Corps de l’article

Introduction

Turkey is an important world producer of loquat, Eriobotrya japonica (Thunb.) Lindl. After the economic value of loquat was realized, demand for commercial production rapidly increased. Total production increased more than four-fold between 1980 (3000 t) and 2003, when it reached about 12 000 t produced from 246 000 trees (Polat and Caliskan 2007). Turkey supplies 27% of the world's total quince (Cydonia spp.) production, with 150 000 t produced annually (Gunes and Dumanoglu 2005).

Fire blight caused by the bacterium Erwinia amylovora (Burr.) Winslow et al. is the most devastating bacterial disease of apple (Malus domestica Borkh), pear (Pyrus communis L), quince and loquat trees worldwide (Jones and Aldwinkle 1990; van der Zwet and Keil 1979; Vanneste 2000). The disease significantly limits the area for successful pome fruit production. It is probable that fire blight will remain a concern as the pathogen may be introduced into new pome fruit growing areas (Jock et al. 2002; van der Zwet 2002; Vanneste 2000). There are several distinct phases of the disease, including blossom blight, shoot blight and rootstock blight. The diversity of host tissues susceptible to infection, combined with the limited number of management tools available to control the disease, has made it difficult to prevent the progress of fire blight epidemics.

Severe fire blight symptoms were observed on quince trees (Cydonia vulgaris (Pers.) cultivars Eşme and Ekmek) in different orchards in Turkey in 2003. The disease had spread to all quince-growing areas located in the Mediterranean, Aegean, Marmara and Middle Anatolia regions of Turkey (Saygili et al. 2006).

The most commonly applied bactericides for controlling fire blight in pome fruit trees are copper compounds and antibiotics (Psallidas and Tsiantos 2000). However, russeting of fruits often results from copper treatments. Furthermore, the authorities in many countries do not permit the use of antibiotics for fire blight control because of potential risks of promoting the development of antibiotic resistance in human pathogens (McManus et al. 2002). Due to the lack of publicly acceptable, effective and non- phytotoxic preparations to control fire blight, there has been an interest in novel control strategies that trigger defense mechanisms in the host plants. Such effects have been reported for the harpin protein, benzothiadiazole (acibenzolar-S-methyl) and prohexadione-Ca (Aldwinckle et al. 2002; Bazzi et al. 2003; Maxson and Jones 2002; McManus et al. 2002; Momol et al. 1999; Norelli et al. 2003; Steiner 2000).

One of the most important factors determining host susceptibility to fire blight shoot infections is the intensity of vegetative shoot growth on the fruit trees (Byers and Yoder 1997). An alternative and promising approach to control fire blight is the use of growth-regulating acylcyclohexanediones, such as prohexadione-Ca. The most notable effect of prohexadione-Ca is the control of shoot elongation growth by inhibiting the formation of growth-active gibberellins from inactive precursors (Evans et al. 1999; Rademacher 2000). Prohexadione-Ca-treated apple and pear trees are significantly less infected by E. amylovora (Aldwinckle et al. 2002; Bazzi et al. 2003; Buban et al. 2002, 2003; Cline and Hunter 2002; Costa et al. 2001a; Deckers and Schoofs 2002; Fernando and Jones 1999; Maxson and Jones 2002; Momol et al. 1999; Norelli et al. 2003; Sobiczewski et al. 2001; Winkler 1997; Yoder et al. 1999) and Venturia inaequalis (Cooke) G. Wint. (Costa et al. 2001b; Spinelli et al. 2002). Another new approach lies in the induction of systemic acquired resistance (SAR), a self-defense mechanism in plants. The SAR response correlates with the accumulation of pathogenesis-related (PR) proteins. The PR proteins can be induced by some plant activators such as benzothiadiazole and harpin protein (Anonymous 2002; Brisset et al. 2000; Fontanilla et al. 2005; Jones 2001; Wei and Beer 1996). In addition, humic acid is reported by the manufacturer to improve plant vigour and natural resistance to plant diseases (Anonymous 2000; Freeman 1969).

In this study, the effectiveness of host resistance inducers (prohexadione-Ca, benzothiadiazole and harpin protein) and a fertilizer (humic acid) was investigated on two hosts that have not been studied so far (quince and loquat cultivars) on the shoot blight phase of fire blight disease.

