Available online at www.sciencedirect.comJournal of

www.elsevier.com/locate/yjipa

Journal of Invertebrate Pathology 97 (2008) 134–141

INVERTEBRATE

PATHOLOGY

Natural occurrence of Beauveria bassiana in Hypothenemushampei (Coleoptera: Curculionidae) populations

in unsprayed coffee fields

Arnulfo J. Monzon a,b,c,*, Falguni Guharay d, Ingeborg Klingen a

a Norwegian Institute for Agricultural and Environmental Research (Bioforsk) Høgskoleveien 7, N-1432 As, Norwayb Norwegian University of Life Sciences, Department of Plant and Environmental Sciences, P.O. Box 5003, N-1432 As, Norway

c National Agricultural University, Nicaragua (UNA) Apdo. 453 Managua, Nicaraguad Tropical Agricultural Research and Higher Education Centre (CATIE), Nicaragua, Apdo. 4830 Managua, Nicaragua

Received 20 April 2007; accepted 20 July 2007Available online 1 August 2007

Abstract

Three unsprayed coffee farms (farm 1, 2 and 3) were studied for the natural occurrence of the insect pathogenic fungus Beauveria

bassiana in Hypothenemus hampei populations throughout the rainy season of 2004 (July–November) and 2005 (July–December).B. bassiana infections were found during most sampling dates in both years, on all three farms. The B. bassiana infection levels werehigher in 2005 than in 2004 with mean prevalence of 12.1% and 2.7%, respectively. No consistent significant differences in infection levelbetween farms were found in any of the years. B. bassiana infection levels fluctuated widely throughout the season, and peaked at 13.5%on farm 3 in 2004 and at 44.0% on farm 1 in 2005. The H. hampei population was significantly higher in 2004 than in 2005, with 6.9% ofthe berries infested in 2004 and only 0.7% in 2005. In both years, the H. hampei infestation level was significantly higher on farm 2. Noconsistent significant differences in H. hampei infestation levels were found between sampling dates on any of the farms. H. hampei infes-tation levels fluctuated throughout both seasons, and peaked at 15.3% on farm 2 in 2004 and 2.2% on farm 2 in 2005. No consistentdensity dependent correlation between H. hampei infestation level and B. bassiana infection level was found. Correlations between cli-matic conditions and B. bassiana or H. hampei were not found.� 2007 Elsevier Inc. All rights reserved.

Keywords: Coffee berry borer; Hypothenemus hampei; Infestation level; Beauveria bassiana; Infection level; Biological control; Natural occurrence;Natural enemy

1. Introduction

The coffee berry borer, Hypothenemus hampei (Coleop-tera: Curculionidae) is the most important pest of coffeethroughout the world (Le Pelley, 1968; Dufour et al.,1999; Soto-Pinto et al., 2002; Bustillo, 2005). In America,H. hampei was first observed in Brazil in 1913 and is nowprevalent in all Central American countries, Colombia,Mexico, and the Dominican Republic (Guharay and

0022-2011/$ - see front matter � 2007 Elsevier Inc. All rights reserved.

doi:10.1016/j.jip.2007.07.008

* Corresponding author. Fax: +505 2632609.E-mail address: arnulfo.monzon@una.edu.ni (A.J. Monzon).

Monterrey, 1997; Borbon et al., 2001; Canet and Garcia,2005).

Hypothenemus hampei causes damage to berries by theboring action of adults and larvae (Le Pelley, 1968). Whenrainy season starts, females emerge from fallen berries andstart boring new berries, about 120 days after floweringwhen berries have about 20% of dry weight. They thenmake galleries where they lay their eggs. Larvae feed onberries from 10 to 26 days (Le Pelley, 1968; Wrigley,1988; Guharay and Monterrey, 1997; Baker, 1999) andoften destroy them completely. This could result in totalbean loss and premature fruit fall if the attack occurs earlyin the season (Guharay and Monterrey, 1997; Damon,

A.J. Monzon et al. / Journal of Invertebrate Pathology 97 (2008) 134–141 135

2000). After hatching from pupae inside the berries, siblingmating occurs, and females leave the berries in search ofberries in which to oviposit (Guharay and Monterrey,1997; Mathieu et al., 2001). The main source of H. hampei

infestation are infested berries that have fallen on theground (Chamorro et al., 1995; Cure et al., 1998; Bustilloet al., 1999).

Hypothenemus hampei is very difficult to control due toits cryptic life cycle, which is spent inside the coffee berry(Damon, 2000; Haraprasad et al., 2000; Rehner et al.,2006). The most common control methods used are chem-ical control and cultural practices. Cultural practicesinclude picking berries that are ripe early in the seasonand picking berries left on the plant after the harvest sea-son. It also includes collection of berries that have fallenon the ground. Manual control reduces H. hampei popula-tions, but it is labor intensive and not economically feasible(Bustillo et al., 1999; Kimani et al., 2002). Few insecticideseffectively control H. hampei (Rhodes and Mansingh, 1981;Damon, 2000). The chlorinated hydrocarbon endosulfan(chemical group: cyclodiene organochlorine), has beeneffectively used against the H. hampei. However, this highlytoxic chemical has negative effects on human health and theenvironment (Brun et al., 1989; Damon, 2000; Wegbe et al.,2003) and has been banned in several countries (Posadaand Vega, 2005) and will soon be banned in Central Amer-ica (CN-MIP et al., 2002). In addition, H. hampei resis-tance to endosulfan has been reported (Brun, 1989). Thedifficulties encountered by growers in controlling the insecthave resulted in a search for effective biological controlstrategies that are sustainable and environmentallyfriendly.

