Occurrence of Fusarium Crown and Root Rot, a New Disease on Greenhouse Cucumbers in British Columbia, and
Methods for Disease Control

Z.K. Punja, M. Parker, S. Rose, and D. Louie

Centre for Pest Management, Department of Biological Sciences, Simon Fraser University,
Burnaby, British Columbia V5A 1S6, Canada

K. Ng

British Columbia Ministry of Agriculture, Fisheries and Food, Abbotsford, B.C. V3G 2M3, Canada

Additional index words. Cucumis sativus, Fusarium oxysporum

Abstract. A crown and root rot of greenhouse cucumbers grown in the Fraser Valley of British Columbia, Canada, has been observed sporadically since 1994, and has increased in frequency and severity. Affected plants wilted at the fruit-bearing stage when ambient temperatures exceeded 27 to 30 °C. Mycelial growth and pronounced-orange spore masses were visible at the crown of the plants, and the pathogen could be isolated from the roots and crown tissue, and from cortical and vascular tissues at distances of greater than 40 cm from the crown. Among several species belonging to the Cucurbitaceae, as well as pepper and tomato, that were artificially inoculated with conidia using a root dip method, Cucumis melo, Cucurbita pepo, Citrullis vulgaris, and Luffa aegyptiaca developed crown and root rot symptoms similar to that on cucumber. Reactions of 25 cucumber cultivars tested ranged from highly susceptible to moderately tolerant to this disease. The most susceptible Long English cultivars were Flamingo, Mustang and Serami. The pathogen has been identified as Fusarium oxysporum, and symptomology and host range studies indicate that this is a new formae specialis, f.sp. radicis-cucurbitacearum (F.o.r.c.). RAPD analysis using eight primers revealed that the new pathogen was distinct from F.o. f.sp. cucumerinum; some intraspecific variation was observed within isolates of F.o.r.c. from B.C. The optimal temperature for growth of three isolates from B.C. on APDA was 24 °C. Pathogenicity experiments revealed that wounding of roots was necessary for disease development, and that various stresses, e.g., drought stress, enhanced disease severity. Disease developed more rapidly at 25 than at 17 °C. The pathogen was recovered on Komada's medium at a high frequency (105 cfu/cm3) from the growing substrate (rockwool blocks and sawdust) and from overdrain water, but at a very low frequency as airborne propagules. The initial sources of inoculum for this disease may be from infested seed or the growing medium, and wounding of roots and subsequent plant stress promote disease development. Following seed inoculation, 15 spores/seed were shown to be sufficient to initiate disease on muskmelon and cause seedling damping-off. This disease has presently only been reported from Greece and B.C., but has the potential to spread to other areas and affect both cucumber and muskmelon, as well as other cucurbits. Disease control measures should include sanitation, use of tolerant cultivars, and pathogen-free seed. Evaluation of potentially antagonistic microorganisms, e.g., Pseudomonas bacteria, added to the growing substrate has demonstrated that there is potential for biocontrol through competition and reduction of infection. Amendment with chitin and digested chitin did not reduce the disease.

A crown and root rot of Long English cucumbers grown in greenhouses in the Fraser Valley of British Columbia was observed sporadically during 1994 and 1995, and has now increased in incidence and severity. Up to 10% of the plants in a greenhouse may be affected, and eight commercial greenhouses have

currently reported losses in yield. Affected plants wilt at the fruit-bearing stage, especially when ambient temperatures exceed 27 to 30 °C. Mycelial growth and pronounced orange spore masses are visible at the crown of the plants, within which Fusarium spores are abundant.


Cucurbitaceae '98

From the descriptions of fungal diseases de scribed in published compendia, none accurately fit the symptomatology of this disease (Blancard et al., 1994; Howard et al., 1994). However, a root and stem rot of cucumber described in 1996 from Greece (19) closely resembled the disease observed in B.C. The causal organism was described by Vakalounakis as Fusarium oxysporum formae specialis radicis-cucumerinum (Vakalounakis, 1996).

The objectives of this study were to 1) establish the identity of the pathogen, to study the host range and infection behavior; 2) establish the extent of pathogen spread; and 3) evaluate the effects of cucumber cultivar, pathogen-infested seed, and amendments to the growing medium using chitin and plant growth promoting rhizobacteria on disease development under B.C. growing conditions.