Materials and Methods

Plant material and growth conditions

A loquat cultivar, Cukurgöbek, and two quince cultivars, Eşme and Ekmek, grown extensively in Turkey, were used in this experiment. The test plants were selected among 3-yr-old saplings showing uniform growth. These saplings were transplanted into plastic pots of 25 cm diam filled with 8 kg of soil, and they were grown for 20 d at 25 ± 5°C, 60-75% RH, and under 12 000-14 000 Lux from tungsten-filament lamps for a 16-h photoperiod. The potted plants were placed on the ground 1 m apart at a field location in the Konya province and were watered regularly throughout the growing season.

After transplantation, the trees were fertilized once a week (each pot) with 25 g ammonium sulfate, 25 g diammonium phosphate, 25 g potassium sulfate, and 50 mL of a liquid fertilizer having 0.05% Mn, Cu, Zn, B, and Mo (Kacar and Katkat 1999). In addition, sulfur dust was applied once (4 g L-1 water) for powdery mildew control.

Erwinia amylovora strain

After conducting virulence tests on cv. Ankara pear trees, a virulent strain of E. amylovora (EAI) was selected for all inoculations (Norelli et al. 1984). Stock cultures were preserved at 4°C on the nutrient agar (NA) medium and transferred to new tubes every 3 mo.

Bacterial suspensions were prepared from growing colonies on NA at 23-25°C and were diluted in sterile distilled water (SDW) to give an absorbance of 0.15 at 660 nm. This represented 108 cfu mL-1 based on viable plate counts. Inoculum was maintained on ice and was used for plant inoculation within 2 h of dilution.

Chemical compounds used in the experiments and their applications

The chemical compounds used in the experiment were: prohexadione-Ca, benzothiadiazole + metalaxyl, harpin protein, humic acid, copper salts and streptomycin. These compounds and their properties are shown in Table 1. Chemical application timing and schedule were based on Momol et al. (1999) (Table 2). Prohexadione-Ca was applied twice when the shoots were 6-12 cm and 15-20 cm long. Benzothiadiazole + metalaxyl and harpin protein were applied when the shoots measured 15-20 cm and 30-35 cm. Copper salts and humic acid were applied three times when the shoots were 6-12 cm, 15-20 cm and 30-35 cm long. Streptomycin was applied twice, 1 d before and 1 d after inoculation (Momol et al. 1999). Streptomycin and copper salts are known to be effective against fire blight and were applied to allow comparison with the host resistance inducer treatments.

Table 1

Active ingredients, formulation and application rate of the chemical compounds used in the experiment

Active ingredient and percentage

 

Commercial name / Firm

Formulation

Application rate (100 L‑1 water)

Prohexadione-Ca

10%

Apogee / BASF

WG1

125 g

Benzothiadiazole

4%

BION MX 44 / Syngenta

WG

135 g

+ Metalaxyl

40%

 

 

 

Harpin protein

3%

Messenger / Eden Bioscience

Powder

50 g2 + 20 mL adjuvant3

Humic acid

55%

K-humate / Hektas A.S.

Powder

20 mL

+ Fulvic acid

30%

 

 

 

+ Potassium hydroxide

8%

 

 

 

Streptomycin sulfate

100%

Streptomycine / I.E. Ulagay

Powder

59 g

Copper salts

51.4%

Tenn Cop 5E / Hektas A.S.

Liquid

250 mL

1 WG: wettable granule.

2 Prepared in distilled water.

3 KINETIC® (non-ionic adjuvant) manufactured for Helena Chemical Company.

-> Voir la liste des tableaux

Table 2

Date of the chemical applications and date of inoculation with Erwinia amylovora to potted loquat and quince trees

 

 

Application time and shoot length (cm)

Treatment

 

June (6‑12 cm)2 

June (15‑20 cm)3 

July (30‑35 cm)4 

July (40‑45 cm)5 

July (40‑45 cm)6 

Prohexadione-Ca

                               

                               

                               

                               

                               

                               

Inoculated1

x

x

 

 

 

 

Uninoculated

x

x

 

 

 

 

Inoculated + Cu

x

x

 

 

 

Benzothiadiazole + Metalaxyl

 

 

 

 

 

 

Inoculated

 

x

x

 

 

 

Uninoculated

 

x

x

 

 

 

Inoculated + Cu

 

x

x

 

 

Harpin protein

 

 

 

 

 

 

Inoculated

 

x

x

 

 

 

Uninoculated

 

x

x

 

 

 

Inoculated + Cu

 

x

x

 

 

Humic acid

 

 

 

 

 

 

Inoculated

x

x

x

 

 

 

Uninoculated

x

x

x

 

 

 

