Integrated Management of Melon Vine Decline Caused by Monosporascus cannonballus Using Grafting and Chemical Treatments Applied Pre- and Postplanting

M. Edelstein, R. Cohen, and S. Shriber

Department of Vegetable Crops, A.R.O., Newe Ya'ar Research Center, P.O.B. 1021,
Ramat Yishay 30095, Israel

S. Pivonia

Arava Research and Development, Sapir Center, Israel

D. Shtienberg

Department of Plant Pathology, the Volcani Center, A.R.O., Bet Dagan 50250, Israel

Abstract. Vine decline of melons induced by Monosporascus cannonballus is a worldwide problem, mainly in arid and semiarid regions. Soil disinfestation by fumigation with methyl bromide (M.Br.) is a common and very effective treatment. However, since M.Br. is expected to be banned from use within the next 5 to 10 years, alternative measures for disease management are urgently needed. In this study the efficacy of M.Br. at reduced rate (15 g·m­2 instead of 50 g·m­2), grafted plants and post application of the fungicide fluazinam, alone or in combination, was examined. Experiments were conducted in the Arava region in southern Israel in naturally infested fields. First, various rootstock species were tested in three experiments. Of the eight rootstocks evaluated, 'Brava' (Cucurbita maxima) was selected to serve as a rootstock in the subsequent experiments due to its effects on wilt reduction (significant [63% to 100%] reduction in two out of three experiments, insignificant [35%] reduction in one experiment) and its horticultural performance. Then, the efficacy of the various control measures was evaluated in two other experiments. Reduced M.Br. rate was effective and suppressed wilt significantly, but results were variable (32% and 91% reduction). Grafting, when applied as a sole means of control (in plots not treated with M.Br. or with fluazinam), significantly reduced wilt incidence (by 84% and 87%) compared with untreated plots, and control efficacy did not differ significantly from plots treated with the full rate of M.Br. Postplanting application of fluazinam significantly reduced wilt incidence (by 88%) when applied to nongrafted plants planted in plots not treated with M.Br. Effects of the treatments on yield resembled those mentioned on wilt incidence and the yield component mostly affected by the treatments was the number of fruits per unit area. Integration of at least two of the control measures provided adequate and reliable solution to vine decline of melon: wilt incidence was always significantly lower (75% to 100% reduction) than that in untreated plots and control efficacy did not differ significantly from that observed by application of M.Br. at full rate. Since control efficacy achieved by each of the measures applied solely may vary, it is preferable to practice integrated application.

Vine decline of melons (Cucumis melo L.), also known as melon collapse or sudden wilt, is a worldwide problem, mainly in arid and semiarid regions (Martyn and Miller, 1966). This disease is a major problem in the arid Arava region of southern Israel, which is the main out-of-season melon production area for the local market and for export. The vine decline disease causes rapid wilt of plants bearing mature fruits and may result in total loss of the crop. Field

trials and inoculation experiments conducted by Reuveni et al. (1983) in the early 1980s showed that Monosporascus eutypoides was a primary agent of melon collapse in Israel. Pathogenicity tests performed in 1995 and 1996 suggested however, that the most virulent species involved in the vine decline syndrome of melons is Monosporascus cannonballus (Kim, 1995; Pivonia et al., 1997). M. cannonballus was reported to induce melon vine decline also in the southern United States, Central

Cucurbitaceae '98


America, Japan, Tunisia, Saudi Arabia and Spain (Bruton, 1997; Karlatti et al., 1997; Kurata, 1994; Martyn et al., 1994; Martyn and Miller, 1996; Stanghellini et al., 1996). Soil disinfestation by fumigation with methyl bromide (M.Br.) is the common treatment used in the Arava to control this disease (Klein, 1996). However, since M.Br. is expected to be banned from use worldwide within the next 5 to 10 years (Ristaino and Thomas, 1997), and because growing melons without M.Br. fumigation in the Arava valley results in severe vine decline and substantial yield loss, alternative measures for disease management are needed urgently.

