Impact of natural enemies on coffee leafminer Leucoptera coffeella (Lepidoptera: Lyonetiidae)...

Author's personal copy

Impact of natural enemies on coffee leafminer Leucoptera coffeella(Lepidoptera: Lyonetiidae) population dynamics in Chiapas, Mexico

J. Refugio Lomeli-Flores a,1, Juan F. Barrera b, Julio S. Bernal a,*

a Department of Entomology, Biological Control Laboratory, Texas A&M University, MS 2475, College Station, TX 77843-2475, USAb El Colegio de la Frontera Sur, Carretera Antiguo Aeropuerto, Km 2.5 Tapachula, Chiapas 30700, Mexico

a r t i c l e i n f o

Article history:Received 11 September 2008Accepted 25 March 2009Available online 13 June 2009

Keywords:Biological controlLife tablesFormicidaeEulophidaeBraconidaePredationParasitismCoffea arabica

a b s t r a c t

Coffee leafminer, Leucoptera coffeella, is a pest in many New World coffee growing areas. Previous studiessuggested that its population dynamics were strongly affected by natural enemies, particularly of larvae,and physical environmental conditions. Our study documented through field surveys and life table anal-yses (i) the natural enemy complex associated with coffee leafminer and (ii) the impacts of natural ene-mies on the population dynamics of coffee leafminer, on coffee (Coffea arabica) at two elevations and tworainfall levels in the Soconusco region of Chiapas, Mexico. Twenty-two larval parasitoid species (includ-ing 14 morphospecies) were collected. Egg and pupal parasitoids were not recovered. Life table analysesshowed that parasitism contributed 610% of real mortality, and parasitism rates were 8–10-fold higher atthe low (<550 m) versus high (>950 m) elevation; parasitism rates were similar under low (<100 mm)and high (>400 mm) rainfall. Seventeen predator species (including five morphospecies) were collected,of which most were ants (Formicidae, 14 species) that contributed >58% of real mortality. Life table anal-yses showed that predation rates were higher at high versus low elevation and under high versus lowrainfall. Independently of elevation and rainfall, egg predation (likely by ants) was the most importantsource of indispensable mortality (range = 0.13–0.30), except at low elevation and high rainfall wherepupal predation (=0.14) was similarly important. Also, predation was the main source of coffee leafminerlarval and pupal mortality during a 13-month period in a low elevation coffee farm and was highest dur-ing the rainy season (>400 mm rainfall/month), when coffee leafminer prevalence was highest. Overall,predation of eggs and pupae (the latter particularly at low elevation), mostly by ants, were the mostimportant sources of coffee leafminer mortality. Because ants were the main source of coffee leafmineregg and pupal mortality, their importance and potential role in coffee pest management strategies werediscussed.

� 2009 Elsevier Inc. All rights reserved.

1. Introduction

Coffee leafminer, Leucoptera coffeella (Guérin-Ménèville) (Lepi-doptera: Lyonetiidae), is a nocturnal microlepidopteran whose lar-vae are known only from Coffea species (Mendoza, 1995; Ramiroet al., 2004). Previous studies addressing the effects of coffee plantcharacteristics on coffee leafminer distribution and abundanceshowed that coffee leafminer prevalence was positively correlatedwith leaf age (= position on a branch) and nitrogen content (Lome-li-Flores et al., unpubl. data), and oviposition with leaf caffeine con-centrations (Magalha~es et al., 2008). In addition, laboratorystudies showed that coffee leafminer larvae grew larger, developedfaster, and survived better on leaves with moderate nitrogen levelscompared to leaves with low or high nitrogen levels, and on tough

versus soft leaves (Walker and Quintana, 1969; Nantes and Parra,1977; Guerreiro-Filho, 2006; Lomeli-Flores et al., unpubl. data),though adult females did not generally select such leaves for ovi-position in laboratory experiments (Lomeli-Flores et al., unpubl.data). Under field conditions, natural enemies and weather vari-ables may have stronger influences than coffee plant characteris-tics on the distribution and abundance of coffee leafminer(Heisswolf et al., 2005). Indeed, prior studies assessing the impactsof different mortality factors suggested that coffee leafminer pop-ulation dynamics were strongly affected both by natural enemies(Flórez and Hernández, 1981; Gravena, 1983; Campos et al.,1989; Paliz and Mendoza, 1993; Pereira et al., 2007; de la Moraet al., 2008), and environmental conditions prevalent at differentelevations (e.g., Souza et al., 2003; Lomeli-Flores et al., unpubl.data).

Previous studies of coffee leafminer documented a diverse suiteof natural enemies, including at least 26 genera and 30 species oflarval parasitoids, all Eulophidae or Braconidae (Hymenoptera),though no egg or pupal parasitoids were documented (Parra

1049-9644/$ - see front matter � 2009 Elsevier Inc. All rights reserved.doi:10.1016/j.biocontrol.2009.03.021

* Corresponding author. Fax: +1 979 845 5926.E-mail addresses: [email protected], [email protected] (J.S. Bernal).

1 Permanent address: Instituto de Fitosanidad, Colegio de Postgraduados, Km 35.5carretera México-Texcoco, Montecillo, Edo. de México 56230, Mexico.

Biological Control 51 (2009) 51–60

Contents lists available at ScienceDirect

Biological Control

journal homepage: www.elsevier .com/locate /ybcon


et al., 1977; Souza, 1979; Villacorta, 1980; Aranda-Delgado, 1986;Gallardo, 1988; Flórez and Hernández, 1981; Mendoza, 1995;Lomeli-Flores et al., unpubl. data). A compilation of predator spe-cies from previous studies includes at least six genera and 11 spe-cies of Vespidae (Hymenoptera), one morphospecies (anunidentified species clearly distinguishable from others by its mor-phology) of Formicidae (Hymenoptera), and two of Chrysopidae(Neuroptera), all predatory on larvae; predators of pupae or eggshave not been reported (Enriquez et al., 1975; Parra et al., 1977;Tozatti and Gravena, 1988; Mendoza, 1995; Fragoso et al., 2002;Carvalho et al., 2002, 2005; Fernández and Bueno, 2002). However,most studies overtly ignored mortality of coffee leafminer eggs andpupae, focusing exclusively on larvae (e.g., Parra et al., 1977; Villa-corta, 1980; Aranda-Delgado, 1986; Reis et al., 2000), though onestudy and observations preliminary to our study showed that eggsand pupae suffered high mortality rates due to predation (Pereiraet al., 2007; Lomeli-Flores et al., unpubl. data).

Coffee leafminer appears to be a secondary pest of coffee inMexico because it is of concern only in occasional years or on spe-cific farms, but few studies have assessed the impact of naturalenemies on its population dynamics. Previously, it was noted inBrazil that the intensity and frequency of coffee leafminer out-breaks increased dramatically with the introduction of improvedcoffee varieties, which required greater agrochemical use andfull-sun conditions (Fragoso et al., 2002), and that such outbreakswere due, in part, to faster coffee leafminer development and lowermortality due to natural enemies (Guharay et al., 2001). Aranda-Delgado (1986) conducted the single known survey of coffee leaf-miner parasitoids in Mexico, but primarily focused on larval para-sitoids, and largely in Veracruz state, which is located on the Gulfof Mexico coast. In prior studies in the Soconusco area of Chiapasstate, located on Mexico’s Pacific coast, coffee leafminer occurredat higher densities in low (<500 m) versus high (>1000 m) eleva-tion farms (Lomeli-Flores et al., unpubl. data), but it was unclearwhether the difference was due to natural enemies or other envi-ronmental variables associated with elevation. Thus, for Mexicofurther research on the impacts of natural enemies on coffee leaf-miner population dynamics is needed, and targeting all leafminerdevelopmental stages at different elevations and under varyingweather conditions. Data derived from such research are necessaryto improve evolving coffee IPM strategies (e.g., Barrera et al., 2000;Segura et al., 2004). Thus, our study sought to: (i) identify the par-asitoid and predator complexes associated with coffee leafminerand assess their impacts on coffee leafminer population dynamics,and (ii) contrast the impacts of natural enemies on coffee leafminerpopulation dynamics at two elevations and two rainfall levels, rep-resenting contrasting environmental conditions in the Soconuscoregion of Chiapas, Mexico.

