Optimizing Efficacy of New Postharvest Fungicides and Evaluation
`of Sanitizing Agents for Managing Citrus Green Mold
`
`Loukas Kanetis, Department of Plant Pathology, University of California, Riverside 92521; Helga Förster, De-
`partment of Plant Pathology, University of California, Davis 95616; and James E. Adaskaveg, Department of Plant
`Pathology, University of California, Riverside
`
`ABSTRACT
`Kanetis, L., Förster, H., and Adaskaveg, J. E. 2008. Optimizing efficacy of new postharvest fun-
`gicides and evaluation of sanitizing agents for managing citrus green mold. Plant Dis. 92:261-
`269.
`
`Three new fungicides, azoxystrobin, fludioxonil, and pyrimethanil, that belong to different
`chemical classes are highly effective in managing citrus green mold and are being registered for
`postharvest use in the United States. Recirculating in-line drenches provided a significantly
`improved efficacy compared with standard low-volume spray applications. To prevent pathogen
`contamination of drench solutions, two oxidizing disinfectants, sodium hypochlorite and hydro-
`gen peroxide/peroxyacetic acid (HPPA) solutions, were evaluated. Inhibition of conidial germi-
`nation of Penicillium digitatum was dependent on the pH of the solution and the exposure time
`for each sanitizing agent. Chlorine (50 mg/liter) and HPPA (2,700 mg/liter) effectively inhibited
`germination in 40- and 240-s exposures, respectively, at pH 7. All fungicides tested were com-
`patible and effective with HPPA, whereas fludioxonil, azoxystrobin, and thiabendazole, but not
`imazalil and pyrimethanil, were compatible with chlorine. In laboratory studies, sodium bicar-
`bonate (SBC, 3%) significantly increased the efficacy of the three fungicides (250 mg/liter) and
`had no adverse effect on their stability in aqueous solutions. Fludioxonil (300 mg/liter)-SBC
`mixtures were still highly effective when applied 24 h after fruit inoculation. In experimental
`packingline studies, SBC or SBC-chlorine improved the efficacy of fludioxonil, whereas azox-
`ystrobin was effective with and without these additives. Heating of drench solutions of fludi-
`oxonil (300 mg/liter) to 50°C did not improve decay control. In conclusion, in-line recirculating
`drench applications and fungicide-sanitizer-SBC mixtures significantly increased fungicide
`efficacy and provide an integrated approach for optimizing fungicide efficacy. These strategies
`also should minimize the selection for resistant isolates of the pathogen.
`
`Additional keyword: sanitation
`
`
`
`Green mold caused by Penicillium digi-
`tatum (Pers.) Sacc. is the most important
`postharvest decay of citrus fruit. In Cali-
`fornia and other citrus production areas,
`postharvest fungicide treatments are an
`integral part of the management of green
`mold. In contrast to other fresh fruit com-
`modities in California, citrus fruit are har-
`vested, stored, and processed almost year-
`round. Because inoculum is continuously
`present and reintroduced, decay manage-
`ment in packinghouses is a challenging
`task. Sodium ortho-phenylphenate, thia-
`bendazole (TBZ), and imazalil have been
`available for postharvest use for many
`years, but resistance to these fungicides is
`very common (5). Currently, three new
`fungicides, azoxystrobin, fludioxonil, and
`pyrimethanil, are being introduced in the
`United States that all are highly effective
`
`Corresponding author: J. E. Adaskaveg
`E-mail: jim.adaskaveg@ucr.edu
`
`Accepted for publication 19 September 2007.
`
`doi:10.1094 / PDIS-92-2-0261
`© 2008 The American Phytopathological Society
`
`in reducing the incidence of green mold
`but differ in their post-infection efficacy
`and antisporulation activity (2,15). Thus,
`when aqueous fungicide solutions were
`applied 9
`to 21 h after
`inoculation,
`pyrimethanil provided the highest level of
`green mold decay control, whereas the
`efficacy of fludioxonil and azoxystrobin
`was very high at the early timings but di-
`minished as application time after inocula-
`tion increased. Pyrimethanil, however, was
`less effective in inhibiting sporulation of
`the pathogen. In addition, azoxystrobin-
`fludioxonil mixtures were significantly
`more effective than single-fungicide treat-
`ments.