The insect pathogenic fungus Beauveria bassiana (Asco-mycota: Hypocreales) has been widely reported as a natu-ral enemy of H. hampei (De La Rosa et al., 1997;Damon, 2000; Samuels et al., 2002; De Oliveira et al.,2003; Bustillo and Posada, 1996; Guharay and Monterrey,1997; Trejo and Funez, 2004; Bustillo, 2005), and is themost common entomopathogen infecting H. hampei (Bus-tillo et al., 2002). According to Damon (2000), natural B.

bassiana infections are common and epizootics may occur,especially on the ground in fallen berries and under coolconditions with high relative humidity and heavy shade.B. bassiana is used as an inundative microbial control agentagainst H. hampei in several countries in Latin America(Bustillo and Posada, 1996; Guharay and Monterrey,1997; Van Lenteren and Bueno, 2003). To our knowledgethere are, however, no systematic studies revealing theimportance, occurrence and dynamics of B. bassiana as anaturally occurring enemy in H. hampei populationsthroughout the season, over several years and in severallocations. The aim of this research was therefore to con-duct a systematic study where the natural occurrence ofB. bassiana in three different unsprayed coffee farms wasobserved regularly throughout the rainy season over twosuccessive years. This information would contributetowards an understanding of the natural control levels

occurring in unsprayed coffee farms which might be usefulin implementing strategies for biological control.

2. Materials and methods

2.1. Location

The study was carried out from July to November 2004and from July to December in 2005 at three different cof-fee farms in Matagalpa, in the northern coffee growingzone of Nicaragua. Farm 1 (La Flor) was located at13�06 00600N, 85�47 03300W, 750 m above sea level and about200 m from farm 2 (Verania) and about 2 km from farm 3(Quitasueno). Farm 2 was located at 13�04 05200N,85�47 03200W, 730 m above sea level and about 2 km fromfarm 3. Farm 3 was located at 13�07 01000N, 85�48 01300W at800 m above sea level. One field on each coffee farm wasselected for the survey. The coffee fields selected on farm 1had an area of 1.5 ha, farm 2 had an area of 1.0 ha andfarm 3 had an area of 2.0 ha. The plant density was about5200 plants per ha and the estimated yield was 750 kg,600 kg and 1500 kg per ha on farm 1, 2 and 3, respec-tively. Neither synthetic pesticides nor B. bassiana hadbeen used in the last 5 years on any of the farms.Mechanical (machete) weed control and pruning of deadand diseased branches was carried out on all farms. Ber-ries that were ripe early in the season and berries left onthe plant after the harvest season were picked to preventbuild up of H. hampei populations only on farm 3. Dur-ing this study, however, berries were not picked in thesampling plots on any of the farms. Fertilization was onlycarried out on farm 3.

On all three farms coffee (Coffea arabica L.) was grownunder shade. In the sampling plot on farm 1, the coffeevarieties were Caturra and Catimor and age of plantswere 9 and 6 years, respectively. Shade species were Ery-

thrina sp., Gliricidia sepium, Citrus sp., and Musa sp. Onfarm 2, the coffee varieties were Bourbon, Caturra andCatimor and age of plants were 26 years (Bourbon andCaturra) and 11 years (Catimor). Shade species were Ery-

thrina sp., Guazuma ulmifolia, Juglans olanchana andMusa sp. On these farms shade regulation was not per-formed. On farm 3 the coffee variety was Catuai andthe age of plants was 15 years. Shade species were Inga

sp., and Musa sp., and shade regulation was performedby pruning shade trees at the beginning of the rainy sea-son. Shade levels were estimated in 2005. Shade estima-tions were conducted at the four sites where H. hampei

was counted by using a spherical densitometer (Lemmon,1956). The shade level on each coffee field was obtainedby averaging the shade percentages measured at the foursites of the field.

The rainy season was from March to November in 2004and from May to December in 2005, and the total precip-itation was 2005 mm in 2004 and 2011 mm in 2005. Higherprecipitation levels were observed in June in both 2004 and2005 at 417 and 470 mm, respectively.

Table 1Mean percent Beauveria bassiana-infected Hypothenemus hampei (infec-tion level) per year and farm in 2004 and 2005

Farmsa Infection level (%)

2004 Mean ± SE 2005 Mean ± SE

Farm 1 2.3 ± 4.2a (n = 482) 13.7 ± 13.4a (n = 573)Farm 2 2.2 ± 3.2a (n = 510) 15.0 ± 12.5a (n = 456)Farm 3 3.7 ± 4.3a (n = 470) 7.6 ± 7.7a (n = 669)Meanb 2.7 ± 3.8a 12.1 ± 11.6b

a Differences between farms within a year: Different letters within acolumn denote significant differences using Tukey’s (a = 0.05) multiplecomparisons.

b Differences between years (Means): Different letters between columnsdenote significant differences using Tukey’s (a = 0.05) multiplecomparisons.

136 A.J. Monzon et al. / Journal of Invertebrate Pathology 97 (2008) 134–141

2.2. Collection of bored berries

To estimate the B. bassiana infection level in H. hampei

populations, 100 infested berries (based on characteristichole bored on the berry by the insect) were collected everysecond week in each field from July to November 2004 andfrom July to December in 2005. Green, semi-red and redberries were collected from at least 50 randomly selectedplants in each field on each date.