Growing grafted melons and watermelons on Cucurbita rootstocks to combat diseases is common in various regions in the Mediterranean Basin and in southeast Asia (Kurata, 1994; Lee, 1994). Grafted cucurbits are used mainly to overcome soilborne diseases caused by Fusarium sp. (Lee, 1994). M. cannonballus has been reported in Japan on a bottle gourd rootstock used for grafting watermelon for control of Fusarium wilt (Kurata, 1994). To the best of our knowledge, however, no data are currently available on the response of Cucurbita species used commercially as rootstocks for control of M. cannonballus damages. Grafting is currently not used in Israel commercially, probably due to the availability and high efficiency of M.Br.

Another means by which M. cannonballus could be suppressed is postplanting application of fungicides. In a recent study it was demonstrated that application of fluazinam to the soil several times during the growing season suppressed the intensity of vine decline significantly (Cohen et al., 1997). Control efficacy achieved by fluazinam was however, not always sufficient, and it varied among experiments (Cohen et al., 1997). It is possible that integration of grafting with postplanting application of fungicide may suppress vine decline as effectively as M.Br. Even if these measures are less effective, it may be possible to integrate them with a reduced rate of M.Br which will be allowed in certain places during the phaseout period (Ristaino and Thomas, 1997). In this study various Cucurbita species were screened as a potential source of rootstock for grafting, and the efficacy of grafting,

preplanting fumigation of infested soil with M.Br. at full and reduced rates, and postplanting treatment of fluazinam in suppression of melon vine decline, were studied solely and in combination.

Materials and methods

Grafting melon plants on Cucurbita rootstocks. Seeds of melon (scion) and Cucurbita (rootstocks) (Table 1) were sown in seedlings trays (Polyvid, Mishmar HaNegev, Israel, 37 mm/cell, 128 cells/tray) filled with a 1:1 (v:v) mixture of peat and vermiculite. The plants were grown in an environment-controlled greenhouse with 25/20 oC day and night respectively, and grafted at the two-leaf stage. The true leaves of the rootstocks were removed with a razor blade creating a V-shaped cut between the rootstock's cotyledons. An inverse V-shaped cut was made on the stem of the scion, 2 cm below the cotyledons, to fit the cut in the rootstock. Scion and rootstock were held with a grafting clip (Sakata Seed Corp., Yokohama, Japan). Grafted plants were transferred to a mist chamber (>95% RH) for 8 days, after which the relative humidity was reduced gradually for acclimatization.

Evaluation of rootstock response to vine decline. The response of eight rootstocks belonging to different species of the Cucurbita (Table 1) to vine decline was evaluated in three field experiments, conducted in fields naturally infested with M. cannonballus (Pivonia et al., 1997). The first two experiments were carried out in Fall 1995. One at the 'Zohar' Experiment Station at 'En Tamar (northern Arava), and the other in an organic field at En Yahav (Central Arava). Melon plants (grafted and nongrafted) were transplanted on 12 Sept. Intrarow spacing was 50 cm and beds were 190 cm apart. Each experimental plot was 3.5 m long and consisted of seven transplants. There were four replications per treatment (rootstocks) and the experiments were laid out in a complete block design. The third experiment was conducted at 'En Yahav in the spring cropping season of 1996, on the same plot as the fall experiment. Melon plants were transplanted on 27 Jan. Plant spacing and the number of transplants per experimental plot were the same as for the fall experiment, but there were only two replications for each of the tested rootstocks.

