring, when soil was moist, and the occurrence and health of perennial plant species situated within a plot of 20 m radius (Havel 1975) near the disease front were determined. For each species, health was rated from all alive to all dead according to the scale described by Shearer and Dillon (1996), and relative importance was rated as occasional (present, observable, but in small numbers only), frequent (common locally, but not uniformly over the whole area), abundant (common over the whole area), and dominant (dominant over the area). The relative importance rating is similar to that of Havel (1975) as outlined by Shearer and Dillon (1995). Voucher specimens are held at the Research Centre, Dwellingup, and the Western Australian Herbarium. Nomenclature follows that of Green (1985) and Marchant et al. (1987). Three recently dead plants of each species, together with adjacent soil, were sampled for isolation of P. cinnamomi in each disease centre. The stumps of large plants were excavated and bark and wood samples were taken from the collar and roots. For small plants, the collar and as much of the root system as possible were removed from the soil. Soil samples, 0-30 cm of surface soil, were taken adjacent to the sampled plants. Samples from plants or soil were bulked to give one plant or soil sample for each plant species sampled in a disease centre. Samples were kept moist in plastic bags until processing, which was generally within a few days of sampling. Utensils were washed with ethanol after collection of each sample. Procedures for isolation of P. cinnamomi from plant material and soil are described by Shearer and Dillon (1995). The data were cross-tabulated and sorted into appropriate groupings. Only species that occurred in three or more disease centres were included in the final groupings, to exclude infrequently occurring species. The number of centres in which a plant species occurred was used to calculate the proportion of centres in which a plant species was a host, in order to avoid distortion caused by the frequency of plant species occurrence. Where appropriate, mean and standard error of the mean were calculated. Relationships between variables were determined by calculating Pearson correlation coefficients.

Results Floristics In the disease centres, dicotyledons outnumbered monocotyledons (Table 1). Forty-four percent of the 177 species assessed were from four families of dicotyledons with the largest number of species from the Myrtaceae, Proteaceae and Papilionaceae. Twelve percent of species were monocotyledons from the Anthericaceae and Cyperaceae. Fifty-four percent of the species occurred in three or more disease centres.

Floristics in Relation to Isolation of P. cinnamomi and Plant Death Phytophthora cinnamomi was isolated from 26 of the 95 species occurring in three or more disease centres (Table 1). For most species, the frequency of isolation of P. cinnamomi from recently dead plants was much less than the frequency of dead plants sampled (I= -0.80 + 0.54 D, 3 = 0.76, P =S 0.01, where I = number of centres in which P. cinnamomi was isolated from recently dead plants for each plant species and D = the number of centres in which a plant species died). Isolation from plants was less frequent than from adjacent soil (I = -0.16 0.81 S, = 0.93, P s 0.01, S = the number of centres in which a plant species

+

Susceptibility of Plants to Phytophthora cinnamomi

Table 1. Numbers of genera and species, and corresponding numbers of hosts (in parentheses), that occurred in the disease centres infested with Phytophthora cinnamomi in Banksia woodland of the Swan Coastal Plain south of Perth Hosts were those taxa from which P. cinnamomi was isolated from recently dead plants Class, Family

Taxa in all centres Genera

Species

Taxa in 3 3 centres Genera

Species

Filicopsida Dennstaedtiaceae Coniferopsida Cupressaceae Cycadopsida Zamiaceae Magnoliopsida Myrtaceae Proteaceae Papilionaceae Epacridaceae Mimosaceae Apiaceae Dilleniaceae Goodeniaceae Casuarinaceae Rhamnaceae Rutaceae Tremandraceae Lamiaceae Lauraceae Rubiaceae Aizoaceae Euphorbiaceae Loganiaceae Loranthaceae Pittosporaceae Polygalaceae Santalaceae Lobeliaceae Liliopsida Anthericaceae Cyperaceae Restionaceae Haemodoraceae Dasypogonaceae Iridaceae Xanthorrhoeaceae Juncaceae Phonniaceae Total

died and P. cinnamomi was isolated from adjacent soil). This pattern of isolation was similar to that found for E, marginata forest (Shearer and Dillon 1995). The pathogen was isolated mainly from species of Proteaceae, Myrtaceae and Papilionaceae and from species of Dasypogonaceae, Iridaceae and Xanthorrhoeaceae (Table 1).