2.3. Diagnosis of B. bassiana infection

After collection, berries were immediately brought tothe laboratory of the National Agricultural University(Universidad Nacional Agraria, Managua, Nicaragua)for processing. Berries were surface sterilized prior to dis-section to avoid contaminants from outside the berry tocome in contact with insects inside the berry and to avoidcross contamination between berries. This was done bywashing berries with sterile water and then by dipping themin 75% ethanol for 2 min. After surface sterilization, berrieswere dissected in a laminar flow hood and H. hampei adultswere removed, rinsed with 75% ethanol by using a smallbrush and then placed individually in 30 ml vials with 2%water–agar. Insects were rinsed to avoid contaminant fromthe insects to grow on the water agar. Vials were then incu-bated in the dark at room temperature (ca. 24 �C). H. ham-

pei was examined every second day and mortality andfungal infection recorded. During the incubation period,no food was provided since H. hampei is able to surviveextended periods of time without food (Posada and Vega,2005). Fungi growing on the insect were isolated on potatodextrose agar (PDA) and B. bassiana was morphologicallyidentified according to De Hoog (1972), Brady (1979) andDomsch et al. (1980). Only dead H. hampei with visiblegrowth of B. bassiana were recorded as killed by thefungus.

2.4. H. hampei population estimate

Percentage of infested berries was used as a measure forH. hampei infestation level (Remond et al., 1996; Baker,1999; Wegbe et al., 2003). To obtain these numbers,infested and non-infested berries were counted at four sam-pling sites in each field every second week from July toNovember 2004 and from July to December in 2005. Eachsampling site consisted of eight plants with about 20 mbetween sites and at least 10 m from the edge of the field.One hundred berries from five branches were counted oneach plant at each sampling date. The counting started inthe upper part of the coffee tree and ended in the lowerpart. If less than 100 berries were present in the fivebranches, the counting continued on new branches until100 berries per tree were counted. To avoid interferingeffects on the population estimate, the counting was con-ducted in parts of the field where no collection of berrieswere conducted.

2.5. Statistical analysis

Percentages of B. bassiana infection level, and H. hampei

infestation level were arcsine square-root transformedbefore a two-way ANOVA (PROC GLM) test was usedto compare B. bassiana infection levels and H. hampeiinfestation levels between farms, sampling dates and years.Means were separated using Tukey’s multiple range tests.A correlation analysis (PROC CORR) was conducted totest density dependence between H. hampei and B. bassi-

ana. A correlation analysis was also conducted to studythe relation between B. bassiana prevalence, temperatureand relative humidity in 2005. The effect of precipitationon H. hampei and B. bassiana was also tested by using acorrelation analysis in 2004 and 2005. Analysis of variance(PROC GLM) and Tukey’s multiple range tests were alsoused to compare shade levels between farms. SAS version9.1 (SAS Institute Inc., 2003) was used for analysis.

3. Results

3.1. B. bassiana infection levels

The B. bassiana infection levels were generally higher in2005 than in 2004, and fluctuated widely throughout bothseasons. The mean infection level in 2005 was 12.1% andsignificantly higher (P = 0.0001) than the 2004 mean infec-tion level of 2.7%. No consistent significant differences ininfection levels were found between farms in any of theyears (Table 1).

3.1.1. B. bassiana infection levels in 2004

No significant differences in B. bassiana infection levelswere found between farms or between sampling dates in2004. The mean B. bassiana infection level on coffee farm1 was 2.3% (n = 482) (Table 1), and the highest infectionlevel was observed November 3 and was 11.5% (n = 61)(Fig. 1). On coffee farm 2, the mean B. bassiana infectionlevel was 2.2% (n = 510) (Table 1) and the highest infectionlevel was 7.8% (n = 51) observed October 20 (Fig. 1). The

Fig. 1. Percentages of berries infested by H. hampei (H. hampei infestationlevel) and percentages of H. hampei infected with B. bassiana (B. bassiana

infection level) in three coffee farms throughout the rainy season in 2004.

Fig. 2. Percentages of berries infested by H. hampei (H. hampei infestationlevel) and percentages of H. hampei infected with B. bassiana (B. bassiana

infection level) in three coffee farms throughout the rainy season in 2005.

A.J. Monzon et al. / Journal of Invertebrate Pathology 97 (2008) 134–141 137

mean B. bassiana infection level was 3.7% (n = 470) in cof-fee farm 3 (Table 1) and the highest infection level wasobserved September 8 and was 13.5% (n = 37) (Fig. 1).

3.1.2. B. bassiana infection levels in 2005No significant differences in B. bassiana infection levels

were found between farms in 2005 (Table 1). Significantdifferences (P = 0.0007) between sampling dates werefound when data from all three farms were pooled, andthe highest infection level was 27.0% on November 25(Fig. 2).

The mean B. bassiana infection level in H. hampei was13.7% (n = 573) on coffee farm 1 (Table 1), and the highestprevalence was observed November 25 and was 44.0%(n = 25) (Fig. 2). The mean B. bassiana infection level inH. hampei was 15.0% (n = 456) on coffee farm 2 (Table1), and the highest infection (37.5%; n = 16) level wasobserved September 30. Peaks of about 30% were also seenon August 18 and September 2 (Fig. 2). On coffee farm 3,the mean B. bassiana infection level was 7.6% (n = 669)(Table 1) and the highest infection level was observedNovember 25 and was 23% (n = 35). A peak of 20.0%was also observed in October 28 (Fig. 2).

3.2. H. hampei infestation levels

The H. hampei infestation level was significantly higherin 2004 than in 2005 (P < 0.0001), and 6.9% of the berrieswere infested in 2004 and 0.8% in 2005. Farm 2 had a sig-nificantly higher H. hampei infestation level than farms 1and 3 both in 2004 and 2005. No consistent significant dif-ferences in the H. hampei infestation level were foundbetween sampling dates on any of the farms in any of theyears.

3.2.1. H. hampei infestation levels in 2004

Table 2 shows that H. hampei infestation levels werehigh on all three farms, and a significant (P < 0.0001)higher H. hampei infestation level was found on farm 2than on farms 1 and 3 in 2004. The highest H. hampei infes-tation levels observed were 15.3% November 3 on farm 2,6.6% October 6 on farm 1 and 4.2% September 21 on farm3 (Fig. 1).