`To maximize the performance of post-
`harvest
`treatments, various approaches
`have been used previously that sometimes
`significantly increased treatment efficacy.
`Thus, for citrus fruit, the addition of so-
`dium bicarbonate (SBC) increased treat-
`ment performance in green mold decay
`management compared with the use of
`TBZ or imazalil alone (25,27). Further-
`more, solutions of TBZ or imazalil heated
`to 41 or 50°C resulted in an improved
`efficacy (7,27). Newer fungicides such as
`fludioxonil and pyrimethanil recently have
`
`treatments
`been evaluated as heated
`(7,22,23,26). Recirculating, in-line drench
`applications to inoculated plum fruit were
`shown to reduce decay incidence signifi-
`cantly compared with in-line, low-volume
`spray applications (12). These strategies
`also need to be evaluated for the three new
`postharvest fungicides for citrus fruit. In
`addition, sanitizing treatments that can be
`used in recirculating fungicide solutions
`have to be evaluated for their interaction
`with fungicide efficacy. Because recircu-
`lating fungicide solutions are being used
`for an extended time period, the build-up
`of microbial populations has to be pre-
`vented to avoid contamination of healthy
`fruit and to minimize selection of fungi-
`cide-resistant pathogen propagules.
`Thus, the objectives of this study were
`to (i) compare the efficacy of azoxystrobin,
`fludioxonil, and pyrimethanil in reducing
`green mold of lemon fruit using different
`postharvest application methods; (ii) com-
`pare the effect of the sanitizing agents
`sodium hypochlorite and hydrogen perox-
`ide/peroxyacetic acid (HPPA) on viability
`of conidia of P. digitatum in regard to ex-
`posure time and pH of the solutions; (iii)
`determine interactions of SBC and sanitiz-
`ing agents on the stability and efficacy of
`citrus postharvest fungicides; and (iv) de-
`termine the efficacy of nonheated and
`heated solutions of fludioxonil and fludi-
`oxonil-SBC.
`
`MATERIALS AND METHODS
`Fungal isolates. Two single-spore iso-
`lates of P. digitatum were used in the stud-
`ies. Isolate Pd was sensitive to imazalil and
`TBZ (50% effective concentration [EC50]
`values for mycelial growth = 0.05 and 0.08
`mg/liter, respectively), whereas
`isolate
`2152 was resistant to both fungicides
`(EC50 values = 0.65 and > 7.82 mg/liter,
`respectively). Isolates were maintained as
`mycelial plugs in sterile water at 4°C for
`up to 1 year. For conidial production, iso-
`lates were grown on potato dextrose agar
`(PDA) (Difco Laboratories, Detroit, MI)
`and incubated at 25°C for 5 to 10 days.
`Spore suspensions were prepared in 0.01%
`Tween 20 (Sigma-Aldrich, St. Louis, MO)
`and adjusted to the appropriate concentra-
`tions with sterile water using a hemacy-
`tometer.
`Fungicides and sanitation agents.
`Fungicides used were formulated products
`of azoxystrobin (Abound 2.08F; Syngenta
`
`Plant Disease / February 2008 261
`
`

`

`Crop Protection, Greensboro, NC), fludi-
`oxonil (Scholar 50WP; Syngenta Crop
`Protection),
`imazalil
`(Freshgard 700;
`Janssen Pharmaceutica, Titusville, NJ),
`pyrimethanil (Penbotec 400SC; Janssen
`Pharmaceutica), and thiabendazole (Decco
`Salt No. 19; Decco-Cerexagri, Monrovia,
`CA). Standard solutions of fungicides for
`gas chromatography were prepared using
`technical-grade materials that were sup-
`plied by the manufacturers. Sodium bicar-
`bonate (Arm and Hammer; Church and
`Dwight Co. Inc., Princeton, NJ) and the
`sanitizing agents sodium hypochlorite
`(household bleach containing approxi-
`mately 5.25% NaOCl) and commercial
`HPPA (ZeroTol, 27% hydrogen peroxide;
`BioSafe Systems, Glastonbury, CT) were
`used alone or in combination with fungi-
`cides. All concentrations were based on the
`active ingredient of each chemical.
`Effect of application method on fun-
`gicide efficacy in reducing citrus green
`mold. Lemon fruit (Citrus limon (L.) N.L.