Cucurbitaceae '98


Integrated management of vine decline. The efficacy of grafting, preplanting fumigation with M.Br. and postplanting treatment with fluazinam in suppression of melon vine decline was tested in two field experiments conducted at the 'Zohar' Experiment Station. The first experiment was carried out in the 1977 spring cropping season (transplanted 23 Jan.; harvested on 26 Apr. to 11 May); the second in the cropping fall season of 1997­98 (transplanted 15 Sept.; harvested on 4 to 21 Nov.). 'Arava', which is highly susceptible to M. cannonballus, was planted in both experiments. The spring experiment consisted of a factorial design with two main factors: 1) Grafting: 'Arava' scions were either not grafted or were grafted on 'Brava' rootstock. 2) Preplant treatment with M.Br. This factor consisted of four levels, as follows: i) untreated; ii) M.Br. that had been applied at a rate of 50 g·m­2 before the previous season (before fall cropping of 1996); iii) M.Br. applied at a reduced rate (15 g·m­2); and iv) M.Br. applied at the full rate (50 g·m­2). The fall experiment consisted of a factorial design with three main factors: 1) Grafting: 'Arava' scions were either not grafted or were grafted on 'Brava' rootstock. 2) Preplanting treatment with M.Br.: This factor consisted of three levels, as follows: i) untreated; ii) M.Br. applied at a reduced rate (15 g·m­2); iii) M.Br. applied at the full rate (50 g·m­2); iv) and postplanting application of fungicide. Plots were either not treated or treated with fluazinam (Ohayo, 50% SC, produced by ISK, Japan) at a rate of 750 g a.i./ha, four times during the season.

M.Br. was applied as follows. Drip-irrigation tubes were placed on beds which were then covered with plastic sheets (Polyon, Kibbutz Barkai, Israel). Methyl bromide (Dead Sea Bromine Group, Be'er Sheva, Israel) was applied through the irrigation tubes at the desired rate (15 or 50 g·m­2) employing the hot gas method (Klein, 1996).

The fungicide fluazinam was applied through the drip irrigation system at a rate of 750 g a.i./ha on four dates, starting 10 days after transplanting and continuing at biweekly intervals. Experimental plots were 7 m long and consisted of 14 transplants per plot. There were four replications per treatment and experiments were arranged in a complete block design.

Disease assessments and yield quality measurements. All the experiments, both at 'En Tamar and 'En Yahav, were conducted in fields with a history of vine decline disease caused by M. cannonballus (Pivonia et al., 1997). The effects of the treatments on wilt and on various horticultural parameters were determined. Wilt was evaluated visually in all experiments starting at the onset of the disease and continuing weekly until harvest. Herein are reported results recorded at the harvest date. A plant was considered dead when the whole plant exhibited irreversible wilt symptoms. The number of wilted plants and the total number of plants per plot were used to calculate the incidence of wilting (%). Upon ripening, the fruits in each plot were counted and weighed. Total soluble solids (TSS) as an indicator of fruit sugar content, was evaluated from a sample of five fruits per plot

Table 1. Wilt incidence (%) induced by Monosporascus cannonballus in nongrafted and grafted melon on various rootstocks.z

'En Tamar 'En Yahav 'En Yahav

Rootstock + scion Rootstock species Rootstock source (Fall 1995) (Fall 1995) (Spring 1996)

Nongrafted, Arava Cucumis melo Hazera, Israel 93 ay 100 a 89 a

TZ-148 + Arava Cucurbita maxima Tezier, France 10 b 96 a 28 b

Brava + Arava Cucurbita maxima Petoseed, U.S.A. 0 b 65 a 33 b

Tetsukabuto + Arava C. maxima x C. moschata Takii Seed, Japan 13 b 65 a 50 b

Marukanpyou + Arava Lagenaria siceraria Takii Seed, Japan 76 a ntx nt

Feggous + Arava Cucumis melo Inter Semences, Moroco nt 100 a nt

Slaoui + Arava Lagenaria siceraria (Unknown) nt 100 a nt

Just + Arava Cucurbita sp. Takii Seed, Japan 14 b nt nt

Butternut + Arava Cucurbita moschata Petoseed, USA 67 a nt nt

zExperiments were conducted in naturally infested soil at 'En Tamar (northern Arava) and at 'En Yahav (central Arava) in southern Israel. Disease incidence was evaluated 60 days after transplanting.

yWithin columns, values with the same letters, do not differ significantly (P = 0.05), as determined by Fisher's protected lsd test.

xnt = not tested.

Cucurbitaceae '98


using an Atago digital refractometer. Results were analyzed statistically by appropriate (one-, two-, or three-way) ANOVA tests. When the F values were significant at P > 0.05, differences among the treatments were determined by Fisher's protected lsd test.