Inoculated + Cu

x

x

x

 

 

Copper salts

 

 

 

 

 

 

Inoculated

x

x

x

 

 

 

Uninoculated

x

x

x

 

 

Streptomycin

 

 

 

 

 

 

Inoculated

 

 

 

x

x

 

Uninoculated

 

 

 

x

x

Control

 

 

 

 

 

 

Inoculated

x

x

x

 

 

 

Uninoculated

x

x

x

 

 

1 Inoculated with Erwinia amylovora after application of chemical, on 16 July 2002 and 8 July 2003.

2 Treatments applied on 15 June 2002 and 9 June 2003.

3 Treatments applied on 25 June 2002 and 20 June 2003.

4 Treatments applied on 10 July 2002 and 2 July 2003.

5 Treatments applied on 15 July 2002 and 7 July 2003.

6 Treatments applied on 17 July 2002 and 9 July 2003.

-> Voir la liste des tableaux

Experimental design and set-up

The experiment was set up in a completely randomizedblock design with three replicates. A single replicate was a mean from nine shoots on three saplings (Duzgunes et al. 1987). Each treatment was applied to five groups of plants (Table 3). The first three groups of plants were treated by the chemicals and inoculated with E. amylovora to see the effects of chemicals on disease severity (first group treated with chemicals + E. amylovora inoculation, second group treated with chemicals + copper salts + E. amylovora inoculation, third group as control 1 treated only with E. amylovora inoculation). The fourth group was treated only with the chemicals, and the fifth group, as control 2, was treated with water to see the effects of treatments on shoot growth of loquat and quince. The different combinations of treatments were all analyzed as separate treatments. The experiment was conducted during two growing seasons, 2002 and 2003.

Table 3

Experimental design for applications on loquat and quince cultivars

Plant groups                

 

Applications

First group

 

Chemicals (PC1, BTHM, HRP, HA, CU, STR)

 

 

+ E. amylovora inoculation

Second group

 

Chemicals (except Copper and Streptomycin)

 

 

+ Copper salts

 

 

+ E. amylovora inoculation

Third group

 

E. amylovora inoculation

(Control1 = for disease severity)

 

 

Fourth group

 

Chemicals

Fifth group

 

Water

(Control2 = for shoot growth)

 

 

1 PC: Prohexadione-Ca; BTHM: Benzothiadiazole + Metalaxyl; HRP: Harpin protein; HA: Humic acid; CU: Copper salts; STR: Streptomycin.

-> Voir la liste des tableaux

Inoculation of the shoots

Actively growing shoot tips of plants were inoculated by inserting a 0.46-mm-diam (26-gauge) hypodermic needle through the stem just above the youngest unfolded leaf. A suspension of 108 cfu mL-1E. amylovora was introduced to fill the wound and leave visible drops at both ends of the wound. The treated shoots were labeled with flagging tape for evaluation purposes (Norelli et al. 1986).

Evaluation of disease severity and shoot growth

The length of visible fire blight lesions and of the current season's shoot growth was recorded after all lesions had ceased to extend, as indicated by the formation of a determinate margin between diseased and healthy tissues.

Disease severity was calculated using the following formula: Disease severity (%) = (a / b) x 100; where a is the length of the blighted part of the shoot (cm), and b is the whole length of the shoot (cm) (Fernando and Jones 1999).

Percent effectiveness of the applications (A) was calculated according to the following formula: A = 100 x (B - C) / B; where B is the percent disease severity in the controls, and C is the percent disease severity in treated shoots.

Percent effectiveness of the treatments on reduction of shoot growth (D) was calculated in a similar way: D = 100 x (E - F) / E; where E is the mean shoot length in the controls, and F is the length of treated shoots (Anonymous 1996).

MINITAB (State College, PA, USA) was the statistical program used. The means (expressed as percent disease) were used to determine significant treatment differences. Data were analyzed using MSTAT software (Michigan State University, MI, USA) and the differences between treatments were determined by Duncan's New Multiple Range Test at P ≤ 0.05.

Results

Control of shoot blight on loquat

Shoot blight caused by E. amylovora was destructive on loquat with a disease severity of 91.06 and 82.29% in control plots in 2002 and 2003, respectively (Table 4). Streptomycin was the most effective treatment in preventing shoot blight. It was followed by benzothiadiazole + metalaxyl with 59.48 and 46.62% effectiveness, which was more (P ≤ 0.05) than all the other treatments during the 2 yr. Prohexadione-Ca, harpin protein and copper salts lacked sufficient efficacy.