2. Materials and methods

2.1. Study area

Studies were conducted on three arabica-coffee (Coffea arabicaL.) farms representing two elevations in the municipality of Cac-ahoatán, Chiapas, Mexico: El Encanto (14� 590 2100 N; 92� 090 5500

W; 480 m) and La Gloria (15� 000 0700 N; 92� 090 2400 W, 550 m),representing low elevation farms, and Alpujarras (15� 040 2400 N;92� 100 1100 W, 960 m) representing high elevation farms. Allthree farms were traditionally managed. That is, the farms wereplanted with tall coffee varieties and a high density of shadetrees (e.g., Inga spp.), had a minimal use of agrochemical inputs,and the fruits were hand-harvested as they matured. Shade coverwas 38–87% in the low elevation farms and 45–91% in the highelevation farm. Rainfall amounts and patterns were similar at the

low and high elevation farms: yearly rainfall was >4000 mm,with a December–March dry season (rainfall < 100 mm/month),and an April–November rainy season (>200 mm/month) (Comi-sión Nacional del Agua, 2006). Yearly average temperature was26.2 ± 0.2 �C at low elevation and 22.6 ± 0.3 �C at high elevation.The main coffee varieties on these farms were Borbon, Typica,and Catuai, but all studies involved Catuai plants. Insecticideswere not used in any of the farms during the study period. Allleaf samples to recover coffee leafminer or natural enemies werecollected randomly (height, cardinal direction) from coffeeplants.

2.2. Natural enemy complex associated with coffee leafminer

Two hundred leaves with active coffee leafminer mines, and100 coffee leafminer pupae were collected monthly at one low ele-vation (El Encanto) and the high elevation farm (Alpujarras) fromJune 2003 to July 2004 to document the parasitoid communityassociated with coffee leafminer. In addition, sporadic collectionsof leaves were made during the summers (May–July) of 2003–2005 in these and other coffee farms in the Soconusco region to re-cover as many parasitoid species as possible. All leaves were takento the laboratory and examined under a microscope at 25�magni-fication to confirm that mines were active or contained coffee leaf-miner pupae. All coffee leafminer eggs, larvae, and pupaerecovered from leaves were incubated individually (on or withina portion of the corresponding leaf) in 5 cm diam petri dishes untiladult parasitoids or adult leafminers emerged. Adult parasitoidswere killed in 80% ethanol upon emergence and properly mountedfor subsequent identification. Parasitoid specimens were identifiedto genus using the keys of Gibson et al. (1997), Burks (2003) andWharton et al. (1997). Parasitoid species were identified using spe-cific keys: Hansson (1987, 1997) for Chrysocharis; Hansson (1994)for Closterocerus; Hansson (1997) for Neochrysocharis; Yoshimoto(1983) for Pnigalio; Gordh (1978) for Zagrammosoma; and van Ach-terberg (1995) for Stiropius.

A preliminary survey of potential predators of coffee leafmin-er was conducted in mid-July 2005. Thirty coffee plants wererandomly selected in a low elevation farm (La Gloria), and allpredators found on each plant during a 10-min per plant obser-vation period were collected, including ants living inside coffeetwigs or branches. Also, any arthropod observed preying on cof-fee leafminers during the parasitoid surveys described above wascollected, though few collections were made in this manner dueto a likely short interaction time between predators and coffeeleafminer prey. Collected predators were killed in 80% ethanoland properly mounted for subsequent identification. The speci-mens thus collected, almost exclusively ants (see Section 3),were identified to genus using the keys of Mackay and Mackay(2004).

Predators were collected also by exposing ‘‘sentinel” coffeeleafminer eggs, larvae, and pupae on coffee plants to collect pre-dators during the last week of July 2005 at a low elevation farm(La Gloria). Sentinel coffee leafminers were reared at El Colegiode la Frontera Sur (hereafter ECOSUR), Tapachula, Chiapas, Mex-ico, on detached coffee leaves, following Reis et al. (2000), withminor modifications. Sentinels were exposed on each of 10 coffeeplants by stapling three coffee leaves with mature mines (minedarea >4 cm2), three with coffee leafminer eggs (<24-h old) andthree with pupae (<24-h old) to leaves of each plant at �1.5 mabove the ground. Altogether, at least 30 eggs, five larvae, andfive pupae were thus exposed to natural enemies per each ofthe 10 plants. Plants with sentinels were in a single, randomlyselected row, and separated from each other by at least five cof-fee plants. Diurnal (0900–1300 h) and nocturnal (2000–000 h)point observations were made to document predation events

52 J.R. Lomeli-Flores et al. / Biological Control 51 (2009) 51–60


on coffee leafminer sentinels. Each sentinel was observed for aperiod of 5 s every 5 min in a rotating fashion. Thus, a total of384 point observations were made during a total observationalperiod of 32 h, i.e. 16 h during the day and 16 h during the night.Predators feeding on sentinels were identified to genus in thefield and collected for subsequent identification, and the frequen-cies of predation events were scored. A predation event was re-corded when a predator was observed consuming/removing asentinel egg, larva, or pupa from a coffee leaf. Leaves with senti-nels were replaced as predation occurred.

Voucher specimens for all parasitoid and predator species ormorphospecies were deposited in the Texas A&M University in-sect collection (voucher specimens #666). A duplicate set ofvoucher specimens were deposited in the ECOSUR insect col-lection (without voucher number, but identified as ‘‘JRLF-voucher”).

2.3. Effects of natural enemies on coffee leafminer populationdynamics

The seasonal dynamics of coffee leafminer were assessedthrough monthly samples at a low elevation farm (El Encanto)from June 2003 to July 2004. Ten coffee plants were randomlyselected each month and the proportion of mined leaves perplant was assessed from nine randomly selected branches perplant. Also, 180 mined leaves and 67–157 coffee leafminer pupaewere collected per month, though pupae were not collected be-tween October 2003 and February 2004 due to logistical limita-tions. Leaves were held in plastic bags, and subsequentlyexamined to assess coffee leafminer survivorship and identifymortality sources. The numbers of mines, and coffee leafminereggs and larvae on each leaf were recorded. Coffee leafminer lar-val and pupal survival rates and mortality factors were assessedin the laboratory through dissection under a microscope at 25�magnification. Egg survival rate could not be assessed via fieldsamples because most of the mortality at this stage is due topredation (see Section 3), and predators did not leave visibleegg remains as evidence of predation (Lomeli-Flores, pers.observ.).

Previous studies (Aranda-Delgado, 1986; Lomeli-Flores, pers.observ.) showed that larval and pupal coffee leafminer mortalitycould be assigned unambiguously to one of three factors. (1) Deathby parasitoid: indicated by the presence of immature parasitoids(larvae, pupae, or their remains) inside coffee leafminer mines orpupae, parasitoid exit holes, or by presence of black, ovipositor-caused scars on coffee leafminer larvae. (2) Death by predator:indicated by torn mines in which the number of larvae withinthe mine was smaller than the number of coffee leafminer egg-shells at the base of the mine, or by torn or empty coffee leafminerpupal remains in the absence of evidence of parasitoid activity oremergence of an adult coffee leafminer. (3) Death by factors otherthan parasitoids or predators (hereafter FOPP): coffee leafminerlarval or pupal mortality was assigned to FOPP when mortalitycould not be unambiguously assigned to either death by parasitoidor by predator.

Monthly per-leaf densities of coffee leafminer larvae were cal-culated by multiplying the proportion of mined leaves per plantby the mean number of coffee leafminer larvae per leaf, whichwere assessed from the monthly samples, as described above.Stepwise linear regression analyses were used to assess the influ-ences (as adjusted partial R2 values) of individual mortality factors(parasitism, predation, FOPP) on the dynamics of total coffee leaf-miner mortality (adjusted total R2 value) during the larval or pupalstages; because pupal and larval stage mortality data wereestimated as proportions, they were transformed (a sin

px) for

analyses.

2.4. Contribution of natural enemies to coffee leafminer mortality attwo elevations and rainfall levels

Life table analyses (Southwood, 1978; Bellows and van Driesche,1999) were used to contrast the impacts of natural enemies on cof-fee leafminer survival between one low elevation farm (La Gloria)and the high elevation farm, during periods of low and high rainfall.Coffee leafminer survival and mortality during the larval stage wereassessed by following coffee leafminer cohorts established on 10randomly selected coffee plants. For cohort establishment, onebranch without mined leaves was selected per plant. The apicalthree pairs of leaves on each branch were enclosed in a 30 cm nylonmesh sleeve. Two male–female pairs of coffee leafminer adults(�24 h-old) were released within each sleeve and held therein for48 h. After this time, the two pairs of adults were removed, andthe number of coffee leafminer eggs laid was recorded. Sleeveswere left on each branch for 1 week to exclude egg predators andallow larvae to initiate mines, and removed afterward to exposecoffee leafminer larvae to natural enemies until they completedtheir development (�1 week under local field conditions). At thesame time, 10 coffee leaves with coffee leafminer eggs or pupaereared in the laboratory were exposed to natural enemies by sta-pling them to coffee leaves on each of the 10 selected plants. At least50 eggs (5–15/leaf), 10 pupae (1 or 2/leaf), and 40 larvae (5–38/co-hort) were exposed per plant. All eggs, larvae, and pupae were ex-posed to natural enemies for 1 week. After 1 week, leaves witheggs, mines, or pupae were taken to the laboratory and examinedunder a microscope at 25� to assess stage-specific survivorshiprate, and mortality rates due to parasitoids, predators, or FOPP.Two life tables were constructed as described above at each eleva-tion in 2005. The first life table was initiated on 27 April, during aperiod of low rainfall (<100 mm/month), and the second on 2 June,during a period of high rainfall (>400 mm/month). For each life ta-ble, the numbers of coffee leafminers surviving and dying in eachstage per each of the 10 cohorts were combined for analyses.