Materials and methods

Source of isolates. Cucumber plants with wilting symptoms and visible infection and mycelial growth and sporulation at the crown were obtained from eight commercial greenhouses located throughout the Fraser Valley of B.C. Infected plants were collected during March to August 1996 and 1997 and brought back to the laboratory. Isolations were made by direct-plating of small (1-mm2) pieces of tissue bearing mycelium or spore masses from the crown area onto potato dextrose agar (Difco Laboratories) containing 200 mg·L­1 of either ampicillin or penicillin to inhibit bacterial growth. Petri dishes were incubated on the laboratory bench (ambient temperature range

of 22 to 25 °C) for 10 d, at which time hyphal-tip transfers were made from the colony margin to fresh PDA plates containing antibiotics. Isolates were designated with different numbers when they originated from different plants and/or greenhouses. A total of 30 isolates were collected over 2 years.

Morphology and growth in culture. Morphological characteristics of the pathogen were determined from three to four 4-week-old PDA colonies and from sporulating colonized host tissues. The distinguishing microscopic features of F. oxysporum, which included shape of microconidia, occurrence of microconidia in false heads, presence of short monophialides and development of chlamydospores (Nelson et al., 1983) were noted. Unique colony characteristics, such as presence of a blue-violet pigment, microsclerotia, and orange sporodochia, were also recorded. Morphological comparisons were made with cultures of other Fusarium species originating from cucurbits (Table 1), including F. oxysporum f.sp. cucumerinum from cucumber (provided by M. Gerlagh, Institute for Plant Protection, Wageningen, The Netherlands), F. oxysporum f.sp. melonis from muskmelon (provided by T.J. Gordon, University of California, Davis), F. oxysporum f.sp. radicis-cucumerinum from cucumber (provided by D. Vakalounakis, Plant Protection Institute, Crete, Greece) and F. solani f.sp. cucurbitae from squash (provided by T. Zitter, Cornell University, Ithaca). A culture of the isolate from B.C. was sent to Dr. Keith Seifert (National Identification Service, Ag

Table 1. Isolates of Fusarium oxysporum included in this study.

Isolate Identity Source

4 F.o. f.sp. radicis-lycopersici Tomato, J. Elmhirst, Abbotsford, B.C. (ATCC 52429)

13 F. oxysporum Soil (saprophytic)

15 F.o. f.sp. radicis-cucurbitacearum Cucumber, this study

16 F.o. f.sp. radicis-cucurbitacearum Cucumber, this study

19 F.o. f.sp. radicis-cucurbitacearum Cucumber, this study

20 F.o. f.sp. radicis-cucurbitacearum Cucumber, this study

21 F.o. f.sp. radicis-cucurbitacearum Cucumber, this study

21A F.o. f.sp. radicis-cucurbitacearum Cucumber, this study

22 F.o. f.sp. radicis-cucurbitacearum Cucumber, this study

23 F.o. f.sp. radicis-cucurbitacearum Cucumber, this study

28 F.o. f.sp. radicis-cucurbitacearum Cucumber, this study

60B F.o. f.sp. radicis-cucumerium Cucumber, D. Vakalounakis, Greece

56 F.o. f.sp. cucumerinum (race 1) Cucumber, ATCC16416

56A F.o. f.sp. cucumerinum Cucumber, M. Gerlagh, The Netherlands

Cucurbitaceae '98

Table 2. Response of various plant species to inoculation with Fusarium oxysporum isolated from cucumber.

Plant Cultivar Disease

species or line ratingz

Pepper (Capsicum annum L.) Valencia 0

Tomato (Lycopersicon esculentum Mill.) Dombito 0

Greenhouse 761 0

Red Giant 0

Vitador 0

Cucumber (Cucumis sativus)

var. sativus Flamingo 3

Mustang 3

Serami 3

Marketmore 86 3

var. hardwickii LJ 90430 3

Gourd (Luffa aegyptiaca Mill.) PI 163295 1

Fletcher 0

Muskmelon (Cucumis melo L.) Caravelle 4

Earligold 4

Impac 4

Magnum 4

Summet 4

Squash (Cucurbita pepo L.) Dixie 1

Watermelon (Citrullis vulgaris)

Eckl. & Zeyh Madera 2

Zucchini (Cucurbita pepo L.) Black Beauty 0

zDisease rating scale is described in the text.