Results

Evaluation of rootstock response to vine decline. Wilt intensity in all three experiments was high: 89% to 100% of the nongrafted 'Arava' plants had wilted (Table 1). Of the eight rootstocks evaluated, only three ('TZ-148', 'Brava' and 'Tetsukabuto') were tested in three field experiments, the others were tested only once, due to a limited amount of seeds. All three rootstocks exhibited the same response to the disease within each experimental location, viz., significant (68% to 100%) reduction in wilt incidence in the experiments conducted at 'En Tamar, and an insignificant (4% to 35% reduction) effect in the 'En Yahav experiment (Table 1). The other rootstocks tested were comparable to or less effective than those mentioned. Since the horticultural performance of the melons grafted on 'Brava' seemed to be somewhat better than that of the other transplant combinations (data not shown), 'Brava' was selected to serve as a rootstock in the next experiments.

Integrated management of vine decline. Severe wilt incidence developed in both experiments in untreated plants (94% to 100% wilt) (Tables 2 and 3). The full rate of M.Br. (50 g·m­2) was highly effective and resulted in 98% to 100% wilt suppression. The reduced M.Br. rate (15 g·m­2) was effective as well and suppressed wilt significantly, but results were variable (32% control in the 1997 spring experiment and 91% control in the 1997­98 fall experiment). A previous season M.Br. treatment was highly effective as well (96% control). Grafting, when employed as the sole measure (in plots not treated with M.Br. or with fluazinam), significantly reduced wilt incidence (by 84% and 87%) in both experiments. Although wilt incidence in that treatment was higher than in the M.Br. treatment (16% vs. 2%, spring experiment; 12% vs. 0%, fall experiment), these values did not differ significantly at P = 0.05. Integration of grafting and reduced rate of M.Br. did not improve wilt suppression further in either experiment when compared with each of the measures applied solely. Postplanting application of fluazinam significantly reduced wilt incidence (by 88%) when applied to nongrafted plants planted in plots not treated with M.Br. However, when applied to M.Br.-treated plots (at both rates) or to grafted plants, any additive contribution of fluazinam was obscured (Tables 2 and 3).

 

Table 2. Effects of methyl bromide and grafting on wilt incidence, induced by Monosporascus cannonballus, and on yield and its components, in an experiment conducted at 'En Tamar in Spring 1997.z

Treatment

Soil Grafting Wilt (%) Fruit/ha Fruit wt (kg) Yield (t/ha)

Untreated ­ 100 8210 0.61 5.1

+ 16 25670 0.71 18.7

M.Br., previous seasony ­ 4 26240 0.65 17.0

+ 2 34620 0.77 26.8

M.Br., 15 g·m­2 ­ 68 20160 0.66 13.3

+ 25 35670 0.79 28.2

M.Br., 50 g·m­2 ­ 2 28910 0.69 20.1

+ 0 36230 0.79 28.7

Analysis of variance

Main effects

Soil (S) 0.0001 0.0225 0.0004 0.0001

Grafting (G) 0.0001 0.0042 0.0001 0.0001

Interaction

S ¥ G 0.0001 0.7738 0.3970 0.8476

Maximal lsd 17.4 7747 0.08 4.93

yM.Br. at the commerical rate of 50 g·m­2 was applied before the Fall 1996 cropping season.

zMelon ('Arava') transplants were grown in naturally infested soil.

Cucurbitaceae '98


Effects on yield, yield components and fruit quality. Yield was significantly affected by the vine decline disease, as reflected by the significant yield increase in treated plots. In the 'En Tamar spring 1997 experiment, for example, the yield in M.Br. grafted plots was 5.7 times higher than that in the untreated plots (28.7 vs. 5.1 t·ha­1, respectively) (Table 2). Each of the control measures, when examined alone in comparison with an untreated plot, resulted in a significant increase in yield (P < 0.05). However, the increase in yield in M.Br. and in grafted plants was higher than that obtained in fluazinam treated plots (Tables 2 and 3). When more than one measure was applied, the results varied among experiments. In the 1997 spring experiment, the main effects of fumigation and of grafting was significant, suggesting that the effects of the two measures on yield were additive (Table 2). On the other hand, in the 1997 fall experiment, yield did not differ significantly in plots treated solely or in combination for each of the three control measures (Table 3).