B.L. Shearer and M. Dillon

Examination of plant death and isolation frequency indicated that the response of the Banksia woodland vegetation to infection by P. cinnamomi can be divided into the following five groups:

(I) no individuals of a species died in disease centres; (2) individuals of a species died in one-third or less of the disease centres in which they occurred, and P. cinnamomi was not isolated from recently dead plants; (3) individuals of a species died in one-third or less of the disease centres in which they occurred, and P. cinnamomi was isolated relatively frequently from recently dead plants; (4) individuals of a species died in one-third or more of the disease centres in which they occurred, and P. cinnamomi was relatively infrequently isolated from recently dead plants; and (5) individuals of a species died in one-third or more of the disease centres in which they occurred, and P, cinnamomi was relatively frequently isolated from recently dead plants. Group 1 No deaths were recorded for 47% of species occurring in three or more disease centres (Table 2). Of these species, 41% were dicotyledons and 65% were monocotyledons. The largest number of species were of the Cyperaceae, Haemodoraceae, Myrtaceae and Papilionaceae. Twenty-four percent of the 45 species in this group occurred in 10 or more centres (Table 2). Schoenus sp. of the Cyperaceae, species of the Restionaceae and Baeckea camphorosmae were abundant in the disease centres in which they occurred. Group 2 Twenty-one percent of the species of dicotyledons and 23% of the species of monocotyledons died in less than a third of disease centres in which they occurred, and P. cinnamomi was not isolated from recently dead plants (Table 3). The largest number of species were of the Myrtaceae, Proteaceae and Restionaceae. Thirty percent of the 20 species in this group occurred in 10 or more centres (Table 3). Lepidosperma angustatum, Phlebocaiya ciliata, Calytrix sp., Hypocalymma angustifolium, Melale_uca species B (Marchant et al. 1987), Scholtzia sp., S. involucrata, Adenanthos cygnorum subsp. chamaephyton and Restio sp. were abundant in the disease centres in which they occurred. Group 3 Only 9% of the species of dicotyledons and 8% of the species of the monocotyledons were included within this group (Table 4). The largest number of species were of the Myrtaceae. All of the eight species could be found in more than 10 sites, and Dasypogon bromeliifolius, Patersonia occidentalis, Kunzea ericifolia and Bossiaea eriocarpa were abundant in the disease centres in which they occurred. Even though the percentage of centres within which the species died was relatively low (mean + s.e., 20 + 4%), P. cinnamomi was frequently isolated from plant samples (Table 4). The plant isolation frequency per species ranged from 12% to loo%, with a mean of 59% + 11%. Acacia stenoptera and Kunzea ericifolia are hosts to P. cinnamomi that are additional to those reported by Podger (1968) and Hill (1990). Group 4 The eight species in this group were dicotyledons and 8% of all species (Table 5). The largest number of species were of the Myrtaceae and Papilionaceae. Four species occurred in 10 or more disease centres, and Platysace compressa, Acacia pulchella, Viminaria juncea and Stirlingia latifolia were abundant in the centres in which they occurred. Although the


Table 2. Plant species of Banksia woodland for which no individuals died within the centres in which they occurred (Group 1) and that occurred in three or more disease centres situated on the Swan Coastal Plain south of Perth Figures shown for each species are the number of disease centres in which the species occurred and the number in which relative importance was rated frequent to dominant Family