For the host range studies, a minimum of eight replicate plants of each species was included, and the experiment was repeated twice. Seeds were planted into sterilized Sunshine Mix growing medium (Sungro Horticulture Canada Ltd., Delta, B.C.) in 10-cm2 pots and maintained under supplemental light (12-h photoperiod) either in a growth room or under greenhouse conditions, with an ambient daytime temperature range of 20 to 25 °C. When the plants were 2 to 3 weeks old or had attained a minimum height of 10 cm, depending on the plant species, they were gently uprooted, the soil was washed off the roots, and the plants were inoculated using a root-dip inoculation procedure. Washed roots were blotted dry on paper towels and immersed for 5 min in a spore suspension containing 105 spores/mL (comprising both macro- and microconidia of isolate 16). Controls included plants that were similarly uprooted but immersed in distilled water. The plants were then replanted into Sunshine Mix following inoculation and maintained under the conditions described above for 30 days. Disease evaluations were made according to the following scale: 0 = all plants healthy; 1 = 5% to 25% mortality; 2 = 25% to 50% mortality; 3 = 50% to 75% mortality; 4 = >75% mortality. The crown and root tissue from diseased plants selected at random, as well as from surviving plants, was surface-sterilized in 0.625% NaOCl for 3 min, rinsed in sterile distilled water, and plated onto PDA containing 200 mg·L­1 ampicillin. The recovery of Fusarium colonies was recorded after 2 weeks.

For cucumber cultivar evaluations, the same inoculation procedure described previously for the host range study was used. A minimum of five replicate plants was included for each cultivar, and the experiment was repeated thrice over an 8-month period. Two weeks after seeds were planted into Sunshine Mix medium, seedlings were uprooted, inoculated, and replanted. The cultivars were grown under greenhouse or growth chamber conditions as described previously for a period of 28 days. Disease assessments were made at two weeks and four weeks after inoculation. A disease severity index (DSI) was calculated as follows:

(% plants dead, 14 d)/14 + (% plants dead, 30 d)/30 + % reduction in height of survivors

riculture and Agri-Food Canada, Ottawa) to confirm the species identification as F. oxysporum.

Colony growth of four isolates (three from B.C. and one from Greece) at eight different temperatures was determined after seven days of growth on PDA from three replicate dishes per isolate.

Host range and cultivar evaluations. The plant species evaluated for susceptibility to Fusarium included members of the Cucurbitaceae (Table 2) as well as pepper and tomato. A total of 25 cucumber (Cucumis sativus L.) cultivars were also evaluated, consisting mostly of the Long English type, although a few fresh market and pickling cucumber cultivars were also included. Five muskmelon (Cucumis melo ssp. melo Cantalupensis Group) cultivars were also tested. Seeds of the cucumber cultivars were obtained from various commercial seed companies, while those of muskmelon were donated by Dennis Lawn (Seminis, Saticoy, Calif.). Seeds of noncultivated Cucurbitaceae originated from the seed repository at North Carolina State University (provided by T.C. Wehner, NCSU, Raleigh, N.C.).

Cucurbitaceae '98

Table 3. Recovery of Fusarium from infected cucumber tissues at two stages of infection.z



infected Stage Ay Stage By

crown (cm) Cortical Vascular Cortical Vascular

0 ++ ++ ++ ++

5 + ++ ++ ++

15 + + ++ ++

25 ­ + ++ ++

35 ­ ­ ++ ++

45 ++ ++

55 + ++

65 ­ +

75 ­ ­

zCross-sectional pieces were removed at designated distances along the main stem and cortical and vascular tissue pieces were plated onto Komada's medium and incubated for 10 days at 22 °C.

yPlants with stages of infection A and B were characterized by A = early colonization of crown, no foliar symptoms and B = advanced colonization of crown, plants wilting.

removed from the sawdust and transported to the laboratory. In addition, samples of the sawdust growth medium and overflow water dripping out of the bags were taken. The stems were cut into 10-cm-long segments, which were dipped in 0.625% NaOCl for 3 min, and then small cross-sections of cortical and vascular tissues were excised at a point midway along the segment and plated onto Komada's selective medium (1995). Petri dishes, each containing six to eight tissue pieces per segment were incubated at 22 to 25 °C for 2 weeks, at which time the appearance of colonies of Fusarium was recorded for each segment.

To quantify the extent of sporulation, samples of the crown tissue (1 cm2) and rockwool blocks (1 cm3) with adhering primary roots were placed in test tubes containing 10 mL of distilled water and shaken at 200 rpm for 5 min. Duplicate samples were transferred to a haemocytometer and spore counts were made.

Segments of secondary roots with small lateral roots were gently removed from the underside of the rockwool blocks, washed under running tap water for 2 to 3 h, and placed directly onto Komada's medium. The appearance of colonies along the root system was rated after 5 days of incubation. The sawdust and water samples were diluted 10-fold in distilled water and 0.5 mL was plated onto Komada's medium. Colonies were recorded after 2 weeks of incubation.