3.2.2. H. hampei infestation levels in 2005In 2005 H. hampei infestation levels were low on all

three farms, but also this year a significant (P = 0.0004)

Table 2Mean percentage of coffee berries infested by Hypothenemus hampei

(infestation level) per year and farm in 2004 and 2005

Farms a Infestation level (%)

2004 Mean ± SE 2005 Mean ± SE

Farm 1 5.0 ± 1.2b (n = 44,587) 0.3 ± 0.2b (n = 50,422)Farm 2 12.0 ± 2.3a (n = 36,313) 1.2 ± 0.7a (n = 47,163)Farm 3 3.5 ± 0.6b (n = 38,421) 0.6 ± 0.3b (n = 51,824)Meanb 6.9 ± 4.0a 0.8 ± 0.6b

a Differences between farms within a year: Different letters within acolumn denote significant differences using Tukey’s (a = 0.05) multiplecomparisons.

b Differences between years (Means): Different letters between columnsdenote significant differences using Tukey’s (a = 0.05) multiplecomparisons.

138 A.J. Monzon et al. / Journal of Invertebrate Pathology 97 (2008) 134–141

higher percentage of bored berries was found on farm 2compared to farm 1 and 3 (Table 2). The highest H. hampei

infestation levels observed were 2.2% July 22 and Decem-ber 8 on farm 2, 1.5% November 25 on farm 3 and 0.8%August 5 on farm 1 (Fig. 2).

3.3. B. bassiana�H. hampei density dependence

On farm 1, the tendency of increasing B. bassiana infec-tion levels and decreasing H. hampei infestation levels,resulted in a negative correlation between these two vari-ables in 2004 (r = �0.89; P = 0.001) and in 2005(r = �0.65; P = 0.02). On farm 2, B. bassiana infection lev-els and H. hampei infestation levels did not correlate in anyof the years (2004: r = 0.59, P = 0.09 2005: r = 0.23,P = 0.47). Results from farm 3, however, show that highB. bassiana infection levels co-occurred with high H. ham-

pei infestation levels, and a positive correlation (r = 0.66;P = 0.01) was found in 2005; in 2004, however, no correla-tion was found (r = 0.36; P = 0.34).

3.4. Shade, H. hampei infestation and B. bassiana infection

levels

On farm 1 and farm 2, mean shade levels were 73.7%and 71.4% respectively, while on farm 3 the shade levelwas 60.7%. Significant differences in shade level were foundbetween farm 2 and 3 (P = 0.04). No significant differenceswere, however, found in shading conditions between farm 1compared to farm 2 and 3. No significant correlationbetween shade levels and H. hampei infestation or betweenshade levels and B. bassiana infection levels were found.

3.5. H. hampei, B. bassiana infection levels and climatic

conditions

No correlation was found between B. bassiana infectionlevels and temperature on any of the farms, and no corre-lation was observed between relative humidity and B. bas-

siana on farms 1 and 2. On farm 3, however, a negative

correlation (r = �0.69; P = 0.01) between relative humidityand B. bassiana was observed.

No correlation between H. hampei infestation levels andprecipitation or between B. bassiana infection levels andprecipitation were found on any of the farms.

4. Discussion

This study demonstrates that the natural occurrence ofB. bassiana in a H. hampei population varies considerablythroughout a season and between years, and that infectionlevels up to 44% can be found in unsprayed fields.Although other studies on natural occurrence of B. bassi-

ana have been conducted earlier (Lacayo et al., 1994;Lacayo, 1995; Barrios et al., 1994; Feliz, 2003), this is thefirst study where natural infection levels of B. bassiana inH. hampei populations are studied systematically through-out two successive seasons.

The natural B. bassiana infection levels found in H. ham-

pei populations in our study are similar or higher thanreported from Nicaragua earlier (Guharay, 1994; Lacayoet al., 1994; Lacayo, 1995), higher than reported from Bra-zil (Alves et al., 2002) and lower than reported from Hon-duras (Trejo and Funez, 2004). Earlier results fromNicaragua show natural B. bassiana infection levels up to18% under humid climatic conditions (Lacayo et al.,1994). Another study carried out over a 3-year-period inNicaragua, based on percentages of infested berries witha white mycelium in the entrance hole, reported up to40% B. bassiana infection the first year, 10% the secondyear and 35% the third year (Guharay, 1994). In Honduras,percentages up to 80% infested berries with white myceliumin the entrance hole have been reported from areas withhigh humidity and temperature (Trejo and Funez, 2004).These berries were dissected and 100% of H. hampei adultsfound in the berries were killed by the fungus. Alves et al.(2002) reported B. bassiana infection levels from severalcoffee growing regions in Brazil, but mortalities of morethan 20% were not found. The B. bassiana infection levelsreported in the studies mentioned above are not directlycomparable due to the different methods used to assessinfection levels. In our study, only dead H. hampei withapparent B. bassiana growth was recorded as killed bythe fungus. Since limited growth and/or toxicosis causedby insect pathogenic fungi may be sufficient to kill theinsect (Butt et al., 1992) dead H. hampei without B. bassi-

ana growth might also have been killed by the fungus.Hence, the infection levels we report probably underesti-mate actual infection levels.

Though it is common to report B. bassiana infection lev-els in H. hampei, based on percentages of infested berrieswith a white mycelium in the entrance hole of the galleries,such estimations may not be accurate. In preliminary stud-ies, the authors have found that in some cases the whitefungal mycelium growing in the boring hole has not beenidentified to be B. bassiana. In other cases, the insect mightbe killed by the fungus, but due to environmental condi-

A.J. Monzon et al. / Journal of Invertebrate Pathology 97 (2008) 134–141 139

tions no fungal growth is seen in the boring hole (Bustillo2005).