Species

Anthericaceae

Thysanotus tenellus Tricoryne eiatior Xanthosia huegelii Lepidosperma sp. Mesomelaena tetragona Schoenus curvijolius S. rigens Lomandra integra L. nigricans L. sonderi Hibbertia racemosa H. vaginata Leucopogon glabellus L. propinquus Dampiera linearis Lechenaultia biloba Scaevola calliptera Anigozanthos manglesii Conostylis aculeata C. serrulata Haemodorum paniculatum Orthrosanthus laxus Hemiandra pungens Cassytha flava C. glabella Nuytsia floribunda Acacia extensa A. huegelii A. semitrullata Agonis flexuosa Baeckea camphorosmae Eucalyptus calophylla Melaleuca preissiana Gompholobium tomentosum Hardenbergia comptoniana Hovea trispenna Kennedia prostrata Dryandra nivea Hakea lissocarpha Persoonia aff. saccata Leptocarpus scariosus Loxocarya flexuosa Lyginia barbata Opercularia echinocephala Eriostemon spicatus

Apiaceae Cyperaceae

Dasypogonaceae Dilleniaceae Epacridaceae Goodeniaceae Haemodoraceae

Iridaceae Lamiaceae Lauraceae Loranthaceae Mimosaceae Myrtaceae

Papilionaceae

Proteaceae Restionaceae Rubiaceae Rutaceae

Occurrence

Relative importance


Table 3. Plant species of Banksia woodland for which individuals died in one third or less of the Phytophthora cinnamomi disease centres in which they occurred, from which the pathogen was not isolated from recently dead plants (Group 2), and that occurred in three or more disease centres situated on the Swan Coastal Plain south of Perth Figures shown for each species are the number of disease centres in which the species occurred, the number in which relative importance was rated frequent to dominant, and the number of times that plant death was recorded and that P. cinnamomi was isolated from dead plants or soil Family

Species

Occurrence

Relative Plants importance dead

Isolation of P. cinnamomi Plants

Anthericaceae Cyperaceae Haemodoraceae Mimosaceae Myrtaceae

Papilionaceae Pittosporaceae Proteaceae

Restionaceae

Soil

Thysanotus sp. Lepidospemza angustatum Phlebocarya ciliata Acacia saligna Calytrix sp. C.jlavescens Hypocalymma angustifolium Melaleuca species B Melaleuca sp. Scholtzia sp. S. involucrata Burtonia conferta Billara'iera variifolia Adenanthos cygnorum A. cygnorum subsp. chamaephyton Persoonia longifolia Xylomelum occidentale Hypolaena fastigiata Loxocarya cinerea Restio sp.

Table 4. Plant species of Banksia woodland for which individuals died in one-third or less of the Phytophthora cinnamomi disease centres in which they occurred, from which the pathogen was isolated relatively frequently from recently dead plants (Group 3), and that occurred in three or more disease centres situated on the Swan Coastal Plain south of Perth Figures shown for each species are the number of disease centres in which the species occurred, the number in which relative importance was rated frequent to dominant, and the number of times that plant death was recorded and that P. cinnamomi was isolated from dead plants or soil Family

Species

Occurrence



--

Dasypogonaceae Epacridaceae Iridaceae Mimosaceae Myrtaceae Papilionaceae Proteaceae

Dasypogon bromeliifolius Conostephiumpendulum Patersonia occidentalis Acacia stenoptera Hypocalymma robustum Kunzea ericifolia Bossiaea eriocalpa Petrophile linearis

Soil


Table 5. Plant species of Banksia woodland for which individuals died in one third or more of the Phytophthora cinnamomi disease centres in which they occurred, from which the pathogen was relatively infrequently isolated from recently dead plants (Group 4), and that occurred in three or more disease centres situated on the Swan Coastal Plain south of Perth

Figures shown for each species are the number of disease centres in which the species occurred, the number in which relative importance was rated frequent to dominant, and the number of times that plant death was recorded and that P. cinnamomi was isolated from dead plants or soil Family

Apiaceae Casuarinaceae Mimosaceae Myrtaceae Papilionaceae Proteaceae

Species

Platysace compressa Allocasuarinafiaseriana Acacia pulchella Calytrixfraseri Eucalyptus marginata Jacksonia horrida Viminariajuncea Stirlingia latifolia