Real mortality (the ratio of those dying in a given stage to thenumber initially entering the first stage in a life table) and indis-pensable mortality (the mortality that would not be replaced inthe population by the subsequent action of other mortality factorsif the factor under consideration were removed) (Bellows and vanDriesche, 1999) were used to compare the impacts of predation,parasitism, and FOPP within each elevation and rainfall level.Apparent mortality (the ratio of the number dying in a stage tothe number entering that stage) (Bellows and van Driesche,1999) was used to make comparisons among localities and rainfalllevels within coffee leafminer stages.

Mutual independence among the distribution of mortality ratesdue to specific factors (death by predation or FOPP of eggs or pu-pae, or death by predation, parasitism, or FOPP of larvae), rainfalllevels, and elevations was tested using a 3-dimensional contin-gency table G-test, followed by partial independence tests for thedistribution of mortality rates and elevations or rainfall levels(Zar, 1996). Specifically, the hypotheses that mortality rate distri-butions were independent of elevation and rainfall levels weretested using frequency data from mortality factors acting on eggs,larvae, or pupae from the four life tables that were assembled, i.e.low elevation + low rainfall, low elevation + high rainfall, high ele-vation + low rainfall, and high elevation + high rainfall. Indepen-dence between mortality rate distributions and rainfall levelswas tested by comparing low versus high rainfall level life tableswithin each elevation, and independence between mortality ratedistributions and elevations was tested by comparing low versushigh elevation life tables within each rainfall level. Comparisonsof mortality rates due to parasitoids, predators, or FOPP withineach of the four life tables were made via G-tests, and Tukey-typepost-hoc tests, as warranted (Zar, 1996).

J.R. Lomeli-Flores et al. / Biological Control 51 (2009) 51–60 53


3. Results

3.1. Natural enemy complex associated with coffee leafminer

3.1.1. ParasitoidsA total of 791 parasitoid specimens were collected during our

study, consisting of at least 22 species or morphospecies of Eulo-phidae and Braconidae (Table 1). The parasitoid genera recoveredmost frequently were Neochrysocharis (three species, two morpho-species, 284 specimens), followed by Stiropius (one species, 113specimens), Closterocerus (one species, two morphospecies, 100specimens), Pnigalio (one species, two morphospecies, 98 speci-mens), and Zagrammosoma (two species, 63 specimens). Theremaining five genera were each represented by fewer than 50specimens (Table 1). The five most abundant genera, as listedabove, accounted for >83% of the specimens recovered. Though�50% of the total number of specimens were collected at one ofthe low elevation farms (El Encanto), and �10% at the high eleva-tion farm (Table 1), almost the same numbers of species/morpho-species were recovered at both elevations, 18 and 19, respectively;Allobracon sp. (45 specimens) was recovered only in the low eleva-tion farm.

All parasitoid specimens recovered from coffee leafminer werelarval parasitoids, and most were recovered from mines (�78%,18 species or morphospecies), though some emerged from coffeeleafminer larvae within cocoons (�22%, four species or morpho-species). Primary parasitoids (18 species or morphospecies) repre-sented �82% of the species recovered, secondary parasitoids werenot recovered, and likely facultative secondary parasitoids (fourspecies or morphospecies) were infrequent. Egg or pupal parasit-oids were not recovered.

3.1.2. PredatorsThirteen ant (Formicidae) species were recovered from the 30

coffee plants that were sampled (Fig. 1). The most frequent antspecies were Pseudomyrmex simplex (Smith) (on 60% of the plants),Nesomyrmex echinatinodis (Forel) (37%), and Brachymyrmex sp.

(30%). The remaining 10 ant species were present on <20% of theplants. Brachymyrmex sp. was the only ant species collected on cof-fee plants that was not observed preying on sentinel coffee leaf-miners during our study (Table 2, and see below).

Seventy-eight predation events on sentinel eggs, larvae, or pu-pae were observed during 16 h of diurnal observations, and twoevents during 16 h of nocturnal observations (Fig. 2). Most preda-tion events observed during the day involved ants (77/78), exceptone involving a spider (Salticidae). One predation event observedduring the night involved an ant, Cephalotes sp., and the other acockroach nymph (Blattaria). Predation events most frequently in-volved Camponotus abscisus Roger (�40% of events), Pseudomyrmexgracilis (F.) (�23%), Azteca velox Forel (�11%), and P. simplex(�10%). The remaining species together accounted for �16% ofthe total predation events. Ten predation events were observedon coffee leafminer eggs by four ant species (A. velox, Crematogasterspp., Pseudomyrmex elongatus (Mayr), and Cephalotes sp.), 13events on larvae by four ant species (P. simplex, P. elongatus, P. grac-ilis, and C. abscisus), and 57 events on pupae by six ant species (P.simplex, P. elongatus, P. gracilis, N. echinatinodis, A. velox, and C.abscisus), one spider species (Salticidae), and one cockroach species(Fig. 2). Overall, 17 predator species were observed preying on cof-fee leafminer eggs, larvae, or pupae, including 14 Formicidae, oneVespidae, one Blattaria, and one Salticidae, and the pupa was thecoffee leafminer stage most frequently used as a prey (Table 2,Fig. 2).

3.2. Effects of natural enemies on coffee leafminer populationdynamics

Coffee leafminer was present throughout the study period, June2003–July 2004 (Fig. 3a). Coffee leafminer monthly density (= lar-vae/leaf) varied between 0.1 and 1.4 larvae/leaf, with an average of0.7 ± 0.4 (Fig. 3a). Coffee leafminer density was generally higherduring the rainy relative to the dry (December 2003–March2004) season, was >0.9 larvae/leaf during most (June–October) ofthe 2003 rainy season, and was highest in June–July of the 2004

Table 1Parasitoid species and numbers of specimens per species recovered from coffee leafminer Leucoptera coffeella (Lepidoptera: Lyonetiidae) from coffee farms in the Soconuscoregion of Chiapas, Mexico, June, 2003–July 2005.

Genus/species Habitb No. of specimens (%) by locality Total

Alpujarras El Encanto Othera

N 15� 040 2400 N 14� 590 2100

W 92� 100 1100 W 92� 0905500

960 m 480 m

EulophidaeCirrospilus spp.c L, FS, Ec 2 (2.5) 4 (1.0) 15 (4.7) 21 (2.7)Closterocerus spp.c L, PP, En 26 (32.9) 11 (2.8) 20 (6.3) 57 (7.2)C. cinctipennis Ashmead L, FS, En 1 (1.3) 35 (8.9) 7 (2.2) 43 (5.4)Elachertus spp.c L, PP, Ec 2 (2.5) 16 (4.1) 29 (9.1) 47 (5.9)Horismenus spp.c L, PP, Ec 3 (3.8) 8 (2.0) 2 (0.6) 13 (1.6)Miotropis sp. L, PP, Ec 3 (3.8) 3 (0.8) 0 (0) 6 (0.8)Neochrysocharis spp.c L, PP, En 10 (12.7) 98 (24.9) 87 (27.3) 195 (24.7)N. arvensis Gram. L, PP, En 1 (1.3) 13 (3.3) 14 (4.4) 28 (3.5)N. chalybea Hansson L, PP, En 3 (3.8) 7 (1.8) 13 (4.1) 23 (2.9)N. formosa (Westwood) L, FS, En 1 (1.3) 27 (6.9) 10 (3.1) 38 (4.8)Pnigalio spp.c L, PP, Ec 6 (7.6) 28 (7.1) 39 (12.2) 73 (9.2)P. sarasolai De Santis L, PP, Ec 3 (3.8) 17 (4.3) 5 (1.6) 25 (3.2)Zagrammosoma lineaticeps (Girault) L, PP, Ec 0 (0) 0 (0) 1 (0.3) 1 (0.1)Z. multilineatum (Ashmead) L, PP, Ec 1 (1.3) 30 (7.6) 32 (10.0) 63 (8.0)