Table 4

Effectiveness of the treatments on disease severity caused by Erwinia amylovora on loquat cv. Cukurgöbek in 2002 and 2003

 

2002

2003

 

Disease severity

Effectiveness

Disease severity

Effectiveness

Treatment1

(%)

(%)

(%)

(%)

Prohexadione Ca

88.54 bcd2

2.76

75.02 cd

8.83

BTHM

36.89 g

59.48

43.92 g

46.62

Harpin protein

83.62 de

8.17

69.00 e

16.15

Humic acid

95.32 a

0.00

80.88 a

1.71

Copper salts

88.79 bcd

2.49

77.59 bc

5.71

Streptomycin

1.27 h

98.60

2.83 h

96.56

PC + CU

84.84 cde

6.83

74.04 d

10.02

BTHM + CU

61.43 f

32.53

49.39 f

39.98

HRP + CU

82.61 e

9.27

70.00 e

14.93

HA + CU

89.20 bc

2.04

79.89 ab

2.91

Control

91.06 ab

0.00

82.29 a

0.00

1 BTHM: Benzothiadiazole + Metalaxyl; PC: Prohexadione-Ca; CU: Copper salts; HRP: Harpin protein; HA: Humic acid.

2 In a column, values followed by the same letter are not significantly different (P ≤ 0.05) as determined by Duncan's New Multiple Range Test.

-> Voir la liste des tableaux

The addition of copper to the various treatments did not increase the effectiveness of the plant growth regulator and activators. Copper compound alone was not effective and only reduced shoot blight by 2.49 and 5.71% in 2002 and 2003, respectively. Disease severity in humic acid applications was 95.32 and 80.88% in 2002 and 2003, respectively. It did not differ significantly from the uninoculated control (Table 4).

Reduced shoot growth on loquat

A 2 to 12% reduction in shoot growth was observed in the treatments including streptomycin, copper, BTH and harpin protein. However, the most important reduction (30%) was only obtained with the Prohexadione-Ca treatment (Table 5). Shoot lengths of loquat saplings treated with Prohexadione-Ca were significantly shortened, measuring 16.36 cm and 19.63 cm in 2002 and 2003, respectively, in comparison with the untreated control (23.55 cm and 27.70 cm in 2002 and 2003, respectively).

Table 5

Effectiveness of the treatments on reducing shoot growth of loquat cv. Cukurgöbek in 2002 and 2003

 

2002

2003

 

Shoot length

Effectiveness

Shoot length

Effectiveness

Treatment

(cm)

(%)

(cm)

(%)

Prohexadione Ca

16.36 d2

30.53

19.63 e

29.13

BTHM1

21.00 c

10.82

29.03 abc

0.00

Harpin protein

20.53 c

12.82

29.21 ab

0.00

Humic acid

23.64 a

0.00

31.11 a

0.00

Copper salts

22.25 b

4.24

26.07 d

5.88

Streptomycin

23.09 ab

1.95

26.76 cd

3.39

Control

23.55 a

0.00

27.70 bcd

0.00

1 BTHM: Benzothiadizole + Metalaxyl.

2 In a column, values followed by the same letter are not significantly different (P ≤ 0.05) as determined by Duncan's New Multiple Range Test.

-> Voir la liste des tableaux

Control of shoot blight on quince

After the streptomycin treatment, applications of harpin protein alone or in combination with the copper compound were the most effective treatments for the shoot blight phase of fire blight for both years. The effectiveness of harpin protein in disease control on cv. Eşme was 25.83 and 23.06% in 2002 and 2003, respectively, and 11.89 and 14.82% on cv. Ekmek. This was greater than copper compound effectiveness, which was 11.87 and 7.34% on cv. Eşme, and 8.00 and 6.42% on cv. Ekmek (P ≤ 0.05). Copper salts did not increase the effectiveness of the chemicals (Table 6).

Copper salts were less effective than prohexadione-Ca and benzothiadiazole + metalaxyl applications in reducing the severity of fire blight on shoots inoculated with E. amylovora in 2003. Disease severity was numerically lower than with the untreated control, but there were no statistical differences (Table 6). Disease severity values resulting from humic acid treatments did not differ from the untreated control saplings (Table 6).