`Burm.) cv. Eureka were used in fruit in-
`oculation studies. Fruit for the laboratory
`studies were obtained from a research
`orchard at the University of California,
`Riverside; whereas, for the experimental
`packingline studies, freshly picked fruit
`were obtained from local packinghouses.
`Lemon fruit used in these studies were not
`treated with preharvest fungicides and
`were randomized for each experiment
`based on size and maturity. The flavedo
`and albedo of lemon fruit were punctured
`(one 1-by-2-mm wound per fruit) with a
`nail-like, stainless-steel probe without
`injuring the juice sacks below the albedo.
`For inoculation, a 20-µl drop of inoculum
`(106 spores/ml) from the imazalil- and
`TBZ-resistant isolate of P. digitatum was
`placed on each wound. Inoculated fruit
`were placed into fruit trays in cardboard
`fruit boxes that were covered with plastic
`bags and incubated at 20°C for 12 to 14 h.
`Postharvest
`fungicide
`treatments were
`conducted on an experimental packingline
`at the University of California Kearney
`Agricultural Center, Parlier. High-volume,
`in-line drenches were applied to lemon fruit
`by pumping a fungicide solution from a 70-
`liter reservoir above a perforated steel dis-
`tribution pan (91-by-91-cm area with 127 5-
`mm-diameter holes evenly distributed) ap-
`proximately 16 cm over a moving roller
`bed. Fungicides were used at 500 mg/liter
`and treatment volumes were equivalent to
`33 liters per 10,000 kg of fruit. Fungicide
`treatments were followed by a low-volume
`spray application with a 1:10 dilution
`(vol/vol) of a storage fruit coating (PacRite
`505-35; Pace International, Seattle, WA)
`using a controlled droplet applicator (CDA;
`Decco-Cerexagri) that was positioned in the
`center of the treatment area and approxi-
`mately 20 cm above the roller bed. Treat-
`ment volumes for the spray applications of
`fruit coatings were adjusted to 8.3 liters per
`10,000 kg of fruit by regulating fruit coating
`
`262 Plant Disease / Vol. 92 No. 2
`
`output volumes and speed of fruit move-
`ment through the treatment area. Low-
`volume fungicide treatments were applied
`in the diluted storage fruit coating over
`either a roller bed or a horse-hair brush bed
`using the same specifications as above for
`the fruit-coating applications with fungicide
`rates of 4,000 mg/liter. Based on these rates
`and treatment volumes, low-volume spray
`applications were done at twice the rate
`compared with the in-line drench applica-
`tions (i.e., 33.2 g per 10,000 kg of fruit for
`the spray applications). Treatment times for
`all application methods were generally be-
`tween 12 and 15 s. Control fruit were
`treated only with water and fruit coating in
`each of the three application methods. Be-
`tween treatments, fungicide reservoirs and
`tubing, as well as treatment beds, were
`cleaned with a commercial alkaline deter-
`gent (PacFoam Plus; Pace International) and
`then thoroughly rinsed with water. Fruit
`then were incubated in fruit boxes for 6 to
`7 days at 20°C. Treatments were random-
`ized among boxes and boxes were covered
`with plastic bags. For evaluation, fruit
`were inspected carefully for green mold
`development that was either easily visible
`as mycelium- or conidia-covered decay or
`present as soft, often watery lesions around
`the inoculation site. Decay incidence was
`based on the number of decayed fruit of
`the total number of fruit inoculated. There
`were four replications of 24 fruit for each
`treatment and the experiment was done
`twice. To determine the standardized im-
`proved efficacy in the comparisons of fun-
`gicide application methods for each fungi-
`cide and application method, the incidence
`of decay first was subtracted from that of
`the appropriate control. This value then
`was calculated as a percentage of the inci-
`dence of the control treatment (i.e., the
`standardized efficacy). The standardized
`improved efficacy was determined by sub-
`tracting the incidence of decay of the least
`effective method from that of each of the
`other two methods.