BraconidaeAllobracon sp. L, PP, En 0 (0) 45 (11.5) 0 (0) 45 (5.7)Stiropius letifer (Mann) L, PP, En 17 (21.5) 51 (13.0) 45 (14.1) 113 (14.3)

Total (% specimens) 79 (�10) 393 (49.7) 319 (40.3) 791

a Localities where collections were unsystematic.b L, larval parasitoid; PP, primary parasitoid; FS, facultative secondary parasitoid; Ec, ectoparasitoid; En, endoparasitoid.c Includes at least two morphospecies.



rainy season (�1.2 larvae/leaf). The lowest coffee leafminer den-sity, �0.1, occurred during the dry season (December 2003–Febru-ary 2004) (Fig. 3a). Total larval mortality was highest during therainy season and declined during the dry season (Fig. 3b). Preda-tion was the mortality factor that most influenced the dynamicsof total larval mortality, followed by parasitism and FOPP (Table3). Total pupal mortality, which was not measured during mostof the dry season, was highest during rainy-season months(>0.45, except July 2004), and lowest during a dry-season month,March 2004 (Fig. 3c). Predation was the factor that most influencedthe dynamics of total pupal mortality, followed by FOPP, and par-asitism (Table 3).

3.3. Contribution of natural enemies to coffee leafminer mortality attwo elevations and rainfall levels

Coffee leafminer total real mortality was lowest at low elevationunder low rainfall (Table 4a), followed by low elevation under high

rainfall (Table 4b), and highest at high elevation independently ofrainfall level (Table 5a and b) (G = 137.25, df = 3, P� 0.001). Eggpredation contributed the most to total mortality rates indepen-dently of elevation and rainfall levels (Fig. 4); independently ofrainfall level, egg predation contributed >90% of total mortality at

Predator species

P. sim

plex

N. ech

ina.

Brach

ymyr

mex

S. pice

a

A. velo

x

C. absc

isus

C. stri

atus

Crem

atog

aster

C. plan

atus

Cephalo

tes

P. elo

ngatu

s

P. gr

acili

s

S. gem

inat

a

M. f

lorico

la

Blatta

ria

Saltici

dae

Pro

port

ion

(+ S

E)

0.0

0.2

0.4

0.6

Fig. 1. Predatory species collected in a 10 min period per each of 30 coffee plants ina low elevation farm (La Gloria, 550 m), Cacahoatán, Chiapas, Mexico, July, 2005.Bars show the proportion of coffee plants on which each species was collected. N.echina. = N. echinatinodis; see Table 2 for complete names of species.

Table 2Predator species observed preying on coffee leafminer Leucoptera coffeella (Lepidop-tera: Lyonetiidae) eggs, larvae, or pupae in the Soconusco region of Chiapas, Mexico,including 16 h each of diurnal and nocturnal observation in a low elevation coffeefarm (La Gloria, 550 m), Cacahoatán, July, 2005.

Genus/species Coffee leafminer stage used as prey

Egg Larva Pupa

Vespidae unidentified XFormicidaeAzteca velox Forel X XCamponotus abscisus Roger X XCamponotus planatus (Roger) XCamponotus striatus Smith XCephalotes sp. XCrematogaster spp.a XNesomyrmex echinatinodis (Forel) XMonomorium floricola (Jerdon) XPseudomyrmex simplex (Smith) X XPseudomyrmex elongatus (Mayr) X XPseudomyrmex gracilis (F.) X XSolenopsis picea (Emery) XSolenopsis geminata (F.) XBlattaria unidentified XSalticidae (Arachnida) unidentified X

a Includes two morphospecies.

0

5

10

15

20

25

30

Pre

dati

on e

vent

s (f

requ

ency

)

0

5

10

15

20

25

30

Predator species

C. absc

isus

P. gr

acili

s

P. sim

plex

A. velo

x

P. elo

ngatu

s

N. ech

ina.

Blatta

ria

Saltici

dae

Cephalo

tes

Crem

atog

aster

Brach

ymyr

mex

C. plan

atus

C. stri

atus

M. f

lorico

la

S. pice

a

S. gem

inat

a 0

5

10

15

20

25

30

(a) Eggs

(b) Larvae

(c) Pupae

Fig. 2. Predation events by individual predator species on sentinel coffee leafminerLeucoptera coffeella (Lepidoptera: Lyonetiidae) eggs (a), larvae (b), and pupae (c).Predation frequencies are from 384 point (5 s) observations during 16 h of diurnaland 16 h of nocturnal observations (July, 2005) in a low elevation coffee farm (LaGloria, 550 m), Cacahoatán, Chiapas, Mexico. Predator species with nil frequenciesare species known to occur in coffee farms in the study area (see Table 2), thoughthey were not observed during the 32 h of observation. All but two events (Blattaria,one Cephalotes) were recorded during daytime. N. echina. = N. echinatinodis; seeTable 2 for complete names of species.



high elevation and 48–64% at low elevation. Larval parasitism ac-counted for <1.5% of total mortality at high elevation independentlyof rainfall level, whereas it was 12–18% at high elevation (Fig. 4).

At the low elevation farm, coffee leafminer total real mortalitywas lower during the low rainfall period (0.78) versus the highrainfall period (0.89) (G = 27.73, df = 1, P� 0.001) (Table 4a andb). Egg (0.37) and pupal (0.15) predation were the main sourcesof real mortality under low rainfall (Table 4a), whereas egg preda-tion (0.57) and death by FOPP (0.11) were the main sources under

high rainfall (Table 4b). The remainder of mortality factors at bothrainfall levels contributed 60.10 to total real mortality. The mainsources of coffee leafminer apparent mortality (i.e. within coffeeleafminer stages) at both low and high rainfall were egg predation(0.37 at low and 0.57 at high rainfall), larval parasitism (0.18, 0.20),and pupal predation (0.39, 0.48) (Table 4a). Egg (0.13 under lowand 0.15 under high rainfall) and pupal (0.14 under low- and0.10 under high rainfall) predation contributed the most indis-pensable mortality independently of rainfall (Table 4a).

Cof

fee

leaf

min

ers

per

leaf

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Lar

val m

orta

lity

(Pro

port

ion

+ SE

)

0.0

0.2

0.4

0.6

0.8

1.0

Col 21 Plot 2 ZeroMonth vs LaParasitized Month vs LaPredated Month vs LaOther Month vs LaMort

Month

Jun03

Jul03

Aug03

Sep03

Oct03

Nov03

Dec03

Jan04

Feb04

Mar

04

Apr04

May

04

Jun04

Jul04

Pup

al m

orta

lity

(Pro

port

ion

+ SE

)

0.0

0.2

0.4

0.6

0.8

1.0

(a)

(b)

(c)

Dry seasonCoffee leafminerdensity

ParasitismPredation

FOPP

Total mortality

Fig. 3. Coffee leafminer Leucoptera coffeella (Lepidoptera: Lyonetiidae) population dynamics (larvae/leaf) (a), and dynamics of larval (b) and pupal (c) within-stage totalmortality, and mortality due to predation, parasitism, or factors other than predation or parasitism (FOPP) (all mortality shown as within-stage proportion ± SE) at a lowelevation farm (El Encanto, 480 m), Cacahoatán, Chiapas, Mexico. The dry season (<100 mm rainfall/month) was between December 2003 and March 2004, and is indicated byshading.



At the high elevation farm, coffee leafminer total real mortalitywas similar during low and high rainfall 0.96, 0.97, respectively(G = 0.69, df = 1, P = 0.406) (Table 5a and b). Egg predation wasthe main source of real mortality independently of rainfall level(0.87 under both high- and low rainfall) (Table 5a and b). Theremainder of mortality factors at both elevations and rainfall levelscontributed 60.10 to total real mortality. Under both low and highrainfall, the main sources of apparent mortality were egg predation(0.87 under low and 0.86 under high rainfall), larval death due toFOPP (0.32, 0.41), and pupal predation (0.28, 0.41) (Table 5a). Eggpredation was the main source of indispensable mortality indepen-

dently of the rainfall level (0.30 under low and 0.22 under highrainfall); the remaining factors each contributed 60.02 of indis-pensable mortality (Table 5a and b).