In our study, high natural B. bassiana infection levelswere often found late in the rainy season (October andNovember). Guharay (1994) and Barrios et al. (1994) alsoreported that higher infection levels were found late in theseason. This is the part of the year when berries ripen (Bar-rios et al., 1991), and a high number of berries fall to theground. A high B. bassiana infection level at this time ofthe year therefore probably affects the mortality of H. ham-

pei in fallen berries. Damon (2000) reports that natural B.

bassiana epizootics in H. hampei populations may occur inberries on the ground under cool conditions, at high rela-tive humidity and under heavy shade. H. hampei popula-tions found late in the season do not complete the lifecycle before berries are harvested (Baker, 1999). It is in thispart of the season when H. hampei infestations often tendto decrease. There might be at least two factors contribut-ing to this: (1) harvesting of berries at this part of the yearalso removes berries infested with H. hampei from the field(Ferreira et al., 2000); and (2) natural B. bassiana infectionkills H. hampei.

Low B. bassiana infection levels were observed at thebeginning of the rainy season in our study. A low B. bassi-

ana control of H. hampei at this time of the year is a disad-vantage, since this is the part of the year when H. hampei

females colonize berries produced from secondary flower-ing (Guharay and Monterrey, 1997). When low B. bassiana

infection levels are found at this time of the year, it mightbe expedient to apply B. bassiana in order to increase thegeneral B. bassiana infection level in the field. Lacayo(1995) suggests that early B. bassiana applications whenH. hampei is emerging from berries after harvest or in ber-ries of secondary flowering can increase the initial infectionand promote early presence of the fungus. The recom-mended application of B. bassiana against H. hampei inNicaragua at present is in June, to control H. hampei inberries that produced early in the season and in September,to control H. hampei in berries produced from the mainflowering (Guharay and Monterrey, 1997). Applying B.

bassiana does not necessarily result in high and stable B.

bassiana infection levels. In one study, where B. bassiana

was applied inundatively to the soil to control H. hampei

in fallen berries, infection levels ranging from 7.2% to29.3% were reported (Bustillo et al., 1999). Bustillo andPosada (1996) also reported that results of B. bassianaapplication against H. hampei are inconsistent and infec-tion levels fluctuate from low (20–30%) to high (60–70%).They also report that in areas where the fungus had beenpreviously applied, it had become established and naturalmean infection levels up to 45.0% were found. Other resultsshow B. bassiana infections up to 40.6% (De La Rosa et al.,2000). After three and six B. bassiana applications, infec-tion levels up to 48.1% and 69.0% after 60 and 119 daysof application respectively were observed by Bustillo andPosada (1996). In studies conducted by Barrios et al.(1994), B. bassiana infection levels of 65%, 37% and 7%

were obtained by applying B. bassiana in September, Octo-ber and November, respectively. Lacayo (1995) foundinfection level up to 40%, 79 days after application, butshe only found significant differences from the control inpercentages of bean damage at harvest time. Guharayet al. (2000) suggested that the effect of B. bassiana onthe H. hampei population is not observed in the percentageof bored berries, but in the yield and quality of coffee atharvest time. The B. bassiana infection levels after it wasapplied in the field (discussed above) are similar or higherthan the natural occurring B. bassiana infection level foundin our study.

The H. hampei infestation level in our study varied con-siderably between years, and between farms; though no sig-nificant differences were found between sampling dates,variations were also observed within seasons. The low H.

hampei infestation in 2005 compared to 2004, illustratesthe cyclic behavior of H. hampei populations, which mightbe associated with the biennial cyclic production of berrieson the plant, where years of high berry production alter-nate with years of low berry production (Chamorroet al., 1995; Kufa et al., 2001). The H. hampei infestationlevel in 2005 in our study did not reach 2%, which is con-sidered the economic damage threshold for H. hampei intraditional coffee production systems (Borbon, 2003). Thethreshold for intensively produced coffee is as high as 5%(Baker, 1999). The only case where H. hampei infestationpassed the 2% threshold level in 2005 was on farm 1 where2.2% of the berries were infested in July and December. InDecember, the production season was in the final phaseand control measures are normally not implemented. In2004, however, the H. hampei infestation level was abovethe 2% threshold level throughout the season on all threefarms.

Consistent host density dependence between the H. ham-

pei infestation level and the B. bassiana infection level wasnot found in our study. These results might suggest that theH. hampei–B. bassiana system is not highly density depen-dent and that factors other than host density are moreimportant to the B. bassiana infection level. This is sup-ported by Bustillo (2005) who found no significant differ-ences in B. bassiana infection level after applying B.

bassiana in fields with different H. hampei infestation levels.De La Rosa et al. (2000) however, found a positive corre-lation between H. hampei infestation levels and B. bassiana

in infested berries. Studies on other insects also report ofnegative correlations between insect density and fungalentomopathogen-caused mortality. Ekbom and Pickering(1990) found a negative correlation between host densitiesof black margin aphid (Monellia caryella) and fungal infec-tion levels by Entomophthora planchoniana and Neozygites

sp. Further, higher prevalence of B. bassiana in Sitona dis-

coideus larvae have been observed in soils with low pestdensity compared to soils with high densities (Aeschlimannet al., 1985). It is generally accepted, however, that highhost densities favor high fungal infection levels in manyhost–pathogen interactions, and it has been suggested that

140 A.J. Monzon et al. / Journal of Invertebrate Pathology 97 (2008) 134–141

a certain host density threshold needs to be exceeded forthe initiation of an epizootic to occur (Steenberg, 1995;Inglis et al., 2001).