`Effect of sanitizing agents on viability
`of P. digitatum conidia. In laboratory stud-
`ies, the efficacy of HPPA was compared
`with that of sodium hypochlorite (prepared
`from commercial household bleach) using
`selected exposure times and acidities of the
`test solutions. Dilutions of both sanitizing
`agents at selected pH values were prepared
`in sterile distilled water with HPPA at
`3,000 mg/liter (based on hydrogen perox-
`ide) and free chlorine (HOCl and OCl–) at
`55.5 mg/liter. The concentration of free
`chlorine was verified using autodilution
`ampoules for the colorimetric analysis of
`chlorine (Vacuettes R-2505D; Chemetrics,
`Inc., Calverton, VA). The acidity of the test
`solutions was adjusted to pH 6, 7, and 8
`using 0.1 N sodium hydroxide for HPPA
`and 1 N hydrochloric acid for sodium hy-
`pochlorite. For testing the effect on viabil-
`ity of conidia, 1 ml of conidial suspension
`(5 × 107 conidia/ml) of the imazalil- and
`
`TBZ-resistant isolate of P. digitatum was
`pipetted into sterile glass test tubes con-
`taining 9 ml of the test solutions, resulting
`in final concentrations of 2,700 mg/liter
`for hydrogen peroxide and 50 mg/liter for
`free chlorine. Control tubes contained 9 ml
`of sterile distilled water. Tubes were vor-
`texed and, after selected times (30, 60,
`120, and 240 s for HPPA; 10, 20, 40, and
`60 s for free chlorine), reactions were
`stopped by adding 2 µl of each reaction
`mixture to 1.998 ml of sterile water in
`microcentrifuge tubes, thus diluting the
`sanitizing agents 1,000 times. For each
`reaction mixture, three 20-µl drops then
`were placed onto PDA in petri dishes. Af-
`ter incubation for 16 to 18 h at 25°C, vi-
`ability of conidia was assessed based on
`conidial germination. Using an inverted
`microscope (Axiovert S100; Zeiss, Ger-
`many), 50 conidia in each drop were
`evaluated for germination. Results were
`expressed as percent of germinated conidia
`of the total conidia evaluated. This experi-
`ment was performed three times.
`Stability of citrus postharvest fungi-
`cides in the presence of sanitizing agents
`and SBC. Concentrations of fludioxonil,
`pyrimethanil, imazalil, and TBZ were de-
`termined by gas chromatography after
`incubation in solutions of HPPA, sodium
`hypochlorite, or mixtures of sodium hy-
`pochlorite with SBC. For this, fungicide
`solutions (each fungicide at 100 mg/liter)
`were prepared without and with the addi-
`tion of free chlorine at 100 mg/liter, HPPA
`at 2,700 mg/liter (based on hydrogen per-
`oxide), or a mixture of free chlorine at 100
`mg/liter and 3% (wt/vol) SBC. After 0.5
`and 8 h at 25°C, fungicide concentrations
`were determined using a gas chromato-
`graph (Model 5890 GC; Agilent, Santa
`Clara, CA) equipped with an autosampler,
`an autoinjector, a nitrogen-phosphorus
`detector, and a Zebron ZB-35 capillary GC
`column (15 m long by 0.53-mm inner di-
`ameter). The following conditions were
`used: 300°C injection port temperature,
`350°C detector temperature, 200°C initial
`temperature, and 260°C final temperature.
`Temperature ramping was 10°C/min. Cali-
`bration was done using fungicide standards
`of 0.5, 1, 2, and 5 µg/ml that were prepared
`from technical-grade material. Fungicide
`concentrations were compared with the
`controls (initial concentration of the fungi-
`cides when mixed in water). This experi-
`ment was performed twice.
`Effect of sanitizing agents on the effi-
`cacy of postharvest fungicides in reduc-
`ing citrus green mold. In laboratory stud-
`ies, aqueous solutions of azoxystrobin,
`fludioxonil, and pyrimethanil (250 mg/liter
`each) were prepared immediately before (0
`h) or 8 h before use either without addi-
`tives or with the addition of free chlorine
`at 100 mg/liter, hydrogen peroxide at
`2,700 mg/liter, 100 mg of free chlorine +
`3% (wt/vol) SBC, or hydrogen peroxide at
`2,700 mg/liter + 3% (wt/vol) SBC. Lemon
`
`

`

`fruit were inoculated with the imazalil- and
`TBZ-resistant isolate of P. digitatum, incu-
`bated at 20°C for 14 to 16 h, and then
`treated with the above solutions (10 liters)
`by dipping for 30 s. Treated fruit were
`allowed to air dry, incubated at 20°C, and
`evaluated for decay development as de-
`scribed above. The experiment was per-
`formed twice, with four replications per
`treatment and 12 fruit per replication.