 

The number of fruits per unit area was the yield component that was affected most by the disease (and increased by the treatments). In general, the effects of the different treatments on the number of fruits paralleled the effects on yield. A large increase (up to 4.4 times) was observed in the 1977 spring experiment with additive effects of M.Br. and grafting, and sole, but nonadditive, effect in the 1997 fall experiment. Fruit weight was affected significantly only in the spring experiment, but it was affected much less than was the number of fruits (up to 1.3x increase). Fruit weight was not affected by the disease in the fall experiment. Fruit quality (as measured by TSS) in the 1997 fall experiment was not affected (Tables 2 and 3).

Discussion

The widespread and destructive nature of the vine decline disease of melon caused by M. cannonballus constitutes a serious threat to melon cultivation in the major production areas of the world. Currently, the only means of controlling

Table 3. Effects of methyl bromide, grafting, and postplanting application of fluazinam on wilt incidence, induced by Monosporascus cannonballus, and on yield and its components and quality in an experiment conducted at 'En Tamar, Fall 1997.z

Treatment

Soil Grafting Fluazinamy Wilt (%) Fruit/ha Fruit wt (kg) Yield (t·ha­1) TSSx (%)

Untreated ­ ­ 93.7 14040 1.00 13.5 10.2

­ + 11.0 18980 1.00 18.9 11.6

+ ­ 12.2 17940 1.23 22.3 11.6

+ + 15.2 16120 1.19 19.7 11.4

M.Br. 15 g·m­2 ­ ­ 8.2 24700 1.06 16.7 11.7

­ + 0.0 24960 1.18 29.6 12.5

+ ­ 0.0 22880 1.19 27.4 10.3

+ + 8.2 21580 1.18 25.7 12.0

M.Br. 50 g·m­2 ­ ­ 0.0 23920 1.07 25.6 11.3

­ + 0.0 26780 1.13 30.2 11.2

+ ­ 0.0 24440 1.08 26.3 11.5

+ + 2.0 24440 1.02 25.6 10.9

Analysis of variance

Main effects

Soil (S) 0.0001 0.0001 0.2240 0.0001 0.3317

Grafting (G) 0.0001 0.3441 0.0517 0.8280 0.5729

Chemical (C) 0.0001 0.4336 0.7560 0.4212 0.0836

Interaction

S ¥ G 0.0001 0.4786 0.0234 0.1159 0.1083

S ¥ C 0.0001 0.6611 0.7552 0.4560 0.0824

G ¥ C 0.0001 0.0815 0.2029 0.0440 0.5142

S ¥ G ¥ C 0.0001 0.5736 0.8297 0.8935 0.1951

Maximal lsd 10.6 5160 0.18 7.4 2.1

zMelon ('Arava') transplants were grown in naturally infested soil.

yThe fungicide fluazinam was applied biweekly (four times) through the drip irrigation system, at a rate of 750 (a.i.) g·ha­1.

xTSS = total soluble solids.

Cucurbitaceae '98


the disease in Israel is by application of methyl bromide as a preplant soil treatment (Klein, 1996). M.Br. is indeed highly effective. In plots treated with the full rate of the fumigant (50 g·m­2), wilt was significantly reduced to very low levels (98% to 100% reduction). When applied at a reduced rate (15 g·m­2), control efficacy was variable: in one experiment it was 91.4% but in another it was 32% (Tables 2 and 3). The use of M.Br. is being phased out and will be banned worldwide within the next 5 to 10 years (Ristaino and Thomas, 1997). Thus, development of alternative control measures is of great importance to the melon industry.

Growing grafted vegetables including cucurbits is a common method employed in Japan, Korea, the Mediterranean Basin and some countries in Europe (Lee, 1994). The purpose of growing grafted melons is mainly to control Fusarium wilt caused by various races of F. oxysporum f.sp. melonis. To the best of our knowledge no data are currently available on the possibility of using grafting against M. cannonballus.