To determine if inoculum produced on infected crown and stem tissues was disseminated aerially, petri dishes containing Komada's medium were suspended vertically and taped onto wooden stakes at various heights (from 0.2 to 2.0 m) from the ground, which were placed at lateral distances of 0.2 and 0.4 m away from severely diseased plants. The lids were removed and dishes were left for periods of 12 to 24 h and then incubated in the laboratory for two weeks to determine if colonies developed.

Role of seed transmission. To establish whether inoculum on seed could initiate disease, muskmelon 'Impac' or 'Summet' was used. Varying concentrations of spore suspensions of isolate 16 (from B.C.) and 60 B (from Greece) were prepared (1 x 102 to 1 x 105 spores/mL), into which preweighed lots of 10 seeds were immersed in petri dishes for 5 min. The inoculum was poured off, the seeds were blotted to remove excess moisture, and reweighed. The difference between preinoculation and postinoculation masses was

where % reduction in height of survivors was relative to the control.

The data from the four experiments were averaged to provide a mean DSI for each cultivar that was used to provide a rank of the susceptibility of the cultivars (Table 2). In each experiment, a known susceptible and tolerant cultivar were included as standards. Controls included plants uprooted and dipped in water. Isolations from diseased plants were made as previously described to confirm that Fusarium was the causal agent.

To determine the effect of temperature on disease development, 'Corona' cucumber was inoculated as described above and 10 replicate pots each were placed at 25 or 17 °C in controlled environment chambers for 28 days and a DSI was determined for each temperature. The experiment was repeated twice.

Pathogen development and spread. To establish the extent to which pathogen colonization occurred on the crown and root tissues, as well as the extent of spread from diseased tissues, several experiments were conducted. Naturally infected plants displaying different stages of disease development, from the appearance of initial infection at the crown (Figure 1a) to extensive colonization and sporulation (Figure 1c) and wilting of plants, were selected. A minimum of two plants at each stage was obtained. Stems up to 100 cm in length were excised at the crown, and the rockwool cubes with adhering roots were carefully

Cucurbitaceae '98

Figure 1. Symptomology of Fusarium crown and root rot on greenhouse cucumbers. a) Wilting of infected plant during hot weather. b) Early infection of crown showing yellowish-orange discoloration. c) Advanced infection showing extensive colonization of stem. d) Close-up of infected stem with masses of orange sporodochia. e) Infection at root tips and root junctions by Fusarium. f) Colonies of Fusarium on PDA showing prolific production of spores in sporodochia at colony margins.

used to estimate the number of spores per seed at each of the concentrations tested. The seeds were allowed to air-dry for five days and then planted into sterilized Sunshine Mix medium. Plants were maintained in a growth room with supplemental lighting (12-h photoperiod) and temperature range of 22 to 25 °C. Percentage plant mortality was rated using the 0 to 4 scale described in the host range study.

Assessment of genetic variation using rapd markers. Nine isolates were used to determine the relationship of the Fusarium isolated from cucumber in this study to those provided by other investigators (Table 1), as well to establish the extent of genetic variation among the isolates from B.C. The isolates were all maintained as single-spore colonies before use. Cultures were grown on PDA containing 200 mg·L­1 streptomycin sulfate for two weeks. Mycelium and spores were scraped off the agar surface and total DNA was extracted following the method described by Urquhart et. al. (1997). RAPD analysis was initially conducted with a subset of the isolates to screen and select primers that yielded polymorphic banding patterns. A set of eight primers from the University of B.C., Biotechnology Lab. NAPS Unit was subsequently used with all 14 isolates. The primers used were 105,438,489,825,834,857,873, and 890. The RAPD­PCR conditions and data analysis were conducted as described by Urquhart et. al. (1997).

Evaluation of disease control methodsgrowth chamber and pot studies. Amendment of growth medium with crab/shrimp shells. The effect of crushed crab/shrimp shells on growth of 'Corona' cucumber plants and on development of Fusarium crown rot and survival of the pathogen was determined. The shells were a by-product of the local shell fish processing industry (provided by International Chitin, Richmond, B.C.). Both raw unprocessed crushed flakes and partially digested material (following enzymatic digestion with bacterial chitinases to produce a viscous paste) were used. These amendments were added to sterilized Sunshine mix growth medium at a rate of 4% (v/v) and the medium was moistened to saturation and allowed to incubate for three days at room temperature. Inoculum of Fusarium was added at the rate of 20 mL of 105 spores/mL

to 250 cm3 of medium 3 days after amendment with the crab/shrimp shells.