`Efficacy of nonheated and heated so-
`lutions of fludioxonil and of mixtures of
`fludioxonil or azoxystrobin with SBC
`and chlorine in reducing citrus green
`mold. The efficacy of fludioxonil, SBC,
`and fludioxonil-SBC mixtures was evalu-
`ated in laboratory studies. The effect of
`heated treatments of fludioxonil and the
`addition of SBC or SBC-chlorine to fludi-
`oxonil or azoxystrobin solutions for con-
`trolling green mold was evaluated in
`experimental packingline studies. In labo-
`ratory studies, lemon fruit were inoculated
`with the imazalil- and TBZ-resistant iso-
`late of P. digitatum as described above and
`treated after 14 or 24 h using sprays of
`aqueous solutions of fludioxonil (300
`mg/liter), 3% (wt/vol) SBC, or a mixture
`of these two materials at the same concen-
`trations. Sprays were applied using an
`atomizer (Model 15-RD; DeVilbiss Health
`Care, Somerset, PA) at approximately 8.3
`liters per 10,000 kg of fruit (0.5 ml/fruit).
`For the packingline studies, fruit in the
`repeated experiments were inoculated with
`the imazalil- and TBZ-resistant or -sensitive
`isolates and treated after 12 to 15 h using
`aqueous
`in-line drenches as described
`above for the following treatments: fludi-
`oxonil at 300 mg/liter, azoxystrobin at 500
`mg/liter, and mixtures of each fungicide
`with 3% (wt/vol) SBC or SBC plus free
`chlorine at 50 mg/liter. Drench solutions
`for fludioxonil were held at ambient tem-
`perature (25°C) or were heated to 50°C
`and fruit to be treated were adjusted to
`25°C. Between replications of heated
`treatments, temperatures were readjusted
`to 50°C. The pH of the aqueous fungicide
`solutions was 6.5 to 7, whereas the fungi-
`cide-SBC mixtures had a pH of 8 to 8.5.
`Fungicide drench treatments were fol-
`lowed by low-volume spray applications
`with diluted storage fruit coating, and fruit
`then were incubated, stored, and evaluated
`as described above. For each of the four
`replications, 12 fruit were used in the labo-
`ratory studies and 24 fruit in the pack-
`ingline studies, and experiments were con-
`ducted twice.
`Statistical analysis of data. Percentage
`data were arcsine transformed. Bartlett’s
`test for homogeneity of variances was
`performed for repeated experiments. Data
`sets with homogeneous variances (P <
`0.05) were combined and then analyzed
`using a one-, two-, or three-way classifica-
`tion of data depending on the experiment.
`In experiments with multiple comparisons,
`a balanced factorial design was used for
`
`treatment comparisons. For error control,
`all treatments were in a randomized com-
`plete-block design. Values were analyzed
`using general linear model or analysis of
`variance and least significant difference
`(LSD) mean separation procedures of SAS
`(version 9.1; SAS Institute, Cary, NC).
`
`RESULTS
`Effect of application method on fun-
`gicide efficacy in reducing citrus green
`mold. The efficacy of azoxystrobin, fludi-
`oxonil, and pyrimethanil against green
`mold of lemon fruit was compared using
`low-volume spray applications over roller
`or brush treatment beds and high-volume,
`recirculated, in-line drench applications.
`These studies resulted in highly significant
`(P < 0.01) differences between treatment
`methods. In addition, there was a highly
`significant interaction (P < 0.01) between
`fungicide and application method, indicat-
`ing that fungicides performed differently
`using the three methods. Thus, data in
`Figure 1 are presented separately for each
`fungicide. As indicated in these compari-
`sons, for all three fungicides the high-
`volume, in-line drenches were significantly
`more effective than the low-volume spray
`applications (CDA). There was, however,
`no
`significant difference
`in efficacy
`whether low-volume sprays were done
`over a roller or a brush treatment bed. Ef-
`fectiveness of azoxystrobin and fludioxonil
`was increased by 15 and 25%, respectively,
`when applied using an in-line drench com-
`pared with the spray application over a
`roller bed. For pyrimethanil, the efficacy
`
`was improved by 15.7% compared with
`the spray application over a brush bed
`(Fig. 1).