Host density is only one factor affecting the fungal infec-tion level and an epizootic development of the fungus in ahost insect population. Other biotic and abiotic factors arealso very important for the epizootic development (Carru-thers and Soper, 1987; Fuxa and Tanada, 1987; Steenberg,1995; Inglis et al., 2001; Klingen and Haukeland, 2006). Incoffee systems, factors such as pesticide application andshade management can affect B. bassiana significantly. Itis known that direct solar radiation reduces germinationof conidia and survival of the insect pathogenic fungusB. bassiana (Inglis et al., 2001; Staver et al., 2001). In addi-tion, shade favors the viability of conidia by providing highhumidity in the field (Inglis et al., 2001; Staver et al., 2001;De Oliveira et al., 2003; Kouassi et al., 2003) and highhumidity is required for sporulation and spore germinationof B. bassiana (Walstad et al., 1970). The differences in H.

hampei infestation levels and B. bassiana infection levelsobserved between farms might be caused by different coffeemanagement practices, such as varieties and shade manage-ment. However, the study was not set up to examine thisaspect and no correlation between percentage of infestedberries and shade levels were found at the narrow shadelevels range measured in the field.

This study suggests that naturally occurring B. bassiana

is prevalent in H. hampei populations where no B. bassiana

has ever been applied. The natural infection level can bequite high, and is probably an important H. hampei mortal-ity factor. The B. bassiana infection level varies consider-ably throughout the season and between years.Knowledge about this variation should be useful todevelop an effective and cost effective practical applicationof B. bassiana in microbial control strategies againstH. hampei.

Acknowledgments

We thank Dr. Fernando E. Vega, Professor TrondHofsvang and Dr. Richard Meadow who peer reviewedthis paper. Thanks to coffee farmers, Victoriano Arauz,Agustın Arauz and Isidoro Herrera on whose farms thisstudy was conducted and Johana Castillo, Lidia Torrez,Patricia Acuna, Wilver Betanco and Vıctor Monzon fortechnical assistance. This study was financially supportedby the NORAD funded CATIE/MIP AF project, NationalAgricultural University (UNA), Norwegian Institute forAgricultural and Environmental Research (BIOFORSK),Tropical Agricultural Research and Higher EducationCentre (CATIE) and the Norwegian University of LifeSciences (UMB).

References

Aeschlimann, J.P., Ferron, P., Marchal, M., Soares, G., 1985. Ocurrenceand pathogenicity of Beauveria bassiana infesting larval Sitona

discoideus (Col.: Curculionidae) in the Mediterranean region. Ento-mopahaga 30, 73–82.

Alves, S.B., Pereira, R.M., Lopes, R.B., Tamai, M.A., 2002. Use ofentomopathogenic fungi in Latin America. In: Upadhyay, R.K. (Ed.),Advances of Microbial Control of Insect Pests. Kluwer Academic/Plenum Publishers, New York, pp. 193–212.

Baker, P., 1999. The coffee berry borer in Colombia; Final Report of theDFID-Cenicafe-CABI Bioscience IPM for coffee project (CNTR 93/1536 A). Chinchina Colombia, DFID-CENICAFE. 154 pp.

Barrios, M., Castillo, L.A., Sequeira, C. A., Ubeda, H.R., 1991.Distribucion de la broca del fruto del cafeto Hypothenemus hampei

Ferr (Coleoptera, Scolytidae) en la Region VI de Nicaragua. Informetecnico del Departamento de Proteccion Vegetal, ciclo 1989–1990. In:Ministerio de Agricultura y Ganaderıa, Managua (Nicaragua) CentroNacional de Investigacion del Cafe, Matagalpa, pp. 47–53.

Barrios, M., Jimenez, C., Guharay, F., 1994. Ecologıa de la interaccion deBeauveria bassiana contra la broca del cafe. In: V Congreso Internac-ional de Manejo Integrado de Plagas, Costa Rica, pp. 54.

Borbon, M.O., Fonseca, C.C., Rojas, H., 2001. Broca del cafeto, unanueva plaga para nuestra caficultura. Boletın Tecnico ICAFE 1, 4–5.

Borbon, M.O., 2003. Metodo de liberacion inundativa para el control dela broca del fruto del cafeto con parasitoides Phymastichus coffea,

Cephalonomia stephanoderis y Prorops nasuta. Boletın informativoICAFE 3, 9–11.

Brady, B.L., 1979. Beauveria bassiana. Commonwealth MycologicalInstitute (CMI). Descriptions of Pathogenic Fungi and Bacteria. No.602. 2 pp.

Brun, L.O., Marcillaud, C., Gaudichon, V., Suckling, D.M., 1989.Endosulfan resistance in Hypothenemus hampei (Coleoptera: Scolyti-dae) in New Caledonia. J. Econ. Entomol. 82, 1311–1316.

Bustillo, A.E., Bernal, M.G., Benavides, P., Chaves, B., 1999. Dynamicsof Beauveria bassiana and Metarhizium anisopliae infecting Hypoth-

enemus hampei (Coleoptera: Scolytidae) populations emerging fromfallen coffee berries. Florida Entomol. 82, 491–498.

Bustillo, A., Posada, F., 1996. El uso de entomopatogenos en el control dela broca del cafe en Colombia. Manejo Integral de Plagas 42, 1–13.

Bustillo, A.E., Cardenas, R., Posada, F., 2002. Natural enemies andcompetitors of H. hampei (Ferrari) (Coleoptera: Scolytidae) inColombia. Neotrop. Entomol. 31, 635–639.

Bustillo, A.E., 2005. El papel del control biologico en el manejo integradode la broca del cafe, Hypothenemus hampei (Ferrari) (Coleoptera:Curculionidae: Scolytinae). Rev. Acad. Colomb. Cienc. 29, 55–68.