4. Discussion

The coffee leafminer natural enemy complex in the Soconuscoregion of Chiapas, Mexico, includes at least 22 parasitoid and 17predator species or morphospecies (Tables 1 and 2, Figs. 1 and2). All parasitoids are parasitic on coffee leafminer larvae, and thusfar parasitoids of eggs and pupae remain unknown (Table 1). Ants(Formicidae) were the most frequently observed predators of sen-tinel coffee leafminer eggs, larvae, and pupae and were the mostfrequently present predators on coffee plants (Figs. 1 and 2, Table2). Coffee leafminer prevalence was generally lower during thedry versus the rainy season in a low elevation farm (Fig. 3a), andpredation was the factor that most influenced total larval and pu-pal mortality (Table 3, Fig. 3b and c). Life table analyses showedthat (i) total real mortality (from egg through larva) was higherat high elevation independently of rainfall level, relative to low ele-vation, and was higher under high rainfall versus low rainfall atlow elevation (Tables 4 and 5); (ii) independently of elevationand rainfall, egg predation yielded the highest real and indispens-able (except at low elevation and low rainfall) mortality (Tables 4and 5); and (iii) predation, generally, was the most importantsource of egg and pupal apparent mortality, and deaths due toFOPP or parasitism were the most important sources of larval

Table 3Statistics corresponding to stepwise linear regression of coffee leafminer (Leucopteracoffeella) (Lepidoptera: Lyonetiidae) mortality during the larval- or pupal-stage due toparasitism, predation, or factors other than parasitism or predation (FOPP) againsttotal larval- or pupal-stage mortality. Data from 14 monthly samples (June 2003–July2004) of larvae, and nine of pupae (June 2003–October 2003, March 2004–July 2004)from a low elevation coffee farm (El Encanto, 480 m) in Cacahoatán, Chiapas, Mexico(see Fig. 3 for 14-month dynamics).

Regression statistic Larval mortality Pupal mortality

F 216.13 90.73df 3, 10 3, 5P <0.0001 0.0001Total R2 (adjusted) 0.980 0.971Predation R2 0.611 0.677Parasitism R2 0.256 0.1181a

FOPP R2 0.113 0.176

a Parasitism by larval–pupal parasitoids.

Table 4Real, apparent, and indispensable mortality rates of cohorts of coffee leafminerLeucoptera coffeella (Lepidoptera: Lyonetiidae) eggs, larvae, and pupae correspondingto predation, parasitism, or death due to factors other than predation or parasitism(FOPP) in a low elevation coffee farm (La Gloria, 550 m), Cacahoatán, Chiapas, Mexico,during a period of (a) low rainfall (April, 2005, <100 mm) or (b) high rainfall (June,2005, >400 mm).

Stage Mortalityfactor

Numberof insects(lx)

Numberof deaths(dx)

Mortality

Real Apparent Indispensable

(a) Low rainfallEgg 677

Predation – 252 0.372 0.37 0.13FOPP – 15 0.022 0.04 0.01

Larva 410Predation – 40 0.059 0.10 0.02Parasitism – 68 0.100 0.18 0.05FOPP – 48 0.071 0.16 0.04

Pupa 254Predation 98 0.145 0.39 0.14FOPP 10 0.015 0.06 0.02

Adulta 146Total – 531 0.784 – –

(b) High rainfallEgg 678

Predation – 385 0.568 0.57 0.15FOPP – 12 0.018 0.04 0.01

Larva 281Predation – 10 0.015 0.04 <0.01Parasitism – 53 0.078 0.20 0.03FOPP – 71 0.105 0.33 0.05

Pupa 147Predation – 70 0.103 0.48 0.10FOPP – 2 0.003 0.03 <0.01

Adulta 75Total – 603 0.889 – –

a Assumes 100% successful adult emergence from pupae not dying from preda-tion or FOPP (Lomelí-Flores et al., unpul. data).

Table 5Real, apparent, and indispensable mortality rates of cohorts of coffee leafminerLeucoptera coffeella (Lepidoptera: Lyonetiidae) eggs, larvae, and pupae correspondingto predation, parasitism, or death due to factors other than predation or parasitism(FOPP) in a high elevation coffee farm (Alpujarras, 960 m), Cacahoatán, Chiapas,Mexico, during a period of (a) low rainfall (April, 2005, <100 mm), or (b) high rainfall(June, 2005, >400 mm).

Stage Mortalityfactor

Numberof insects(lx)

Numberof deaths(dx)

Mortality

Real Apparent Indispensable

(a) Low rainfallEgg 697

Predation 606 0.869 0.87 0.30FOPP 2 0.003 0.02 <0.01

Larva 8Predation – 3 0.004 0.03 <0.01Parasitism – 7 0.010 0.08 <0.01FOPP – 25 0.036 0.32 0.02

Pupa 54Predation – 15 0.022 0.28 0.02FOPP – 8 0.011 0.21 0.01

Adulta 31Total – 666 0.956 – –

(b) High rainfallEgg 488

Predation – 422 0.865 0.86 0.22FOPP – 4 0.008 0.06 <0.01

Larva 62Predation – 4 0.008 0.06 <0.01Parasitism – 4 0.008 0.07 <0.01FOPP – 22 0.045 0.41 0.02

Pupa 32Predation 13 0.027 0.41 0.02FOPP 2 0.004 0.11 <0.01

Adulta 17Total – 471 0.965 – –

a Assumes 100% successful adult emergence from pupae not dying from preda-tion or FOPP (Lomelí-Flores et al., unpul. data).



apparent mortality (Fig. 4). While destructive host feeding by someparasitoids was likely common, e.g., Cirrospilus sp. and Pnigalio sp.(Urbaneja et al., 2001), it could not be assessed from field-collectedmaterial in our study, so an unknown proportion of coffee leafmin-er mortality assigned to FOPP may have been due to parasitoids.

The number of natural enemy species recorded in our study (Ta-bles 1 and 2) was higher than previously reported for coffee leaf-miner in other areas (Villacorta, 1980; Flórez and Hernández,1981; Aranda-Delgado, 1986; Gallardo, 1988; Tozatti and Gravena,1988; Campos et al., 1989; Mendoza, 1995; Pereira et al., 2007).Neochrysocharis, Stiropius, Closterocerus, Pnigalio, and Zagrammoso-ma were the parasitoid genera most frequently collected (Table 1).Some species, such as Stiropius letifer (Mann) and Pnigalio sarasolaiDe Santis have been reported only on coffee leafminer (Whitfield,1988; Noyes, 2003), but other species, such as Closterocerus cincti-pennis Ashmead and Zagrammosoma multilineatum (Ashmead),have broad host ranges (Noyes, 2003). Most parasitoid speciescould not be identified beyond the generic level with the availablekeys, and it is likely that undescribed species are present in thestudy area. The biology of coffee leafminer parasitoids is poorlyknown; overall, 18 of the species recovered were primary parasit-oids, and four were likely facultative secondary parasitoids (Table1).

Coffee leafminer predation has been less studied than parasit-ism, and most prior studies did not consider predation of eggs orpupae (Villacorta, 1980; Tozatti and Gravena, 1988; Reis et al.,2000; Pereira et al., 2007). Previous studies conducted in Brazil,Colombia, Ecuador, and Peru recorded 11 species of Vespidae,one genus of Formicidae, and two of Chrysopidae preying on coffeeleafminer larvae, but pupal and egg predators were not recorded oridentified (Enriquez et al., 1975; Parra et al., 1977; Tozatti and Gra-vena, 1988; Mendoza, 1995; Fragoso et al., 2001; Carvalho et al.,2002, 2005; Fernández and Bueno, 2002; Pereira et al., 2007). Antswere the coffee leafminer predators most frequently encounteredin our study (Table 2, Figs. 1 and 2). At least four species in fourant genera were observed preying on coffee leafminer eggs, fourin two ant genera on larvae, and six in four ant genera on pupae(Fig. 2). Three additional species were collected from coffee plantsduring the course of our study, viz. Pseudomyrmex ejectus (Smith),P. oki (Forel), and P. pallidus (Smith) but were not observed feedingon coffee leafminer (data not shown). Four species of arboreal,twig-nesting ants, C. abscisus, P. gracilis, P. elongatus, and A. velox,were the species most frequently observed preying on coffee leaf-

miner. Quantitative data on the effects of predators on coffee leaf-miner dynamics, available only from Brazil, suggested thatpredatory wasps (Vespidae) would be the main source of coffeeleafminer mortality (Tozatti and Gravena, 1988; Villacorta, 1980;Reis et al., 2000; Pereira et al., 2007). However, predatory waspswere not important mortality factors in our study. Only once wasa predatory wasp observed preying on coffee leafminer larvae innumerous field observations in the Soconusco region between2003 and 2005 (Lomeli-Flores, unpubl. data), including 32 h ofobservation on sentinels, whereas 78 predation events by ants onsentinels were observed. Thus, predatory ants in Soconusco regioncoffee farms may be ecologically equivalent to vespid wasps in Bra-zilian farms, but it is unclear whether studies of coffee leafminerpredation conducted in Brazil, or elsewhere, deliberately ignoredor considered ants. Moreover, the differences in predator speciescomposition also could be due to differences in coffee farm man-agement regimes because Brazilian coffee production is mainly un-der full-sun conditions and farms are intensively managed,whereas Soconusco production is under shade conditions andfarms are traditionally managed. As suggested in prior studies,management regimes likely affect both the microclimatic condi-tions and biodiversity present in coffee farms (Monterrey et al.,2001; Nestel et al., 1994; Philpott, 2005). Accordingly, differencesin natural enemy community compositions and their influenceson coffee leafminer population dynamics between our study andBrazilian studies could be due to differences in coffee farm man-agement regimes and/or microclimatic conditions, but this aspectrequires direct testing.