`Effect of sanitizing agents on viability
`of P. digitatum conidia. The effect of chlo-
`rine and HPPA on the viability of conidia of
`P. digitatum was evaluated for different
`exposure times and pHs. For both sanitizers,
`solutions at pH 6 and 7 were significantly
`(P < 0.01) more effective in killing conidia
`than solutions at pH 8. Because a significant
`interaction (P < 0.01) occurred between pH
`and exposure time, comparisons of exposure
`times of each sanitizer then were done for
`each pH value.
`Exposures in solutions of free chlorine
`at 50 mg/liter for 10 s at pH 6, 7, or 8 re-
`duced conidial germination to 5.6, 11.7,
`and 35.1%, respectively, compared with
`>95% germination in the water control
`(Fig. 2A). Less than 1% of the conidia
`germinated after 20- or 40-s exposures at
`pH 6 and 7, respectively; whereas, after a
`60-s exposure at pH 8, 5.5% of the conidia
`still germinated. For HPPA, higher concen-
`trations and longer exposure times were
`required to reduce viability of P. digitatum
`conidia compared with chlorine. At pH 6
`and 7, 240-s exposures resulted in no ger-
`mination whereas, at pH 8, 6.9% of the
`conidia were still viable compared with
`>95% germination in the water control
`(Fig. 2B). At pH 5, the acidity level of a
`nonadjusted aqueous dilution of HPPA, no
`germination was observed after a 120-s
`exposure (data not shown).
`Stability of citrus postharvest fungi-
`cides in the presence of sanitizing agents
`
`Fig. 1. Comparative efficacy of postharvest fungicides against green mold of lemon fruit using three
`application methods. In the low-volume sprays, fungicides at 4,000 mg/liter were applied in diluted
`storage fruit coating using a controlled droplet applicator (CDA) that was positioned over either a
`brush or roller treatment bed. In the in-line drench applications, aqueous fungicide applications at 500
`mg/liter were followed by a low-volume spray application with diluted storage fruit coating. For each
`fungicide, a standardized improved efficacy was calculated (as indicated in the Materials and Methods)
`and horizontal bars with the same letter indicate that treatment means were not significantly different
`(P < 0.05) following an analysis of general linear models and least significant difference mean separa-
`tion test procedures.
`
`Plant Disease / February 2008 263
`
`

`

`and SBC. The stability of fludioxonil,
`pyrimethanil, imazalil, and TBZ in solu-
`tions of HPPA (hydrogen peroxide at 2,700
`mg/liter), free chlorine (100 mg/liter), or
`mixtures of SBC (3%) and free chlorine
`(100 mg/liter) was evaluated after 0.5- and
`8-h incubations using gas chromatography.
`Both fludioxonil and TBZ were stable in
`the three sanitizing solutions, as indicated
`by the presence of fungicide residues in
`the solutions similar to the controls (i.e.,
`the initial concentration of the fungicides
`when mixed in water; Fig. 3). In addition,
`concentrations of both fungicides remained
`quite stable for 7 days after preparation
`(data not shown). At this time, a 7.5%
`decrease in fungicide concentration was
`observed only in solutions containing both
`chlorine and 3% SBC. Pyrimethanil and
`
`imazalil also were found to be stable in
`HPPA but not in chlorine or chlorine-SBC.
`After 0.5 and 8 h of exposure in chlorine,
`the pyrimethanil concentration was re-
`duced to 60 and 45% and the imazalil con-
`centration was reduced to 90 and 5%,
`respectively, of the water control. Degrada-
`tion of these fungicides in chlorine-SBC
`mixtures was significantly higher in some
`cases than in chlorine alone. Thus, after
`0.5 and 8 h of exposure, the pyrimethanil
`concentration was reduced to 40 and
`17.5% and the imazalil concentration was
`reduced to 35 and 5%, respectively, com-
`pared with the water control (Fig. 3).