Occurrence


9 11 31 5 26 4 3

4 5 20 3 11 3 2

6 6 21 3 12 2 2 14

17

7


Soil

0 1 1 0 2 0 0 1

1 1 1 0 2

1 0 4

species died in a third or more of disease centres in which they occurred (55% 2 5 % ) , P. cinnamomi was infrequently isolated from plant samples (Table 5). Plant isolation frequency per species ranged from 0% to 20%, with a mean of 7% 3%. Group 5 The 14 species included in this susceptible group were 15% of all species found in three or more disease centres. Sixteen percent were dicotyledons and only 4% monocotyledons (Table 6). The largest number of species were of the Proteaceae. Most of the species occurred in 10 or more disease centres, with 77% of the species being abundant in the centres in which they occurred (Table 6). The percentage of centres within which the species died ranged from 35% to loo%, with a mean of 72% 6%. Phytophthora cinnamomi was frequently isolated from plant and soil samples (Table 6) and isolation frequency from plants per species ranged from 24% to loo%, with a mean of 53% 6%. Jacksonia furcellata, J. sternbergiana, Melaleuca thymoides and Pericalymma ellipticum are hosts to P. cinnamomi that are additional to those reported by Podger (1968) and Hi11 (1990).

*

*

Families and Susceptibility Group The distribution of species among susceptibility groups was plotted for each of seven families having the largest number of species occurring in three or more disease centres (Fig. 1). Species of the Proteaceae were mainly included in Group 5, but taxa from this family were also in Groups 1 and 2. Species of the Papilionaceae, Myrtaceae and Mimosaceae were included in all groups, but mainly in Groups 1 and 2. By contrast, all species of the Restionaceae, Cyperaceae and Haemodoraceae were included in Groups 1 and 2. Host Species Occurring in Less than Three Centres Callitris preissii, Andersonia lehmanniana, Chorizema sp. and Gompholobium capitatum occurred in less than three centres and were hosts to P. dnnamomi. ~ a l l i t r i s~reissiiand Gompholobium capitatum are hosts to P. cinnamomi that are additional to those reported by Podger (1968) and Hill (1990).


Table 6. Plant species of Banksia woodland for which individuals died in one third or more of the Phytophthora cinnamomi disease centres in which they occurred, from which the pathogen was relatively infrequently isolated from recently dead plants (Group S), and that occurred in three or more disease centres situated on the Swan Coastal Plain south of Perth

Figures shown for each species are the number of disease centres in which the species occurred, the number in which relative importance was rated frequent to dominant, and the number of times that plant death was recorded and that P. cinnamomi was isolated from dead plants or soil Family

Species

Occurrence



Soil

Dilleniaceae

Hibbertia hypericoides H. aff. subvaginata Epacridaceae Leucopogon conostephioides Melaleuca thymoides Myrtaceae Pericalymma ellipticum Papilionaceae Jacksonia furcellata J. sternbergiana Proteaceae Adenanthos obovatus Banksia attenuata B. grandis B. ilicifolia B. menziesii Xanthorrhoeaceae Xanthorrhoea preissii Zamiaceae Macrozamia riedlei

Susceptibility in Banksia Woodland and E. marginata Forest The susceptibility grouping for plant species that were assessed both in E. marginata forest (Shearer and Dillon 1995) and Banksia woodland are compared in Tables 7 and 8. Fifty-seven percent of the species were included in the same susceptibility group in both biomes and 77% were included in the same or the next group. Major exceptions were that P. cinnamomi was isolated more frequently from recently dead plants of Lomandra sonderi and Leucopogon propinquus occurring in E. marginata forest than from those in Banksia woodland (Table 7). In addition, individuals of Dryandra nivea, Loxocarya cinerea, Patersonia occidentalis and Persoonia longifolia died in a greater proportion of disease centres in E. marginata forest than did those in Banksia woodland, whereas individuals of Hibbertia hypericoides died in a greater proportion of disease centres, and P. cinnamomi was isolated more frequently from recently dead plants in Banksia woodland than in E. marginata forest.