Ants are the most diverse and abundant organisms in tropicalhabitats, making up >85% of arthropod biomass in forest canopies(Davidson et al., 2003), and may be the most important source of in-sect mortality in tropical rainforests (Way and Khoo, 1992; Florenet al., 2002; Armbrecht and Perfecto, 2003). Correspondingly, antsare one of the most abundant groups of arthropods in traditionallymanaged coffee farms (Ramos-Suarez et al., 2002; Armbrecht andPerfecto, 2003; Philpott and Foster, 2005), where they likely areimportant sources of herbivore mortality (Lachaud et al., 1995; Bus-tillo et al., 2002). Previous studies in Chiapas coffee farms uncov-ered at least 26 ant species associated with coffee plants, most ofthem with predatory habits (Armbrecht and Perfecto, 2003; Phil-pott and Foster, 2005). While those studies did not investigate theimpacts of ants on populations of coffee pests, our study docu-mented 14 ant species predating coffee leafminer, and four addi-

Low/Low

Low/High

High/Low

High/High

Percentage of total mortality

a ef cde bc cd b f

a d d bc b bc d

a c c bc b bc c

a bcbc bc b bc c

Elevation/Rainfall

Larval predation

Egg predation

Egg FOPP

Larval parasitism

Larvae FOPP

Pupal predation

Pupae FOPP

0 20 40 60 80 100

Fig. 4. Mortality factors acting on cohorts of coffee leafminer Leucoptera coffeella (Lepidoptera: Lyonetiidae) eggs, larvae, and pupae at two elevations and two rainfall levels inCacahoatán, Chiapas, Mexico. Mortality factors were predation, parasitism, or death due to factors other than predation or parasitism (FOPP) in a low elevation (La Gloria,550 m) or high elevation (Alpujarras, 960 m) coffee farm, during a period of low rainfall (April, 2005, <100 mm/mo) or high rainfall (June, 2005, >400 mm/month). G-tests andTukey-type post-hoc tests were used for comparisons among proportions of mortality due to parasitoids, predators, or FOPP on coffee leafminer eggs, larvae, and pupaewithin each elevation and rainfall level. Bar segments within each elevation and rainfall level do not differ significantly if the same letter appears above the segments.



tional species present on coffee plants. One recent investigationconducted in a Soconusco region coffee farm showed a significantlynegative relationship between abundance of arboreal, twig-nestingants and coffee leafminer prevalence on coffee plants (de la Moraet al., 2008). The results of that investigation coincide with our re-sults in that twig-nesting ants, in the genera Pseudomyrmex, Neso-myrmex, Camponotus, and Azteca, were the most frequentlyobserved on coffee plants (Fig. 1), and most frequently seen preyingon coffee leafminer (Fig. 2). Overall, the results of our study supportprior suggestions (e.g., Perfecto and Snelling, 1995; Philpott andFoster, 2005; de la Mora et al., 2008) that ants are important preda-tors of coffee pests, including coffee leafminer.

While several studies suggested that parasitism was the mainmortality factor determining coffee leafminer distribution andabundance (e.g., Flórez and Hernández, 1981; Gravena, 1983;Aranda-Delgado, 1986; Campos et al., 1989; Paliz and Mendoza,1993), the results of our study suggested that in the Soconusco re-gion, predation is more important than parasitism. However,though the results of our study showed that total mortality dueto parasitism represented �1/4 of that due to predation, it is likelythat we did not fully quantify the impact of the parasitoids. Underfield conditions, predation of previously parasitized coffee leafmin-er (e.g., Reis et al., 2000) and destructive parasitoid host feeding(Urbaneja et al., 2001) are difficult to quantify, though they likelyaffected the parasitoid contribution to total coffee leafminer mor-tality, as measured in our study.

Previous studies showed that weather conditions influencedcoffee leafminer distribution and abundance, and those of its nat-ural enemies (Reis et al., 1976; Villacorta, 1980; Nestel et al.,1994; Pereira et al., 2007). The life table analyses of our studyshowed that total mortality rates were higher in the high versusthe low elevation farm, and that mortality rates at high elevationwere not affected by rainfall levels, whereas mortality rates atlow elevation were higher under high versus low rainfall. More-over, reproductive rates per generation (R0), calculated from thelife table results (Tables 4 and 5) and a lifetime fecundity valueof 41.9 eggs/female (at constant 25 �C in the laboratory, Lomeli-Flores et al., unpubl. data), are near replacement values in the highelevation farm (R0 = 0.7 under low rainfall, and R0 = 0.9 under highrainfall), whereas they are doubling to quadrupling values in thelow elevation farm (R0 = 2.3 under low rainfall, and R0 = 4.5 underhigh rainfall). In addition, prior studies showed that coffee leafmin-er fecundity was near nil at 20 �C but increased with temperatures>20 �C (Parra et al., 1995; Lomeli-Flores et al., unpubl. data), andthat while rainfall levels and patterns were similar in high andlow elevation Chiapas farms, average temperatures were consis-tently higher (by �4 �C) in low elevation farms (Lomeli-Floreset al., unpubl. data). Together, these results suggested that loweraverage temperatures may partially account for the lower coffeeleafminer densities in the high elevation farm: Lower temperaturesin the high elevation farm likely decreased coffee leafminer fecun-dity and lengthened its developmental time, thus reducing itsreproductive rate and prolonging its exposure to natural enemiesand other mortality factors.

Overall, the results of our study showed that predation of eggsand pupae were the main sources of coffee leafminer mortality inthe Soconusco region of Chiapas, Mexico, and suggested that antswere the main source of coffee leafminer egg and pupal mortality.In high elevation coffee farms, it is likely that high levels of indis-pensable egg mortality by predators coupled with cool tempera-tures lead to lower reproductive rates per coffee leafminergeneration in comparison with low elevation farms. The resultsof our study are in line with hypotheses pointing to ants as themost important mortality factor affecting herbivore populationsin tropical systems (e.g., Way and Khoo, 1992). Moreover, someauthors recommended augmenting ant populations to increase

predation rates and reduce pest population densities in coffeefarms (Philpott and Foster, 2005). At the same time, other authorsshowed that ants interfered with the actions of other natural ene-mies and were associated with higher pest populations suggestingthat biological control of coffee pests could be enhanced by reduc-ing ant densities (Reimer et al., 1993; Infante et al., 2003). The re-sults of our study point to the importance of ants for biologicalcontrol of coffee leafminer, and to the potential of managing antpopulations for enhancing biological control in low elevationfarms. However, because our study was limited in scope (e.g., 13-month population dynamics, three farms), more work is neededto ascertain the net impact of ants on pests, and herbivores gener-ally, including coffee leafminer, in Chiapas coffee farms beforeadvocating their manipulation within the framework of evolvingIPM programs for coffee.