`Effect of sanitizing agents on the effi-
`cacy of postharvest fungicides in reduc-
`ing citrus green mold. The stability of
`azoxystrobin, fludioxonil, and pyrimetha-
`
`
`Fig. 2. In vitro effect of chlorine and a mixture of hydrogen peroxide and peroxyacetic acid (HPPA) on
`viability of conidia of Penicillium digitatum. Conidia were incubated in solutions of A, free chlorine at
`50 mg/liter or B, 2,700 mg of HPPA (based on the concentration of hydrogen peroxide) at pH 6, 7, or 8
`for selected times. For each sanitizing agent at a specific pH, horizontal bars with the same letter indi-
`cate that treatment means were not significantly different (P < 0.05) following an analysis of variance
`and least significant difference mean separation test procedures.
`
`264 Plant Disease / Vol. 92 No. 2
`
`nil also was evaluated in experiments
`where fruit were treated 14 to 16 h after
`inoculation with fungicide-sanitizer solu-
`tions that were prepared either immedi-
`ately before (0 h) or 8 h before treatment.
`A change in fungicide efficacy from that of
`the aqueous control solution (i.e., without
`additives) was considered indicative of a
`positive or negative reaction between fun-
`gicide and sanitizer. For azoxystrobin and
`fludioxonil, no significant interaction was
`found between preparation time and treat-
`ment efficacy for each fungicide. Thus,
`data for the two timings were combined in
`Figure 4A and B. For pyrimethanil, there
`was a significant
`interaction between
`preparation time and fungicide efficacy
`and, thus, data are presented for each for
`the two preparation times (Fig. 4C). Using
`the aqueous fungicide solutions alone at
`reduced concentrations of 250 mg/liter,
`decay incidence was 37.5% for azox-
`ystrobin, 55.0% for fludioxonil, and 5.5 to
`10% for pyrimethanil compared with
`100% in the untreated control.
`For azoxystrobin and fludioxonil, there
`were no significant differences in decay
`incidence between aqueous preparations
`and preparations in free chlorine at 100
`mg/liter or HPPA (hydrogen peroxide at
`2,700 mg/liter), indicating that the two
`sanitizers did not affect fungicide perform-
`ance (Fig. 4A and B). Efficacy for both of
`these fungicides, however, was improved
`significantly when SBC was added to the
`sanitizers, as
`indicated by
`low
`levels
`(≤9.9%) of decay incidence. For pyri-
`methanil, at both timings the addition of
`HPPA or HPPA-SBC resulted in decay
`levels similar to the aqueous fungicide
`treatment, indicating no interaction. The
`efficacy of this fungicide, however, was
`reduced significantly in solutions contain-
`ing chlorine in the 0- and 8-h-old prepara-
`tions, except when chlorine was mixed
`with SBC and used immediately after
`preparation. Thus, decay incidences for the
`0- and 8-h treatments with solutions con-
`taining chlorine were 49.5 and 72.4% and
`for solutions containing chlorine and SBC
`were 8.7 and 32.3%, respectively (Fig.
`4C).
`Efficacy of nonheated and heated so-
`lutions of fludioxonil and of mixtures of
`fludioxonil or azoxystrobin with SBC
`and chlorine in reducing citrus green
`mold. In laboratory studies using fludi-
`oxonil, SBC, and fludioxonil-SBC mix-
`tures, treatment (P < 0.01), treatment tim-
`ing (14 and 24 h; P < 0.02), and their
`interaction (P < 0.01) were significant and,
`thus, treatments were compared for each
`timing separately (Fig. 5). Both fludioxonil
`and SBC significantly reduced the inci-
`dence of green mold of lemon fruit com-
`pared with the untreated control when
`applied 14 or 24 h after inoculation with P.
`digitatum (Fig. 5). Treatment efficacy at
`both timings was increased significantly
`when fludioxonil-SBC mixtures were used
`
`

`

`compared with the single treatments (i.e.,
`fludioxonil or SBC). Decay incidence in
`these latter treatments was 5.2 and 10.4%
`for the 14- and 24-h timings, respectively,
`compared with 100% incidence in the
`control.