Table 6

Effectiveness of the treatments on disease severity caused by Erwinia amylovora on quince cultivars Eşme and Ekmek in 2002 and 2003

 

cv. EŞME

cv. EKMEK

 

2002

2003

2002

2003

 

Diseasess

Effectiveness

Disease

Effectiveness

Disease

Effectiveness

Disease

Effectiveness

 

severity

 

severity

 

severity

 

severity

 

Treatment1

     (%)     

     (%)     

     (%)     

     (%)     

     (%)     

     (%)     

     (%)     

     (%)     

Prohexadione‑Ca

53.59 ab2

8.42

53.66 cde

16.14

59.69 ab

6.98

64.17 bc

10.36

BTHM

54.12 ab

7.51

54.67 bcde

14.56

62.27 ab

2.96

64.83 bc

9.44

Harpin protein

43.40 c

25.83

49.23 e

23.06

56.54 b

11.89

60.98 c

14.82

Humic acid

56.21 ab

3.94

60.96 ab

4.73

62.88 ab

2.01

69.03 ab

3.57

Copper salts

51.57 b

11.87

59.29 abc

7.34

59.03 ab

8.00

66.99 abc

6.42

Streptomycin

11.56 d

80.24

7.98 f

87.52

13.29 c

79.28

10.25 d

85.68

PC + CU

50.30 b

14.04

51.93 de

18.84

61.30 ab

4.47

63.14 bc

11.80

BTHM + CU

51.19 b

12.52

55.92 bcde

12.61

63.11 ab

1.65

67.32 abc

5.96

HRP + CU

41.72 c

28.70

51.80 de

19.04

57.43 ab

10.50

61.59 c

13.96

HA + CU

53.93 ab

7.84

57.89 abcd

9.53

60.90 ab

5.09

68.96 ab

3.67

Control

58.52 a2

0.00

63.99 a

0.00

64.17 a

0.00

71.59 a

0.00

1 BTHM: Benzothiadiazole + Metalaxyl; PC: Prohexadione-Ca; CU: Copper salts; HRP: Harpin protein; HA: Humic acid.

2 In a column, values followed by the same letter are not significantly different (P ≤ 0.05) as determined by Duncan's New Multiple Range Test.

-> Voir la liste des tableaux

When the quince cultivars were evaluated independently, disease severity on cv. Ekmek was 56.42 and 60.80% in 2002 and 2003, respectively, and 47.83 and 51.57% on cv. Eşme. These results show that disease severity was less important on 'Eşme' than on 'Ekmek'. The data also show that there are numerical differences in disease severity between the cultivars, but these differences are not significant (P ≤ 0.05).

Reduced shoot growth on quince

Prohexadione-Ca was the only treatment that significantly reduced shoot length on quince cultivars, and its effectiveness was the highest (30.96%) on the susceptible cv. Ekmek in 2003 (Table 7).

Table 7

Effectiveness of the treatments on reducing shoot growth of quince cultivars Eşme and Ekmek in 2002 and 2003

 

cv. EŞME

cv. EKMEK

 

2002

2003

2002

2003

 

Shoot

Effectiveness

Shoot

Effectiveness

Shoot

Effectiveness

Shoot

Effectiveness

 

length

 

length

 

length

 

length

 

Treatment

     (cm)     

     (%)     

     (cm)     

     (%)     

     (cm)     

     (%)     

     (cm)     

     (%)     

Prohexadione‑Ca

33.25 e2

22.78

33.94 e

26.10

32.18 e

24.28

32.48 e

30.96

BTHM1

41.58 d

3.43

46.02 abcd

0.00

43.63 bcd

0.00

48.71 a

0.00

Harpin protein

41.71 d

3.13

48.02 ab

0.00

43.84 bcd

0.00

47.98 ab

0.00

Humic acid

43.82 bcd

0.00

47.16 abc

0.00

44.83 abcd

0.00

46.97 abc

0.17

Copper salts

41.58 d

3.43

44.32 abcd

3.50

41.63 d

2.04

45.17 abcd

3.99

Streptomycin

41.79 d

2.94

45.73 abcd

0.43

41.73 d

1.81

45.05 abcd

4.25

Control

43.06 cd

0.00

45.93 abcd

0.00

42.50 cd

0.00

47.05 abc

0.00

1 BTHM: Benzothiadizole + Metalaxyl.

2 In a column, values followed by the same letter are not significantly different (P ≤ 0.05) as determined by Duncan's New Multiple Range Test.

-> Voir la liste des tableaux

Discussion

The shoot blight phase of fire blight caused by E. amylovora is highly destructive within the current and subsequent growing seasons, and improved strategies are required for the control of fire blight on pome fruits. Danovan (1991) and Byers and Yoder (1997) reported that the first factor determining the susceptibility of the host plant against shoot infections of fire blight was rapid shoot growth.