Butt, T.M., Barrisever, M., Drummond, J., Schuler, T.H., Tillemans, F.T.,Wilding, N., 1992. Pathogenicity of the entomogenous, hyphomycetefungus Metarhizium anisopliae against the chrysomelid bettles Psylli-

odes chrysocephala and Phaedon cochleariae. Biocontrol Sci. Technol.2, 327–334.

Canet, B.G., Garcia, A., 2005. PROMECAFE en marcha, acciones dePROMECAFE en Panama. Boletın PROMECAFE-IICA 104, 4–5.

Carruthers, R.I., Soper, R.S., 1987. Fungal diseases. In: Fuxa, J.R.,Tanada, Y. (Eds.), Epizootiology of Insect Diseases. John Wiley &Sons Inc., New York, pp. 357–416.

Chamorro, G., Cardenas, R., Herrera, A., 1995. Evaluacion economica yde la calidad en taza del cafe proveniente de diferentes sistemas derecoleccion manual, utilizables como control en cafetales infestados deHypothenemus hampei. Cenicafe 46, 164–175.

Comite Nacional de Manejo Integrado de Plagas (CN-MIP), ProgramaCATIE/MIP-AF, Red de Accion en Plaguicidas y sus Alternativas enAmerica Central (RAPAC), 2002. Alternativas comprobadas parasustituir plaguicidas de la nueva Docena Sucia. Managua, Nicaragua,48 pp.

Cure, J.R., Santos, R.H.S., Moraes, J.C., 1998. Ecologia, comportamentoe bionomia fenologia e dinamica populacional da broca do cafeHypothenemus hampei (Ferr.) relacionadas as fases de desenvolvimentodo fruto. An. Soc. Entomol. Brasil 27, 325–335.

Damon, A., 2000. A review of the biology and control of the coffee berryborer, Hypothenemus hampei (Coleoptera: Scolytidae). Bull. Entomol.Res. 90, 453–465.

A.J. Monzon et al. / Journal of Invertebrate Pathology 97 (2008) 134–141 141

De Hoog, G.S., 1972. The genera Beauveria, Isaria, Tritirachium andAcrodontium gen. nov. Stud. Mycol. 1, 1–41.

De La Rosa, W., Alatorre, R., Trujillo, J., Barrera, J.F., 1997. Virulence ofBeauveria bassiana (Deuteromycetes) strains against the coffeeberry borer (Coleoptera: Scolytidae). J. Econ. Entomol. 90,1531–1538.

De La Rosa, W., Alatorre, R., Barrera, J.F., Toriello, C., 2000. Effect ofBeauveria bassiana and Metarhizium anisopliae (Deuteromycetes) uponthe coffee berry borer (Coleoptera: Scolytidae) under field conditions.Econ. Entomol. 93, 1409–1414.

De Oliveira, C.N., Oliveira, J.P.M.N., Sueki, K.L., 2003. Compatibilitybetween the entomopathogenic fungus Beauveria bassiana and insec-ticides used in coffee plantations. Sci. Agric. 60, 663–667.

Dufour, B., Barrera, J., Decazy, B., 1999. La broca de los frutos del cafeto:La lucha biologica como solucion. In: Bertrand, B., Rapidel, B. (Eds.),Desafios de la Caficultura en Centroamerica. IICA-PROMECAFECIRAD- IRD-CCCR. San Jose, Costa Rica. pp. 293–326.

Domsch, K.H., Gams, W., Anderson, T.H., 1980Compendium of soilfungi, Vol. 1. Academic Press, London, UK, pp. 137-139.

Ekbom, B., Pickering, J., 1990. Pathogenic fungal dynamics in a fallpopulation of the blackmargined aphid (Monellia caryella). Entomol.Exp. Appl. 57, 29–37.

Feliz, D.A., 2003. Incidence of the borer Hypothenemus hampei Ferr. 1867and their natural controllers in coffee plants under different shade typesin San Marcos, Nicaragua. MSc. thesis, CATIE, Turrialba, CostaRica.

Ferreira, A.J., Bueno, H.P.V., Moraes, J.C., Carvalho, G.A., BuenoFilho, J.S. de S., 2000. Ecologia, comportamiento e bionomia.Dinamica populacional da broca-do-cafe Hypothenemus hampei (Ferr.)(Coleoptera: Scolytidae) em Lavras, MG. An. Soc. Entomol. Brasil 29,237–244.

Fuxa, J.R., Tanada, Y., 1987. Epidemiological concepts applied to insectepizootiology. In: Fuxa, J.R., Tanada, Y. (Eds.), Epizootiology ofInsect Diseases. John Wiley & Sons Inc., New York, pp. 3–21.

Guharay, F., Monterrey, J., 1997. Manejo Ecologico de la broca del cafeto(Hypothenemus hampei) en America Central. CATIE, Hoja TecnicaMIP No. 22, 8 pp.

Guharay, F., 1994. Perspectivas sobre el uso de hongos entomopatogenospara manejo de plagas en Nicaragua. In: CATIE-INTA/MIP (Ed.),Informe Final Proyecto Hongos Entomopatogenos. Managua Nica-ragua, pp. 48–52.

Guharay, F., Monterrey, J., Monterroso, D., Staver, Ch., 2000. Manejointegrado de plagas en el cultivo de cafe, CATIE, Managua. 272 p.

Haraprasad, N., Niranjana, S.R., Prakash, H.S., Shetty, H.S., Wahab, S.,2000. Beauveria bassiana a potencial mycopesticide for the efficientcontrol of coffee berry borer Hypothenemus hampei (Ferrari) in India.Biocontrol Sci. Technol. 11, 251–260.