Acknowledgments

We thank Messrs. Joel Herrera M., Enrique López P. RománMontes, Guillermo López G., and Heber García (all at ECOSUR)for their help with the work conducted in Chiapas, Mexico, andFederico Ochoa (Finca Alpujarras), Alejandro Álvarez (Finca LaGloria), Humberto Prada (Finca El Encanto), and Joaquín Erhlich(Finca Lindavista) for permission to work on their farms. We alsothank Eduardo Aránda Delgado (Terranova Lombricultores, Xalá-pa, Mexico) for lending his collection of coffee leafminer parasit-oids, Drs. James Woolley and Robert Wharton (both at TAMU)for their help in identifying coffee leafminer parasitoids, and Dr.Jorge Valenzuela and Mr. Luis Quiroz Robledo (both at Institutode Ecología, Xalapa, Mexico) for their help in identifying ants.Also, the critical reviews of the manuscript by Drs. Carlos Bográn,Roel López, R. Wharton, J. Woolley (all at TAMU), Les Ehler (Uni-versity of California, Davis), and two anonymous reviewers helpedus produce an improved final version. We greatly appreciate thetimely assistance of Editor Dr. Harry Kaya who expedited themanuscript’s handling after significant early delays, for which hewas not responsible. This work was supported in part by CONA-CyT (Mexico) (Doctoral Fellowship to JRLF, #91539), El Colegiode la Frontera Sur (Fundación Produce-Chiapas project Bioecologíay manejo de plagas del café en el Soconusco y Sierra de Chiapas),Texas Agri-Life Research (Hatch Project H-8707), Texas A&M Uni-versity-CONACyT Collaborative Research Grants Program, andColegio de Postgraduados.

References

Aranda-Delgado, E., 1986. Control Natural del Minador de la Hoja del Cafeto enMéxico, Leucoptera coffeella (Guer-Men 1842) (Lep-Lyonetiidae). Bachelor’sthesis Universidad Veracruzana, Xalapa, Veracruz, Mexico.

Armbrecht, I., Perfecto, I., 2003. Litter-twig dwelling ant species richness andpredation potential within a forest fragment and neighboring coffee plantationsof contrasting habitat quality in Mexico. Agriculture, Ecosystems andEnvironment 97, 107–115.

Barrera, J.F., Infante, F., de la Rosa, W., Castillo, A., Gómez, J., 2000. Control biológicode broca del café. In: Badii, M.H., Flores, A.E.L.J., Galan, W. (Eds.), Fundamentos yperspectivas de control biológico. Universidad Autónoma de Nuevo León,Mexico, pp. 211–229.

Bellows, T.S., van Driesche, R.G., 1999. Life table construction and analysis forevaluating biological control agents. In: Bellows, T.S., Fisher, T.W. (Eds.),Handbook of Biological Control: Principles and Applications of BiologicalControl. Academic Press, San Diego, CA, pp. 199–223.

Burks, R.A., 2003. Key to the Nearctic genera of Eulophidae, subfamiliesEntedoninae, Euderinae, and Eulophinae (Hymenoptera Chalcidoidea).University of California, Riverside, CA, 92521; USA. Available from: <http://cache.ucr.edu/%7Eheraty/Eulophidae/> (accessed December 2005).

Bustillo, A.E., Cardenas, R., Posada, F.J., 2002. Natural enemies and competitors ofHypothenemus hampei (Ferrari) (Coleoptera: Scolytidae) in Colombia.Neotropical Entomology 31, 635–639.

Campos, O.G., Decazy, B., Carrillo, E., 1989. Dinámica poblacional del minador de lahoja del cafeto Leucoptera coffeella y sus enemigos naturales en la zona deNuevo San Carlos, Retalhuleu, Guatemala. Turrialba 3, 393–399.



Carvalho, C.E., Silva, R.A., Louzada, J.N.C., Morales, J.C., Barbosa, L.R., Ambrogi, B.G.,2002. Predation of eggs, and larvae, and pupae of coffee leafminer Leucopteracoffeella (Lepidoptera: Lyonetiidae) by Chrysoperla externa (Hagen, 1861)(Neuroptera: Chrysopidae). Ciencia Agrotecnologica. Lavras 26, 318–324.

Carvalho, G.A., Miranda, J.C., Moura, A.P., Rocha, L.C.D., Reis, P.R., Vilela, F.Z., 2005.Controle de Leucoptera coffeella (Guérin-Ménèville and Perrottet, 1842)(Lepidoptera: Lyonetiidae) com insecticides granulados e sus efeitos sobrevespas predadoras e parasitoids. Arquivos do Instituto do Biologia Sao Paulo 72,63–72.

Comisión Nacional del Agua, 2006. Normales climatológicas estándar yprovisionales 1961–1990. Comisión Nacional del Agua, Subdirección GeneralTécnica, Unidad del Servicio Meteorológico Nacional, Mexico. Available from:<http://smn.cna.gob.mx/productos/normales/estacion/normales.html>(accessed 20.11.06).

Davidson, D.W., Cook, S.C., Snelling, R.R., Chua, T.H., 2003. Explaining the abundanceof ants in lowland tropical rainforest canopies. Science 300, 969–972.

de la Mora, A., Livingston, G., Philpott, S.M., 2008. Arboreal ant abundance and leafminer damage in coffee agroecosystems in Mexico. Biotropica 40, 742–746.

Enriquez, E., Bejarano, S., Vila, V., 1975. Morfología, ciclo de vida y comportamientode Leucoptera coffeella Gue.-Men. Revista Peruana de Entomología 18, 79–81.

Fernández, O.A., Bueno, A.F., 2002. MIP-Café. Available from: <www.fcav.unesp.br/Departamentos/fitossanidade/caract/Laboratorios/ECOLAB/cursos/praga-I/cafe.pd>. (accessed 20.01.05).

Floren, A., Biun, A., Linsenmair, K.E., 2002. Arboreal ants as key predators in tropicallowland rainforest trees. Oecologia 131, 137–144.

Flórez, E., Hernández, M.R., 1981. Fluctuación de la población del minador de lashojas del cafeto, Leucoptera coffeella y de sus enemigos naturales en el Valle delCauca. Revista de la Sociedad Colombiana de Entomología 7, 29–38.

Fragoso, D.B., Jusselino-Filho, P., Guedes, R.N.C., Porque, R., 2001. Seletividade deinseticidas a vespas predadoras de Leucoptera coffeella (Guer-Menev.)(Lepidoptera: Lyonetiidae). Neotropical Entomology 30, 139–144.

Fragoso, D.B., Guedes, R.N.C., Picanco, M.C., Zambolim, L., 2002. Insecticide use andorganophosphate resistance in the coffee leaf miner Leucoptera coffeella(Lepidoptera: Lyonetiidae). Bulletin of Entomological Research 92, 203–212.

Gallardo, C.F., 1988. Faunal survey of the coffee leafminer, Leucoptera coffeella,parasitoids in Puerto Rico. Journal of Agriculture of the University of Puerto Rico72, 255–263.

Gibson, G.A.P., Huber, J.T., Woolley, J.B., 1997. Annotated Keys to the Genera ofNearctic Chalcidoidea (Hymenoptera). NRC Research Press, Ottawa, Ontario,Canada.

Gordh, G., 1978. Taxonomic notes on Zagrammosoma, a key to the Nearctic speciesand descriptions of new species from California (Hymenoptera: Eulophidae).Proceedings of the Entomological Society of Washington 80, 344–359.

Gravena, S., 1983. Tácticas de manejo integrado do bicho mineiro do cafeeiroPerileucoptera coffeella (Guerin-Meneville, 1842): I. Dinamica populacional einimigos naturals. Anais da Sociedade Entomologica do Brasil 12, 61–71.

Guerreiro-Filho, O., 2006. Coffee leaf miner resistance. Brazilian Journal of PlantPhysiology 18, 109–117.

Guharay, F., Monterroso, D., Staver, C., 2001. El diseño y manejo de la sombra para lasupresión de plagas en cafetales de América Central. Agroforesteria en lasAméricas 8, 22–29.

Hansson, C., 1987. Revision of the New World species of Chrysocharis Förster(Hymenoptera: Eulophidae). Entomologica Scandinavica Supplement 31, 1–87.

Hansson, C., 1994. Re-evaluation of the genus Closterocerus Westwood(Hymenoptera: Eulophidae), with a revision of the Nearctic species.Entomologica Scandinavica 25, 1–25.

Hansson, C., 1997. Survey of Chrysocharis Förster and Neochrysocharis Kurdjumov(Hymenoptera, Eulophidae) from Mexico, including eight new species.Miscellania Zoologica (Barcelona) 20, 81–95.

Heisswolf, A., Obermaied, E., Poethke, H.J., 2005. Selection of large host plants foroviposition by a monophagous leaf beetle: nutritional quality or enemy-freespace. Ecological Entomology 30, 299–306.

Infante, F., Mumford, J., García-Ballinas, A., 2003. Predation by native arthropods onthe African parasitoid Prorops nasuta (Hymenoptera: Bethylidae) in coffeeplantations of Mexico. Florida Entomologist 86, 86–88.