'Corona' cucumber seeds were pregerminated by incubating them on moistened filter paper for five days and the planted into growth medium that had been amended with crab/shrimp shells and also received inoculum of Fusarium, as described above. Eight replicate pots per treatment were used and seedlings were maintained in a growth room at 22 to 25 °C with supplemental lighting. Roots were wounded one week later by inserting a spatula into the growth medium at five arbitrary locations. Measurements of plant growth, including height, leaf area, fresh mass, and stem diam were made after 24 days of growth. Disease incidence was expressed as the percentage of plants dead or wilting. The experiment was repeated twice.

Evaluation of rhizobacteria. The effect of two plant growth promoting rhizobacteria on growth of muskmelon 'Summet' plants and development of Fusarium crown rot was determined. The bacteria tested were Burkholderia cepacia (Ral 3) and Pseudomonas fluorescens (G3-28), both formulated in liquid suspension at 1010 cells/mL by Agrium Biologicals Inc. (provided by E. Pederson, Agrium, Saskatoon, SK). The bacteria were applied to seeds either before or following inoculation with Fusarium. Seeds were inoculated with 103 spores/mL to provide an extrapolated density of 15 spores/seed as described previously. The bacteria were applied one day before or after Fusarium inoculation by immersing seeds in a suspension of 108 cells/mL for 5 min and then air drying for 1 day before planting into Sunshine mix. The seedlings were maintained in a growth room at 22 to 25 °C with supplemental light for 30 days. The extent of disease development was expressed as percentage plant mortality.

Evaluation of disease control methodslarge scale greenhouse trials. Trial 1. Amendment of growth medium with crab/shrimp shells and rhizobacteria. The trial was conducted in a semi-commercial size greenhouse located in Abbotsford, B.C. with built-in temperature and humidity control. The growing substrate was rockwool cubes in which seeds of cultivar Odessa were directly germinated. After two weeks, the cubes were placed into bags

Cucurbitaceae '98

The chitin was mixed into the sawdust around the base of the rockwool cubes, while the bacterial suspension was poured around the base of the cube and mixed in. Noninoculated bags received 100 mL of distilled water, which was mixed into the sawdust. Twice weekly, the cucumber fruit was harvested from each plant, and the commercial grade and fresh mass (kg) of each fruit was recorded. The data on cumulative fruit number and yield in Fusarium-inoculated and noninoculated plots were analyzed for significant differences in response to treatment by ANOVA and means separation was achieved using the Tukey-Kramer hsd test at P = 0.0.5.

Trial 2. Amendment of growth medium with rhizobacteria. Seedlings were initiated as described in Trial 1, and before transplanting, bacterial inoculum was added in December 1997, as previously described. There were eight replicate bags for each treatment, with two plants in each bag. Four weeks later, inoculum of Fusarium was applied following the procedure described in Trial 1. A second application of the bacteria was made 4 weeks later, while control plants received distilled water. Fruit number and yield were measured as described previously.

Results and discussion

Symptomology and pathogen description. Infected plants displayed wilting symptoms (Figure 1a) and there was a yellow-orange discoloration with visible mycelial growth at the base (Figure 1b). Over time, fungal growth progressed up the stem, where at advanced stages of infection, orange masses of spores were produced (Figure 1 c and d). Destruction of the cortical tissues resulted in a stringy appearance to the stem. Plating of root segments onto Komada's medium revealed sites of infection originating at the root tips and at root junctions (Figure 1e). On PDA, the Fusarium colonies had a pinkish color, and within 2 to 3 weeks, oranges masses (sporodochia) developed at the colony margins (Figure 1f). The sporodochia contained characteristic macroconidia of Fusarium (Figure 2a) as well as microconidia, which were borne in false heads. On infected host tissue, macroconidia and characteristic short phialides could be seen (Figure 2b). When cultures were 2 to

Figure 2. (a)Macroconidia and microconidia of Fusarium from 3-week-old PDA culture. (b) Macroconidia and phialides produced on host stem.

containing hemlock sawdust. The following treatments were applied to the sawdust in December 1996, prior to transplanting: chitin flakes or partially digested chitin at 1% (v/v); one of P. fluorescens or B. cepecia (200 mL of 108 cells/mL), nonamended control. All amendments were mixed into the sawdust. Eight replicate bags were assigned to each treatment (each with two plants) and these were randomized within a split-plot design. Plants were fertilized using a recirculating hydroponic nutrient solution and trellised according to standard commercial practices (Anonymous, 1996). Four weeks after treatments were applied, inoculum of Fusarium was applied to four out of the eight replicates (block 1), with the remaining four bags serving as noninoculated amendments (block 2). The inoculum was applied using a syringe to the base of the rockwool cube of each plant (50 mL of 106 spores/mL). Four weeks after inoculation, a second treatment of the amendments was made.