Inglis, G.D., Goettel, M.S., Butt, T.M., Strasser, H., 2001. Use ofhyphomycetous fungi for managing insect pests. In: Butt, T.M.,Jackson, C.W., Magan, N. (Eds.), Fungi as Biocontrol Agents,Progress, Problems and Potentials. CABI Publishing, Wallinford,U.K, pp. 23–69.

Kimani, M., Little, T., Vos, J.G.M., 2002. Introduction to coffeemanagement through discovery learning. Farmer Participatory Train-ing and Research Program. CABI Bioscience Kenya. 35 pp.

Klingen, I., Haukeland, S., 2006. The soil as a reservoir for naturalenemies of pest insects and mites with emphasis on fungi andnematodes. In: Eilenberg, J., Hokkanen, H.M.T. (Eds.), An Ecologicaland Societal Approach to Biological Control. Springer, Dordrecht,The Netherlands, pp. 145–211.

Kouassi, M., Coderre, B., Todorova, S.I., 2003. Effects of the timing ofapplications on the incompatibility of three fungicides and one isolateof the entomopathogenic fungus Beauveria bassiana (Balsamo) Vuille-min (Deuteromycotina). J. Appl. Ent. 127, 421–426.

Kufa, T., Shimber, T., Yilma, A., Netsere, A., Taye, E., 2001. The impactof close spacing on yield of arabica coffee under contrasting agro-ecologies of Ethiopia. Afr. Crop Sci. J. 9, 401–409.

Lacayo, L., Barrios, M., Jimenez, C., Sandino, V., 1994. Uso de hongosentomopatogenos para el control de la broca del cafeto (Hypothenemus

hampei) en Nicaragua. In: CATIE-INTA/MIP (Ed.), Informe FinalProyecto Hongos Entomopatogenos. Managua Nicaragua, pp. 48–52.

Lacayo, L., 1995. Use of Beauveria bassiana(Bals.) Vuill. for control ofHypothenemus hampei Ferr. and Plutella xylostella (L.) in Nicaragua.PhD. thesis, Department of Biology, Imperial College, London, UK.

Lemmon, P.E., 1956. A spherical densitometer for estimating forestoverstory density. Forest Sci. 2, 314–320.

Le Pelley, R.H., 1968. Pests of Coffee Tropical Science Series. Longmans,London, UK.

Mathieu, F., Gaudichon, V., Brun, L.O., Frerot, B., 2001. Effect ofphysiological status on olfactory and visual responses of femaleHypothenemus hampei during host plant colonization. Physiol. Ento-mol. 26, 189–193.

Posada, F., Vega, F., 2005. A new method to evaluate the biocontrolpotential of single spore isolates of fungal entomopathogens. J. InsectSci. 5 (37), 10.

Rehner, S.A., Posada, F., Buckley, E.P., Infante, F., Castillo, A., Vega,F.E., 2006. Phylogenetics origin of African and Neotropical Beauveria

bassiana s.l. pathogens of the coffee berry borer Hypothenemus hampei.J. Invertebr. Pathol. 93, 11–21.

Remond, F., Cilas, C., Vega-Rosales, M.I., Gonzalez, M.O., 1996.Methodologie dechantillonnage pour estimer les attaques des baiesdu cafeier par les scolytes (Hypothenemus hampei Ferr.). Cafe, cacao,the 37, 35–52.

Rhodes, L.F., Mansingh, A., 1981. Susceptibility of the coffee berry borerHypothenemus hampei to various insecticidal formulations. Insect. Sci.Appl. 2, 227–231.

Samuels, R.I., Pereira, R.C., Gava, C.A.T., 2002. Infection of the coffeeberry borer Hypothenemus hampei (Coleoptera: Scolytidae) by Brazil-ian isolates of the entomopathogenic fungi Beauveria bassiana andMetarhizium anisopliae (Deuteromycotina: Hyphomycetes). BiocontrolSci. Technol. 12, 631–635.

SAS Institute Inc., 2003. SAS/STAT User’s Guide. Version 9.1. SASInstitute Inc. Cary, North Carolina, USA.

Soto-Pinto, L., Perfecto, I., Caballero-Nieto, J., 2002. Shade over coffee:its effects on berry borer, leaf rust and spontaneous herbs in Chiapas,Mexico. Agroforest. Syst. 55, 37–45.

Staver, Ch., Guharay, F., Monterroso, D., Muschler, R.G., 2001.Designing pest-suppressive multistrata perennial crop systems: shade-grown coffee in Central America. Agroforest. Syst. 53, 151–170.

Steenberg, T., 1995. Natural occurrence of Beauveria bassiana (Bals.)Vuill. with focus on infectivity to Sitona species and other insects inLucerne. Ph.D. thesis, Royal Veterinary and Agricultural University,Copenhagen, Denmark.

Trejo, S.A.R., Funez, C.R., 2004. Manejo Integrado de la broca de cafe,basado en criterios bioecologicos de la broca y del cultivo de cafe—Manual Tecnico. Instituto Hondureno del Cafe (IHCAFE), 44 pp.

Van Lenteren, J.C., Bueno, V.H.P., 2003. Augmentative biological controlof arthropods in Latin America. BioControl 48, 123–139.

Walstad, J.D., Anderson, R.F., Stambaugh, W.J., 1970. Effects ofenvironmental conditions on two species of muscardine fungi (Beau-

veria bassiana and Metarhizium anisopliae). J. Invertebr. Pathol. 16,221–226.

Wegbe, K., Cilas, Ch., Decazy, B., Alauzet, C., Dufour, B., 2003.Estimation of production losses caused by the coffee berry borer(Coleoptera: Scolytidae) and calculation of an economic damagethreshold in Togolese coffee plots. J. Econ. Entomol. 96, 1473–1478.

Wrigley, G., 1988. Coffee. Longman Scientific & Technical, New York,USA.