Discussion The 177 perennial native species recorded in the 46 disease centres were 95% and 14% of the 187 and 1313 species recorded for the Swan Coastal Plain by Dodd and Griffin (1989) and Gibson et al. (1994), respectively. The number of species found by Gibson et al. (1994) included annual and perennial species, whereas only perennial species were assessed in this study. In addition, the area surveyed by Gibson et al. (1994) included coastal plain north and south of Perth, which is much larger than the area covered by the 46 disease centres of this study south of Perth. This study assessed only areas infested with P. cinnamomi and does not represent a random sampling of the Swan Coastal Plain. For example, no disease centres were found in vegetation on the Spearwood and Quindalup dune systems (Shearer and Dillon


Proteaceae Papilionaceae

Mimosaceae

m Rest~onaceae Cyperaceae

I

1 Haemodoraceae

i I

1

2 3 4 Susceptibility group

5

Fig. 1. Percentage distribution of species among susceptibility groups for each of seven families having the largest number of plant species occurring in three or more disease centres of Phytophthora cinnamomi in Banksia woodland on the Swan Coastal Plain south of Perth.

1996). Dodd and Griffin (1989) and Gibson et al. (1994) also found that the Myrtaceae, Proteaceae and Papilionaceae were the most diverse families. Assigning levels of susceptibility in disease centres having a diversity of site characteristics, has the disadvantage of being affected by uncontrolled variation, both within and between centres, that influences plant infection by the pathogen and subsequent symptom expression. Variation in many factors, such as temperature, moisture, soil type, nutrition, microbial population, patchy occurrence of inoculum and rooting habit, influences plant infection, as outlined in a number of reviews presented by Erwin et al. (1983). Once a plant is infected, symptom expression is affected by variation in extrinsic environmental factors within and between disease centres in addition to the innate ability of the plant to contain or express disease. Determining the levels of susceptibility of inoculated plants that have been incubated under controlled conditions has the advantage of minimising variation of the environmental factors affecting infection and host response. However, assessment of susceptibility under controlled conditions suffers from the disadvantages of: (1) the difficulty of accurately reproducing environments similar to natural fluctuating conditions; (2) determining whether host response under limited controlled conditions accurately reflects normal response in disease centres; and (3) the demanding requirement for labour and equipment. Despite the lack of control of variation within and between disease centres, a measure of the relative susceptibility of hosts in the natural environment can be obtained from the consistency of their response to infection over a range of centres and a comparison with known responses under controlled conditions. Assessment within disease centres has the advantage of being able to identify plant species that have variable or inconsistent responses, and narrows down the number of plant species that would need to be tested under controlled


Table 7. Susceptibility group for plant species occurring in Phytophthora cinnamomi disease centres in Banksia woodland on the Swan Coastal Plain south of Perth and also in Eucalyptus marginata forest (Shearer and Dillon 1995) Family

Plant species

Susceptibility group Banksia woodland

Apiaceae Casuarinaceae Cyperaceae Dasypogonaceae Dilleniaceae Epacridaceae Goodeniaceae Haemodoraceae Iridaceae Mimosaceae Myrtaceae

Papilionaceae Proteaceae

Restionaceae Rubiaceae Rutaceae Xanthorrhoeaceae Zamiaceae

E. marginata forest

Platysace compressa Allocasuarina fraseriana Mesomelaena tetragona Lepidospenna angustatum Lomandra integra L. sonderi Hibbertia hypericoides Leucopogon propinquus Dampiera linearis Lechenaultia biloba Conostylis aculeata C. serrulata Patersonia occidentalis Acacia extensa A. pulchella Baeckea camphorosmae Eucalyptus calophylla Hypocalymma angustifolium H. robustum E. marginata Kennedia prostrata Dryandra nivea Hakea lissocarpha Persoonia longifolia Banksia grandis Loxocarya cinerea Opercularia echinocephala Eriostemon spicatus Xanthorrhoea preissii Macrozamia riedlei

Table 8. Cross-tabulation of the number of plant species in each combination of susceptibility groups for those species occurring in Phytophthora cinnamomi disease centres in Banksia woodland on the Swan Coastal Plain south of Perth and also in Eucalyptus marginata forest (Shearer and Dillon 1995) Banksia woodland