`In an experimental packingline study,
`the efficacy of nonheated (25°C) and
`heated (50°C) drench
`treatments with
`fludioxonil and
`fludioxonil-SBC was
`evaluated. There were significant (P <
`0.01) differences between the treatments
`but no significant differences between
`temperatures (P = 0.44), and there was no
`interaction between temperature and fun-
`gicide efficacy (P = 0.89). As in the labora-
`tory study, the efficacy of fludioxonil was
`improved significantly when mixed with
`SBC (Fig. 6A). When applied 14 h after
`inoculation, fludioxonil reduced the inci-
`dence of decay to 17.7%, whereas the
`mixture of fludioxonil with SBC or SBC-
`chlorine reduced the incidence to 3.9 and
`4.0%, respectively, compared with the
`nontreated control with 68.1% decay.
`Drench applications with azoxystrobin,
`azoxystrobin-SBC, or azoxystrobin-SBC-
`chlorine reduced decay incidence to 4.2,
`4.2, or 3.7%, respectively, compared with
`the nontreated control with 74.8% decay
`(Fig. 6B).
`
`DISCUSSION
`Three new postharvest fungicides, azox-
`ystrobin, fludioxonil, and pyrimethanil, are
`being introduced currently for the man-
`agement of citrus green mold. Because
`resistance against the previously registered
`fungicides is widespread in populations of
`P. digitatum (5,10), the judicious use of the
`new compounds is critical to ensure their
`lasting efficacy. For this, the utilization of
`fungicide mixtures or rotations of fungi-
`cides have been suggested if fruit receive
`more than one treatment, as is done com-
`monly with stored lemon fruit in California
`(15). In the current study, additional strate-
`gies were investigated such as the optimi-
`zation of fungicide application methods.
`Furthermore, we evaluated the addition of
`SBC and sanitation agents to improve
`fungicide efficacy and to reduce the spread
`of pathogen inoculum. Thus, the overall
`aim of this study was to define conditions
`that provide the highest efficacy of the new
`postharvest fungicides and address the
`need of the citrus industry for reliable,
`integrated green mold management.
`Low- and ultra-low-volume in-line fun-
`gicide spray applications to wet, washed
`fruit have been the standard treatment
`method of the California fruit industries
`for many years. These treatments are eco-
`nomical and environmentally sound be-
`cause run-off is limited and, consequently,
`few disposal problems arise. Low-volume
`spray applications, however, can have limi-
`tations on types of fruit where adequate
`residues are difficult to achieve. Similar to
`a previous study with plum fruit (12),
`
`comparisons of postharvest application
`methods in the current study showed that,
`for all three fungicides evaluated, high-
`volume, in-line drench applications were
`significantly more effective in reducing
`decay of inoculated lemon fruit than low-
`volume spray applications. In-line drench
`applications were superior in performance,
`although fungicides in the drench applica-
`tions were used at half rates of the spray
`applications (as calculated based on fruit
`weight treated). Thus, fungicide coverage
`probably is improved using drench appli-
`cations and fungicide deposition
`into
`wound sites may have increased.
`Postharvest fungicides commonly have
`been applied to citrus fruit after harvest as
`nonrecovery sprays, dips, or foams over
`brushes or rollers (1). There is currently an
`
`increasing trend in the California posthar-
`vest citrus industry for the use of recircu-
`lating, in-line drench applications. Thus,
`these applications are proving their practi-
`cability
`in commercial packinghouses.
`Because fungicide solutions for in-line
`drenches can be recirculated for extended
`use, these high-volume applications can
`still be done economically. By treating
`only fruit that are washed and sanitized,
`contamination of the fungicide solutions
`with pathogen propagules can be mini-
`mized.
`To reduce the amount of remaining in-
`fective inoculum in the fungicide solu-
`tions, we evaluated the use of sanitation
`treatments. For TBZ, chlorine has been
`used routinely to disinfect postharvest
`treatment solutions (27). Chlorine and
`
`Fig. 3. Effect of sanitizing agents and sodium bicarbonate (SBC) on the stability of citrus postharvest
`fungicides. TBZ = thiabendazole. Fungicides (each 100 mg/liter) were incubated in solutions of hy-
`drogen peroxide/peroxyacetic acid (HPPA; hydrogen peroxide at 2,700 mg/liter), free chlorine (100
`mg/liter), or mixtures of SBC (3%, wt/vol) and free chlorine (100 mg/liter). After 0.5 and 8 h, fungi-
`cide concentrations were determin