Cucurbitaceae '98

was Cucumis melo, followed by C. sativus vars. sativus and hardwickii. Some plant mortality and stunting and yellowing of leaves was observed on Cucurbita pepo, Citrullis vulgaris, and Luffa aegyptiaca. The pathogen was recovered from the roots of all infected species when plated onto Komada's medium. These results are consistent with the findings of Vakalounakis (1996), who reported disease development on five cucurbit species in some of the pathogenicity tests conducted, although C. pepo was not reported to develop disease. Our pathogenicity tests suggest that the host range of the F. oxysporum causing crown and root rot is more extensive than that previously reported (Vakalounakis, 1996). Additional members of the Cucurbitaceae will need to be evaluated to determine the potential host range of this pathogen. Given that at least five members are shown to be susceptible, and in accordance with the argument presented by Gerlagh and Blok (1988), we propose that the pathogen be named F. oxysporum f.sp. radicis-cucurbitacearum, which differs slightly from the combination f.sp. radicis-cucumerinum proposed earlier (Vakalounakis, 1996).

The extreme susceptibility of all cultivars of muskmelon tested suggest that this pathogen has the potential to become a threat to this crop, causing pre- and postemergence damping-off, and crown rot of infected seedlings.

When the disease severity index of 'Corona' was rated in response to temperature, it was found that infection rate and severity were greater at 25 °C (DSI = 4) compared to 17 °C (DSI = 2). Similar results were obtained with 'Summet' muskmelon. These findings are in contrast to those reported earlier from Greece (Vakalounakis, 1996), where infection was higher at 17 °C. The optimal temperature range for growth of all of the isolates in vitro was, however, 24 to 27 °C. The majority of diseases caused by F. oxysporum on a broad range of crops are reported to be favored by warm temperatures (above 25 °C) (Agrios, 1997; Cook and Baker, 1983). Our results are consistent with this general observation. The onset of disease development under commercial greenhouse conditions is also consistent with the requirement for high temperatures.


Figure 3. Effect of temperature on colony growth of four isolates of F. oxysporum f.sp. radicis-cucurbitacearum on PDA. Data were recorded after 7 days.

3 months old, blue-black microsclerotia developed and the colonies had a purplish tinge when viewed from the underside. All of these morphological features are distinctive of Fusarium oxysporum (Nelson et al., 1983) and the identity was confirmed by K. Seifert. The description of the fungus provided by Vakalounakis (1996) is also consistent with our observations. Comparisons with an isolate obtained from diseased cucumbers with root and stem rot from Greece revealed that the isolates from B.C. were morphologically identical to that isolated from Greece, but differed from other Fusarium species from cucurbits that were included for comparison. Colony growth of three isolates from B.C. and one from Greece (60 B) at eight temperatures revealed the optimal temperature range of all isolates to be 24 to 27 °C (Figure 3).

Host range and cultivar evaluations. The responses of the plant species tested to root-dip inoculation with F. oxysporum isolate 16 are summarized in Table 2. Pepper and tomato did not develop symptoms. The most susceptible species

Cucurbitaceae '98

Dispersal by water, however, is a likely means of spread within the greenhouse, in addition to providing a source of inoculum for contamination of the growing medium. In other diseases caused by F. oxysporum, e.g., on basil, tomato, spores have been reported to be disseminated aerially (Gamliel et al., 1996; Katan et al., 1997; Rowe and Farley, 1981).

The distribution of infected plants within a greenhouse tended toward randomness, and there were no instances of two adjacent plants showing symptoms unless they were grown in the same bag, suggesting the pathogen had spread via root contact. This is consistent with the observed lack of secondary spread of inoculum, indicating that the disease is single-cycle and that primary inoculum early during the growing season is most important for disease development.

Role seed transmission. The application of spores of Fusarium to seed of 'Impac' or 'Summet' muskmelon at varying concentrations resulted in extrapolated spore densities of 15, 150, 1500 spores/seed. At these levels, the incidence of plant mortality caused by isolate 16 was 50, 100, and 100%, respectively. Isolate 60B caused 15, 75 and 100% plant mortality, suggesting it may be less virulent than isolate 16. These differences may account for the variation in results from host range studies discussed earlier.