Genetic sources for rootstocks resistant or tolerant to M. cannonballus may exist in various plants of the Cucurbitaceae. Cucurbita, as a genus was defined by Mertely et al. (1993) as a host for M. cannonballus. Indeed, perithecia of M. cannonballus could be detected on various Cucurbita roots recovered from the soil after harvest, although no wilt symptoms were evident throughout the growing season (S. Pivonia, unpublished). Little is known, however on the rate of infection of Cucurbita compared with melons or other cucurbits, and the response of different Cucurbita accessions to the pathogen.

In general, results obtained in this study suggest that grafting is an effective measure against melon vine decline. When 'Arava' scions were grafted on 'Brava' rootstocks, wilt incidence was significantly reduced ( by 63% to 100%) in four of the five experiments. In one experiment ('En Yahav, fall 1995) where disease pressure was very high, wilt incidence in the grafted plants did not, however, differ significantly (35% reduction) from that in nongrafted plants (Table 1). The variable performance of grafting is a known phenomenon (Lee, 1994). Moreover, a significant reduction in wilt incidence in a certain year does not necessarily

reflect the whole picture. The pathogen population can still increase gradually in the soil and once it reaches a certain level (in the following seasons), the grafted plants might be endangered. In addition, differences in environmental conditions, especially high temperatures, can enhance symptoms development (Cohen et al., 1996) and cause disease in grafted plants. The differences in the response to the disease of grafted plants in the fall of 1995 at 'En Yahav versus 'En Tamar (Table 1) are an example of one such possible variation. Thus, occasional incidence of wilt should not be unexpected and additional measures for disease control should be implemented.

Application of the fungicide fluazinam as a postplanting treatment was recently shown to reduce significantly the incidence of vine decline in melons (Cohen et al. 1997). This treatment was tested in the current study in one experiment, as a further repetition of the previous study. Moreover, it was integrated with the other control measures (i.e., M.Br. and grafting). The results obtained here corroborated those obtained in the previous study: wilt incidence in plots treated postplanting with fluazinam was significantly reduced (by 88%), in comparison with nontreated plots (Table 3). Since fluazinam alone was not always as effective as M.Br., it was recommended to be used as one measure in an integrated control strategy (Cohen et al., 1997).

Effects of the treatments on yield resemble those obtained on wilt suppression. In general, treatments that resulted in reduced wilt, increased yield significantly as well. The main yield component that was affected by the disease (and thus was increased by the control treatments) was the number of fruits per unit area. Fruit quality, which is determined mostly by its TSS, was not affected (Tables 2 and 3). These results reflect the expression of the epidemic of the disease in the field, where infected plants collapse before harvest and do not produce any marketable fruit.

Part of the yield increase observed following soil fumigation with M.Br. and the use of grafted plants is attributed to disease control. Increased growth response (IGR), which is not related to the control of a major pathogen, is a well known phenomenon in both soil fumigation and grafted

Cucurbitaceae '98


plants. The IGR effect following soil fumigation results largely from the decrease in or elimination of minor pathogens. The IGR resembles a fertilizer response because the healthier roots are more absorptive and also more effective in manufacturing growth factors for transfer to the plant top (Cook and Baker, 1983). In grafted plants, the rootstock's vigorous root system is often capable of absorbing water and nutrients more efficiently than scion roots, and serves as a good supplier of endogenous plant hormones. The rootstock effect varies greatly with scion cultivar and growing season (Lee, 1994). An IGR effect my have been more pronounced in the Spring 1997 experiment, when wilt incidence in the grafted and nongrafted plots treated with two of the M.Br. treatments (previous season, and full rate M.Br) was similar (0% to 4% wilt). Yield in the grafted plots was, however, significantly higher (by 8.7 to 9.8 t·ha­1) than in the nongrafted plots (Table 2). The IGR effect if grafting was not apparent in the Fall 1997 experiment (Table 3).

Grafted plants are expected to be expensive. They originate from rootstock and the scion seeds both of which may be expensive hybrids. The preparation of the transplant includes manual labor and expensive postgrafting handling (Kurata, 1994; Lee, 1994).