The preventive effects of prohexadione-Ca on shoot growth and fire blight development on apples is a well-known phenomenon (Deckers and Daemen 1999). Prohexadione-Ca was tested on two different rosaceous plants, loquat and quince, at the same rates as those applied to apples. The primary effect of prohexadione-Ca is the control of shoot growth (Evans et al. 1997; Greene 1999; Rademacher and Kober 2003; Yoder et al. 1999) in apple, pear, plum and sweet and sour cherry trees (Basak and Rademacher 2000; Elfving et al. 2003). A suppression of disease incidence was not associated with substantial growth control on pears (Buban et al. 2002), which was comparable to the results on loquat and quince in this study. These results indicate that the effects of prohexadione-Ca on the incidence of fire blight are primarily the result of physiological changes (Rademacher 2004). In addition, Unrath (1999) pointed out that varied results were obtained with prohexadione-Ca in climatically different regions.

Benzothiadiazole (acibenzolar-S-methyl) is known to induce systemic acquired resistance (SAR) against fire blight. Benzothiadiazole mimics the role of salicylic acid in defense reactions, and treated plants produce pathogenesis-related proteins, which are able to degrade some of the bacterial cell walls (Kessmann et al. 1996; Oostendorp et al. 2001; Thomson et al. 1999a, b). Weekly applications of benzothiadiazole (Actigard 50 WG, Novartis) provided 81% control compared with 97.6% with streptomycin on apples (Maxson and Jones 1999). In this study, benzothiadiazole + metalaxyl provided on average 53% disease control for the 2 yr on the loquat cultivar Cukurgobek, but its effectiveness was very low on the two quince cultivars tested. This suggests that the mode of action of the products does not work in the same way on different hosts. The other aspect that should be considered is the very low rate of benzothiadiazole in the mixture we used (4%). A higher degree of control could be expected with a higher concentrated form of the chemical, provided that there is no synergistic effect of metalaxyl. This hypothesis should be further tested on loquat and quince.

Harpin protein provided broad-spectrum protection of plants against fungal, bacterial and viral pathogens (Anonymous 2002; Fontanilla et al. 2005; Jones 2001; Momol et al. 1999; Wei and Beer 1996). However, it was not as effective on loquat compared to our previous studies on other pome fruits, notably pears (Bastas and Maden 2004). On quince cultivars, harpin protein generally provided better control than the other treatments, except streptomycin.

Streptomycin was effective in preventing the shoot blight phase of the disease on quince and loquat; however, the use of this chemical must be limited to high disease pressure conditions. In the control of fire blight, copper compounds can be effective only at low and medium disease severities (van der Zwet and Keil 1979), and the rate of control is lower on susceptible host pears (Dimova 1990). These alternative chemicals to copper have shown some promise for controlling the shoot blight phase of fire blight on quince and loquat.

We obtained very low disease control of the shoot blight phase of fire blight with the copper compound alone or in combination with benzothiadiazole and harpin protein on quince. The addition of copper to benzothiadiazole reduced the effectiveness of the latter on loquat. Romero et al. (2001) also found that the addition of a copper compound to plant activators did not affect the performance of the plant activators. In contrast to our data from this study, some researchers obtained increasing yield and lower disease onset with the application of plant activator + fungicide mixtures (Anonymous 1997). Addition of copper salts did not improve the effectiveness of some of the chemicals but rather reduced it. All the chemicals except humic acid were more effective than copper compound alone.

Even when they were more effective than copper salts, the low disease control obtained from treatments with prohexadione-Ca, benzothiadiazole and harpin protein can be attributed to the inoculation method, high inoculation density, and host/cultivar susceptibility. When these situations are taken into consideration, improved performance may occur in natural infections with repeated applications during the growing season. Therefore, repeated applications should be considered in situations where disease epidemics are anticipated. Humic acid applications were ineffective as a fire blight disease control method both on loquat and quince cultivars. Humic acid should not be used as foliar application on loquat and quince during the growing season. Its negative effect should be further tested under different climatic conditions and with different application doses.

It is important to note that host resistance inducers have to be applied prophylactically against pathogen infections; they should be used 1-3 wk prior to a possible infection risk or inoculation by Erwinia amylovora. It will be necessary to find the right strategy for the application of these compounds in different areas. Prohexadione-Ca, benzothiadiazole and harpin protein application should be seen as a complementary action in the whole process of fire blight control measures.