Assessment of genetic variation using RAPD markers. The use of the eight-primer set adequately distinguished isolates of F.o.r.c. from F.o. f.sp. cucumerinum (two isolates) as well F.o. f.sp. radicis-lycopersici and a saprophytic F. oxysporum (Figure 4). Within F.o.r.c., there was some variation among isolates obtained from different greenhouses and in some cases from different plants (Figure 5). Interestingly, the isolate from Greece (60 B) showed a very close relationship to the B.C. isolates. These findings illustrate that the level of genetic variation currently present within F.o.r.c. overlaps across broad geographic regions.

Evaluation of disease control methods

Figure 4. RAPD banding patterns obtained with two primers. For isolate identification, refer to Table 1.

Pathogen development and spread. The pathogen was recovered from diseased plants to different extents depending on the stage of infection (Table 3). At early stages, the fungus was recovered up to 25 cm distal to the crown, while at advanced stages, Fusarium was recovered from the cortical and vascular tissues up to 65 cm from the crown (Table 3). This finding suggests that while the external sporulation may be localized to the crown and adjacent tissues, internal colonization can be more extensive, although the pathogen was not localized to just the vascular tissue, thus distinguishing it from Fusarium wilt caused by F.o. f.sp. cucumerinum (Blancard et al., 1994; Howard et al., 1994).

The extent of sporulation was estimated to be about 5 x 107 spores (macro and microconidia)/cm2 of crown tissue at the advanced stages of disease development (Figure 1d). On the rockwool blocks, Fusarium populations were around 1 ¥ 105/cm3. On the root system, colonies were readily observed developing on Komada's medium and originated from root tips and at root junctions (Figure 1e). The pathogen was also recovered from the sawdust-growing medium and from water samples draining out of the bags. However, on petri dishes suspended at various distances from diseased plants, colonies only developed when dishes were placed closest to the plant (0.2 m). At all other distances, no colonies were recovered. These findings suggest that while diseased tissues support profuse sporulation, there is negligible airborne inoculum, perhaps due to the slimy compact nature of the sporodochia.

Cucurbitaceae '98

1995), which would account for the marked growth stimulation observed in this study by providing an additional source of nitrogen to the plants. However, the incidence of Fusarium was increased. Fertilization with ammonium nitrogen has been shown to increase the susceptibility of crop plants to diseases caused by Fusarium (4). Thus, the benefits of chitin amendment to soil for control of other fungal diseases (Hampson and Coombes, 1991, 1995; Mitchell and Alexander, 1962) were not observed on cucumber in this study.

Application of either bacterial species to seed of muskmelon following treatment with Fusarium spores did not increase survival of seedlings. However, when bacteria were applied one day prior to application of Fusarium spores (15 spores/seed), the survival of seedling was increased by 30%. When inoculum of Fusarium was increased to 150 spores per seed or higher, the bacteria had no effect. These results indicate that the bacteria could provide protection against Fusarium infection if applied as a protective treatment and provided the inoculum density of the pathogen is not excessive. These bacteria have been shown to reduce diseases due to a number of fungal pathogens on a broad range of crop species through several mechanisms, including occupation of infection sites and production of antibiotic substances (Punja, 1997).

Figure 5. Dendrogram derived from cluster analysis of RAPD data depicting the relationship of F.o.r.c. isolates to other formae specialis of Fusarium. For isolate identification, refer to Table 1.

growth chamber and pot studies. The response of cucumber plants to amendment with crab/shrimp shells and inoculation with Fusarium is summarized in Table 4. The overall growth of the plants was significantly increased by the amendment, including plant height, fresh mass and leaf area. Chemical analysis of the shells revealed that the N content was 6.4% and the Ca content was 2%. During the breakdown of chitin contained in the crab/shrimp shells in soil, release of NH3 is reported to occur (Hampson and Coombes, 1991;

Table 4. Effect of crab/shrimp shell amendment to growth medium on cucumber (Cucumis sativus cv. Corona) growth and incidence of Fusarium crown rot.z

Plant Stem Fresh mass Total leaf Fresh mass Disease

Treatment height (cm) diam (cm) top(g) area (cm2) roots (g) incidence (%)

None 12.5 b 0.4 b 7.2 b 250 b 3.6 a 51 b

Crab/shrimp shellsy 18.5 a 0.55 a 15.3 a 529 a 3.0 b 86 a

zData were recorded after 24 days of growth at 22 to 25 °C. Means followed by different letters are significantly different according to the Tukey-Kramer HSD test at P = 0.05.

yPartially digested chitin was added at a rate of 4% (v/v) 3 days before pregerminated seeds were planted into the medium.

Cucurbitaceae '98

Figure 6. Yield response of cucumber inoculated with F.o.r.c. and subjected to treatments with crab/shrimp shell chitin or one of two bacterial strains (Ral 3 and 63-28) (Trial 1).