Extensive efforts to overcome the labor problem are being made in Japan, where private companies, institutions and universities are attempting to develop feasible grafting robots. In the near future, it will be common practice for growers to buy seedlings grafted by robots (Kurata, 1994).

In this study, integration of three control measures was tested, viz., grafting, preplanting measure of M.Br. at full and reduced rate, and postplanting application of fluazinam. Integration of at least two of the control measures provided adequate and reliable solution to vine decline of melon. Wilt incidence in plots treated with these measures was always significantly lower (by 75% to 100%) than that in untreated plots. Although the efficacy of these measures did not differ significantly from application of M.Br. at full rate, it was sometimes less effective. Nevertheless, yield in these plots was always similar to (or even higher than) yield in the full rate M.Br. plots

(Tables 2 and 3). The overall conclusion is that reduced rates of M.Br., grafting and postplanting application of fungicides, can be used as alternative measures to full rate application of M.Br. Since control efficacy of each of these measures may vary, it is preferable to use them in an integrated treatment as insurance.

Literature cited

Bruton, B.D. 1997. Occurrence of vine decline disease on muskmelon in Guatemala. Plant Dis. 81:694.

Cohen, R., Y. Elkind, Y. Burger, R. Offenbach, and H. Nerson. 1996. Variation in the response of melon genotypes to sudden wilt. Euphytica 87:91­95.

Cohen, R., S. Pivonia, D. Shtienberg, D. Raz, S. Shraiber, Z. Gerstl, and J. Katan. 1997. Chemical control of Monosporascus, the causal agent of the root rot and vine decline disease of melons. Phytoparasitica 25:255 (Abstr.).

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

Karlatti, R.S., F.M. Abdeen, and M.S. Al-Fehaid. 1997. First report of Monosporascus cannonballus on melons in Saudi Arabia. Plant Dis. 81:1215.

Kim, D.H., S.L. Rasmussen, and M.E. Stanghellini. 1995. Monosporascus cannonballus root rot of muskmelon: Root infection and symptoms development in relation to soil temperature. Phytopathology 85:1195. (Abstr.)

Klein, L. 1996. Methyl bromide as soil fumigant, p. 191­235 In: C.H. Be, N. Price, and B. Chakrabarti (eds.). The methyl bromide Issue. Wiley, New York.

Kurata, K. 1994. Cultivation of grafted vegetables II. Development of grafting robots in Japan. HortScience 29:240­244.

Lee, J.M. 1994. Cultivation of grafted vegetables I. Current status, grafting methods, and benefits. HortScience 29:235­239.

Martyn, R.D., B.R. Lovic, D.A. Maddox, A. Germash, and M.E. Miller. 1994. First report of Monosporascus root rot/vine decline of watermelon in Tunisia. Plant Dis. 78:1220.

Martyn, R.D. and M.E. Miller. 1996. Monosporascus root rot and vine decline, an emerging disease of melons worldwide. Plant Dis. 80:716­725.

Mertely, J.C., R.D. Martyn, M.E. Miller, and B.D. Bruton. 1993. An expanded host range for the muskmelon pathogen Monosporascus cannonballus. Plant Dis. 77:667­673.

Pivonia, S., R. Cohen, U. Kafkafi, IS. Ben-Ze'ev, and J. Katan. 1997. Sudden wilt of melons in southern Israel: Fungal agents and relationship with plant development. Plant Dis. 81:1264­1268.

Reuveni, R., J. Krikun, and U. Shani. 1983. The role of Monosporascus eutypoides in a collapse of melon plants in arid area of Israel. Phytopathology 73:1223­1226.

Ristaino, J.B. and W. Thomas. 1997. Agriculture, methyl bromide and the ozone hole, can we fill the gap? Plant Dis. 81:964­977.

Stanghellini, M.E., D.H. Kim, and S.L. Rasmussen. 1996. Ascospores of Monosporascus cannonballus: Germination and distribution on cultivated and desert soils in Arizona. Phytopathology 86: 509­514.

Cucurbitaceae '98