Group 1 Group 2 Group 3 Group 4 Group 5

E. marginata forest Group 1

Group 2

Group 3

Group 4

Group 5

8 0 0 0 0

5 2

2 0 1 0 1

1

2 0 4

0 0 1 0 2

0 0 0

1


conditions. It also gives an indication of the normal variation of host response occurring in disease centres for those species that respond to P. cinnamomi infection in a manner that makes them useful indicators of the presence of the pathogen. In addition to variation in the factors affecting disease expression, susceptibility rating is also affected by the efficiency of isolation of P. cinnamomi from infected tissue. Variation in isolation frequency is affected by sampling method in relation to rooting habit, and timing of sampling associated with host death and seasonal conditions, as discussed by Shearer and Dillon (1995). Factors influencing isolation frequency were minimised by sampling woody collar and root tissue of recently dead plants during the moist times of the year. The susceptibility groups to which plant species were assigned are simiiar to those used by Shearer and Dillon (1995) to describe the relative susceptibility of plant species in E. marginata forest to infection by P. cinnamomi. The groupings are consistent with host responses to infection being gradational between the most resistant to the most susceptible types, as demonstrated by Cahill et al. (1989) who examined the cellular and histological changes induced by P. cinnamomi. None of the hosts tested by Cahill et al. (1989) were resistant to initial infection and colonisation, but subsequent colonisation was confined in resistant hosts and progressed in susceptible hosts, resulting in associated secondary symptoms of infection. Species in Group 1 (Table 2) are probably comparatively resistant as they did not die in infested areas. Within this group, Eucalyptus calophylla is known to be moderately resistant from inoculation studies (Tippett et al. 1985; Cahill et al. 1989) and eight species were considered resistant in disease centres in E. marginata forest (Tables 7, 8). Podger (1968) observed no deaths of Dampiera linearis in disease centres in E, marginata forest, but isolated P. cinnamomi from this host in a centre in Banksia woodland. By contrast to this study, Podger (1968) isolated P. cinnamomi from Hibbertia vaginata and Acacia huegelii. Species in Group 2 (Table 3), characterised by few deaths in disease centres within which they occurred and no isolation of P. cinnamomi, could also be considered to express resistance to infection, but possibly not to the same degree as those in Group 1. Some of the deaths within Group 2 could be due to causes other than P. cinnamomi infection-Podger (1968) observed Lepidosperma angustatum dying in areas not infested by the pathogen. Phytophthora cinnamomi was not isolated from Hypocalymma angustifolium, which is in agreement with findings of Shearer and Dillon (1995) for E. marginata forest, but in contrast to those of Podger (1968) who isolated the pathogen relatively frequently from the stems of this species. Also in contrast to findings of this study, Podger (1968), isolated P. cinnamomi frequently from Calytrixflavescens at one centre, infrequently from Burtonia conferta and Persoonia longifolia, and isolated the pathogen from Xylomelum occidentale in one centre but not another. Most of the species in Group 3 (Table 4) were considered by either Podger (1968) or Hill (1990) as susceptible, and the species within this group would be useful indicators of infection. Isolation of P. cinnamomi from a low number of deaths may be due to intraspecific variation in susceptibility. Alternatively, variation in plant rooting habit, associated microflora or extrinsic environmental factors may cause variable disease expression within disease centres (Shearer and Dillon 1995). Group 4, in which species died frequently but from which P, cinnamomi was isolated infrequently, probably represents variable responses to infection. Death of some of the group may be due to causes other than infection. Although Acacia pulchella is a resistant host to P. cinnamomi (Tippett and Malajczuk 1979; Cahill et al. 1989), plants of the species died frequently in disease centres. For plant species having a deep tap root similar to that of Stirlingia latifolia (Dodd et al. 1984) or the deep sinker roots of E, marginata (Shea et al. 1982), low isolation frequency may be due to the difficulty in sampling infected tissue below the soil surface. ~ l t h o u ~sometimes h killed by P. cinnamomi, the formation of necro-