Evaluation of disease control methodslarge scale greenhouse trials. The cumulative and total fruit number and yield were significantly reduced in Fusarium-inoculated plants (Figure 6). Amendment of the growing medium with crab/shrimp shells or paste did not improve the yield of inoculated plants. Following addition of the two bacterial strains, a significant yield response was only obtained with B. cepacia Ral 3 (Figure 6), which increased yield to that of the noninoculated control.

In Trial 2, both bacterial strains showed a positive response and yield was significantly enhanced over that in the Fusarium-only plants (Figure 7). These results illustrate the potential of these bacterial biological control agents to reduce Fusarium crown and root rot. Among other treatments tested against this disease, fungicide application to muskmelon seed (captan + thiram) has also been shown to be extremely effective in pre

venting infection. A combination of sanitary practices (Menzies and Bélanger, 1996), use of disease tolerant cultivars, and pathogen-free or treated seed should prove to be effective against this disease.

Literature cited

Agrios, G.N. 1997. Plant pathology. 4th ed. Academic Press, San Diego, Calif.

Anonymous. 1996. Greenhouse vegetable production guide for commercial growers. British Columbia Ministry of Agriculture, Fisheries and Food, Victoria, B.C.

Blancard, D., H. Lecoq, and M. Pitrat. 1994. A colour atlas of cucurbit diseases. Observation, identification and control. Manson Publishing Ltd., London .

Cook, R.J. and K.F. Baker. 1983. The nature and practice of biological control of plant pathogens. Amer. Phytopathol. Soc., St. Paul, Minn.

Gamliel, A., T. Katan, H. Yunis, and J. Katan. 1996. Fusarium wilt and crown rot of sweet basil: Involvement of soilborne and airborne inoculum. Phytopathology 86:56­62.

Cucurbitaceae '98

Figure 7. Yield response of cucumber inoculated with F.o.r.c. and subjected to treatment with bacteria (Ral 3 and 63-28) (Trial 2).

Gerlagh, M. and W.J. Blok. 1988. Fusarium oxysporum f.sp. cucurbitacearum n.f. embracing all formae specialis of F. oxysporum attacking Cucurbitaceous crops. Neth. J. Plant Path. 94:17­31.

Hampson, M.C. and J.W. Coombes. 1991. Use of crabshell to control potato wart in Newfoundland. Can. J. Plant Pathol. 13:97­105.

Hampson, M.C. and J.W. Coombes. 1995. Reduction of potato wart disease with crushed crabshell: Suppression or eradication? Can. J. Plant Pathol. 17:69­74.

Howard, R.J., J.A. Garland, and W.L. Seaman (eds.). 1994. Diseases and pests of vegetable crops in Canada. Can. Phytopathol. Soc.­Entomol. Soc. Can.

Jarvis, W.R. 1988. Fusarium crown and root rot of tomatoes. Phytoprotection 69:49­64.

Katan, T., E. Shlevin, and J. Katan. 1997. Sporulation of Fusarium oxysporum f.sp. lycopersici on stem surfaces of tomato plants and aerial dissemination of inoculum. Phytopathology 87:712­719.

Komada, H. 1995. Development of a selective medium for quantitative isolation of Fusarium oxysporum from natural soil. Rev. Plant Prot. Res. 8:114­125.

Menzies, J.G. and R.R. Bélanger. 1996. Recent advances in cultural management of diseases of greenhouse crops. Can. J. Plant Pathol. 18:186­193.

Mitchell, R. and M. Alexander. 1962. Microbiological processes associated with the use of chitin for biological control. Proc. Soil Sci. Soc. Amer. 26:556­558.

Nelson, P.A., T.A. Tousson, and W.F.O. Marasas. 1983. Fusarium species. An illustrated manual for identification. Pa. State Univ. Press.

Punja, Z. K. 1997. Comparative efficacy of bacteria, fungi and yeasts as biological control agents for diseases of vegetable crops. Can. J. Plant Pathol. 19:315­323.

Rowe, R.C. and J.D. Farley. 1981. Strategies for controlling Fusarium crown and root rot in greenhouse tomatoes. Plant Dis. 65:107­112.

Urquhart, E.J., L.J. Sun, and Z.K. Punja. 1997. Identification of species of Tilletiopsis using random amplified polymorphic DNA analysis. Can. J. Plant Pathol. 19: 380­389.

Vakalounakis, D.J. 1996. Root and stem rot of cucumber caused by Fusarium oxysporum f.sp. radicis-cucumerinum f.sp. nov. Plant Dis. 80:313­316.

Cucurbitaceae '98