phylactic and exophylactic periderm is an important resistance mechanism in E. marginata (Tippett et al. 1983). Thus, the response to infection of the species within Group 4 probably ranges from moderately susceptible to moderately resistant. Species in Group 5 are the most susceptible as they died frequently in disease centres within which they occurred and P. cinnamomi could readily be isolated from recently dead plants (Table 6). Most of the species in this group were considered as susceptible by either Podger (1968) or Hill (1990) and would be good indicators of P. cinnamomi infestation. Although the susceptible species accounted for only 15% of the total number of species occurring in three or more disease centres, most of the susceptible species occurred frequently and were often structuraily dominant in Banksia woodland. As a consequence of infection, death of susceptible plant species results in reduced floristic diversity and a conspicuous decline in biomass (Shearer and Dillon 1996). Infected susceptible hosts would be important as a food base for inoculum production, and for the survival and spread of the pathogen in roots, similar to that observed for Banksia attenuata (Hill et al. 1994). Elimination of susceptible species by infection and the associated reduction in the food base would probably lead to a decline in P. cinnamomi inoculum, as has been observed in E. marginata forest (Shearer and Shea 1987) and forests in Victoria (Weste et al. 1973). Weste (1981) suggests that in infected sclerophyll woodland in Victoria, this may lead to a host-pathogen cyclic interaction of disease eliminating susceptible host biomass with an associated reduction in pathogen inoculum that may allow regeneration of susceptible hosts, with subsequent increase in inoculum and expression of disease. Susceptible hosts were observed to regenerate in 71% of the disease centres assessed in Banksia woodland on the Swan Coastal Plain (Shearer and Dillon 1996). However, the length of periods between disease cycles will depend on the rate of regeneration of susceptible plant species and the distribution of inoculum of P,cinnamomi within the soil profile. A number of plant species expressed a susceptibility in E. marginata forest different from that found in Banksia woodland (Tables 7, 8). This was especially the case for Hibbertia hypericoides, which died infrequently and from which P. cinnamomi was not isolated from recently dead plants in disease centres in the E, marginata forest (Shearer and Dillon 1995), by contrast to both frequent death and isolation of the pathogen from this species in centres in Banksia woodland. Differences between biomes may be due to differences in environmental influences on infection and symptom expression, as already discussed, or may reflect different geographic genetic entities included within the one species. Considerable variation in susceptibility to P. cinnamomi exists within families, similar to that observed by Wills (1993) in the Stirling Range National Park in the Southern Sandplain and Shearer and Dillon (1995) in E. marginata forest. Species of Papilionaceae and Proteaceae are well represented in the susceptible Group 5, as well as the resistant Groups 1 and 2. As noted by Shearer and Dillon (1995), a greater understanding of the variation in host response within taxa may help elucidate the basis of resistance to P. cinnamomi. There is also a need to determine the variation in susceptibility within genera that contain rare and endangered species such as Adenanthos, Darwinia, Daviesia, Dryandra, Lambertia, Leucopogon, Myrtaceae and Verticordia (Hopper et al. 1990) in order to optimise strategies for protection, seed collection and storage of germplasm of taxa threatened by extinction.

Acknowledgments Our thanks to D. Devlin and J. Webster for help in site assessment and sample processing, and R. Buehrig for arranging the database entry programs. References Cahill, D., 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.


Dodd, J., and Griffin, E. A. (1989). Floristics of the Banksia woodlands. Journal of the Royal Society of Western Australia 71, 89-90. Dodd, J., Heddle, E. M., Pate, J. S., and Dixon, K. W. (1984). Rooting patterns of sandplain plants and their functional significance. In 'Kwongan, Plant Life of the Sandplain'. (Eds J. S. Pate and J. S. Beard.) pp. 146-177. (University of Western Australia Press: Nedlands.) Erwin, D. C., Bartnicki-Garcia, S., and Tsao, P. H. (Eds) (1983). 'Phytophthora. Its Biology, Taxonomy, Ecology, and Pathology.' (The American Phytopathological Society: St Paul.) Gibson, N., Keighery, B. J., Keighery, G. J., Burbidge, A. H., and Lyons, M. N. (1994). A floristic survey of the Southern Swan Coastal Plain. Unpublished report to the Australian Heritage Commission, Department of Conservation and Land Management and the Conservation Council of Western Australia, Perth. Green, J. W. (1985). 'Census of the Vascular Plants of Western Australia.' 2nd Edn (Western Australian Herbarium: Perth.) Havel, J. J. (1975). Site-vegetation mapping in the northern jarrah forest (Darling Range). 1. Definition of site-vegetation types. Forests Department Bulletin No. 86, Perth. Hill, T. C. J. (1990). Dieback disease and other Phytophthora species in the Northern Kwongan. In 'Nature Conservation, Landscape and Recreation Values of the Lesueur Area'. (Eds A. A. Burbidge, S. D. Hopper and S. van Leeuwen.) pp. 89-97. Bulletin No. 424, Environmental Protection Authority, Perth. Hill, T. C. J., Tippett, J. T., and Shearer, B. L. (1994). Invasion of Bassendean Dune Bankria woodland by Phytophthora cinnamomi. Australian Journal of Botany 42, 725-738. Hopper, S. D., van Leeuwen, S., Brown, A. P., and Patrick, S. J. (1990). 'Western Australia's Endangered Flora and Other Plants in Need of Special Protection.' (Department of Conservation and Land Management: Perth.) Marchant, N. G., Wheeler, J. R., Rye, B. L., Bennett, E. M., Lander, N. S., and Mcfarlane, T. D. (1987). 'Flora of the Perth Region.' (Western Australian Government Printer: Perth.) Podger, F. D. (1968). Aetiology of jarrah dieback. A disease of dry sclerophyll Eucalyptus marginata Sm. forests in Western Australia. MSc Thesis, The University of Melbourne. Shea, S. R., Shearer, B., and Tippett, J. (1982). Recovery of Phytophthora cinnamomi from vertical roots of jarrah (Eucalyptus marginata Sm.). Australasian Plant Pathology 11,25-28. 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 Dillon, M. (1996). Impact and disease centre characteristics of Phytophthora cinnamomi infestations of Banksia woodlands on the Swan Coastal Plain, Western Australia. Australian Journal of Botany 44,79-90. Shearer, B. L., and Hill, T. C. (1989). Diseases of Banksia woodlands on the Bassendean and Spearwood Dune Systems. Journal of the Royal Society of Western Australia 71, 113-1 14. Shearer, B. L., and Shea, S. R. (1987). Variation in seasonal population fluctuations of Phytophthora cinnamomi within and between infected Eucalyptus marginata sites of south-western Australia. Forest Ecology and Management 21,209-230. Tippett, J. T., and Malajczuk, N. (1979). Interaction of Phytophthora cinnamomi and a resistant host Acaciapulchella. Phytopathology 69,764-772. Tippett, J. T., Shea, S. R., Hill, T. C., and Shearer, B. L. (1983). Development of lesions caused by Phytophthora cinnamomi in the secondary phloem of Eucalyptus marginata. Australian Journal of Botany 31,197-210. Tippett, J. T., Hill, T. C., and Shearer, B. L. (1985). Resistance of Eucalyptus spp. to invasion by Phytophthora cinnamomi. Australian Journal of Botany 33,409-418. Weste, G . (1981). Changes in the vegetation of sclerophyll shrubby woodland associated with invasion by Phytophthora cinnamomi. Australian Journal of Botany 29,261-276. Weste, G., Cooke, D., and Taylor, P. (1973). The invasion of native forest by Phytophthora cinnamomi. 11. Post-infection vegetation patterns, regeneration,decline in inoculum, and attempted control. Australian Journal of Botany 21, 13-29. Wills, R. T. (1993). The ecological impact of Phytophthora cinnamomi in the Stirling Range National Park, Western Australia. Australian Journal of Ecology 18, 145-159. Manuscript received 20 November 1995, accepted 28 May 1996