Lachaud, J.P., Ibarra, G., García, J.A., López, J.A., Moreno, M., 1995. Impacto dedepredación de dos hormigas ponerinas y de algunas arañas tejedoras en losagroecosistemas de café en el Soconusco, Chiapas, México. Memoria del XVIIICongreso Nacional de Control Biológico, Tapachula, Chiapas, México. pp. 105–106.

Mackay, W.P., Mackay, E.E., 2004. Clave de los géneros de de hormigas en México(Hymenoptera: Formicidae). Available from: <http://www3.utep.edu/leb/ants/Mexicoants.doc> (accessed 10.07.04).

Magalha~es, S.T.V., Guedes, R.N.C., Demuner, A.J., Lima, E.R., 2008. Effect of coffeealkaloids and phenolics on egg-laying by the coffee leaf miner Leucopteracoffeella. Bulletin of Entomological Research 98, 483–489.

Mendoza, M.J., 1995. El Minador de la Hoja del Café Perileucoptera cofeella y suControl. Instituto Nacional Autónomo de Investigaciones Agropecuarias, Boletíndivulgativo No. 247. Estación Experimental Tropical Pichilingue, Ecuador.

Monterrey, J., Suarez, D., Gonzalez, M., 2001. Comportamiento de insectos ensistemas agroforestales con café en el Pacífico sur de Nicaragua. Agroforesteriaen las Americas 8, 15–21.

Nantes, J.F.D., Parra, J.R.P., 1977. Queda de folhas em mudas de tres variedades decafé, provocadas por Perileucoptera coffeella (Guérin-Ménèville) [Lepidoptera:Lyonetiidae]. Agros 12, 55–58.

Nestel, D., Dickschen, F., Altieri, M.A., 1994. Seasonal and spatial population loads ofa tropical insect: the case of the coffee leaf-miner in Mexico. EcologicalEntomology 19, 159–167.

Noyes, J.S., 2003. Universal Chalcidoidea Database. Available from:<www.nhm.ac.uk/entomology/chalcidoids/index.html> (accessed 20.01.07).

Paliz, V., Mendoza, J., 1993. Plagas del cafeto. In: Sotomayor, H.I. (Ed.), Manual delCultivo del Café. Estación Experimental Pichilingue, Quevedo, Ecuador, pp. 144–166.

Parra, J.R.P., Gonçalves, W., Gravena, S., Marconato, A.R., 1977. Parasites epredadores do bicho-mineiro do cafeeiro, Perileucoptera coffeella (Guérin-Ménèville, 1842). Anais da Sociedade Entomologica do Brasil 6, 138–143.

Parra, J.R.P., Haddad, M.L., Silveira Neto, S., 1995. Tabela de vida de fertilidade dePerileucoptera coffeella (Guérin-Ménèville, 1842) (Lepidoptera: Lyonetiidae) emtre s temperaturas. Revista Brasileira de Entomologia 39, 125–129.

Pereira, E.J.G., Picanco, M.C., Bacci, L., Crespo, A.L.B., Guedes, R.N.C., 2007. Seasonalmortality factors of the coffee leafminer, Leucoptera coffeella. Bulletin ofEntomological Research 97, 421–432.

Perfecto, I., Snelling, R., 1995. Biodiversity and the transformation of a tropicalagroecosystem: ants in coffee plantations. Ecological Applications 5, 1084–1097.

Philpott, S.M., 2005. Changes in arboreal ant populations following pruning of coffeeshade-trees in Chiapas, Mexico. Agroforestry Systems 64, 219–224.

Philpott, S.M., Foster, P., 2005. Nest-site limitation in coffee agroecosystems:artificial nest maintain diversity of arboreal ants. Ecological Applications 15,1478–1485.

Ramiro, D.A., Guerreiro-Filho, O., Queiroz-Volta, R.B., Matthiesen, S.C., 2004.Caracterizacao anatomica de folhas de caffeiros resistentes e suscetiveis aobicho-mineiro. Bragantia 63, 363–372.

Ramos-Suarez, M.P., Morales, H., Ruiz-Montoya, L., Soto-Pinto, L., Rojas-Fernandez,P., 2002. Se mantiene la diversidad de hormigas con el cambio de bosquemesófilo a cafetales? Revista Protección Vegetal, Actas del Simposio Café yBiodiversidad 12, 17–30.

Reimer, N.J., Cope, M.L., Yasuda, G., 1993. Interference of Pheidole megacephala(Hymenoptera: Formicidae) with biological control of Coccus viridis(Homoptera: Coccidae) in coffee. Environmental Entomology 22, 483–488.

Reis, R.P., Gomez, L.J., Souza, L.J., 1976. Fluctuacao populacional do ‘‘bicho mineiro”das folhas do cafeeiro, Perileucoptera coffeella (Lep: Lyonetiidae), nas regioescaffeiras do Estado de Minas Gerais e identificacao de inimigos naturais.Congreso Brasileiro de Pesquisas Cafeeiras, vol. 4. Resumenes, Caxambú, Brasil.pp. 105–106.

Reis Jr., R., Lima, E.R., Vilela, E.F., Barros, R.S., 2000. Method for maintenance ofcoffee leaves in vitro for mass rearing of Leucoptera coffeellum (Guérin-Ménèville) (Lepidoptera: Lyonetiidae). Anais da Sociedade Entomologica doBrasil 29, 849–854.

Segura, H.R., Barrera, J.F., Nazar, A., Morales, H., 2004. Farmers’ perceptions,knowledge and management of the coffee pests and diseases and their naturalenemies in Chiapas, Mexico. Journal of Economic Entomology 97, 1491–1499.

Southwood, T.R.E., 1978. Ecological Methods, with Particular Reference to the Studyof Insect Populations. Chapman & Hall, London.

Souza, J.C., 1979. Levantamento, identificacao e eficencia dos parasitoi do ‘‘bicho-mineiro” das folhas do cafeeiro, Perileucoptera coffeella (Guérin- Ménèville,1842) no Estado de Minas Gerais. p. 91. Thesis M.S. Piracicaba, Escuela Superiorde Agricultura Luíz de Queiroz, Brasil.

Souza, T., de Oliveira, R.N., Carvalho, G., Magalhaes, L.C., 2003. Ocorrencia de bicho-mineiro do cafeeiro (Leucoptera coffeella) influenciada pelo periodo estacional epela altitude. Acta Scientiarum: Agronomy 25, 119–124.

Tozatti, G., Gravena, S., 1988. Natural mortality factors of Perileucoptera coffeellaGuérin-Ménèville (Lepidoptera, Lyonetiidae) on coffee in Jaboticabal State ofSao Paulo, Brazil. Científica 16, 179–187.

Urbaneja, A., Llácer, E., Garrido, A., Jacas, J., 2001. Effect of variable photoperiod ondevelopment and survival of Cirrospilus sp. nr. lyncus (Hymenoptera:Eulophidae), an ectoparasitoid of Phyllocnistis citrella (Lepidoptera:Gracillariidae). Florida Entomologist 84, 305–307.

Van Achterberg, C., 1995. Generic revision of the subfamily Betylobraconinae(Hymenoptera: Braconidae) and other groups with modified fore tarsus.Zoologische Verhandelingen 298, 1–242.

Villacorta, A., 1980. Alguns factores que afetam a populacao estacional dePerileucoptera coffeella Guerin-Meneville, Norte do Parana, Londrina Pr. Anaisdo Sociedad Entomologica do Brasil 9, 23–32.

Walker, D.W., Quintana, V., 1969. Mating and oviposition behavior of the coffee leafminer, Leucoptera coffeella (Lepidoptera: Lyonetiidae). Proceedings of theEntomological Society of Washington 71, 88–90.

Way, M.J., Khoo, K.C., 1992. Role of ants in pest management. Annual Review ofEntomology 37, 479–503.

Wharton, R.A., Marsh, P.M., Sharkey, M.J., 1997. Manual of the New World Genera ofthe Family Braconidae (Hymenoptera). Special Publications of the InternationalSociety of Hymenopterists 1, London, United Kingdom.

Whitfield, J.B., 1988. Revision of the Nearctic species of the genus StiropiusCameron (=Bucculatriplex Auct.) with the description of a new relatedgenus (Hymenoptera: Braconidae). Systematic Entomology 13, 373–385.

Yoshimoto, C.M., 1983. Review of North American Pnigalio Schrank (Hymenoptera:Eulophidae). Canadian Entomologist 115, 971–1000.

Zar, J.H., 1996. Biostatistical Analysis, vol. 622. Upper Saddle River, Prentice Hall,London, UK.


Impact of natural enemies on coffee leafminer Leucoptera coffeella (Lepidoptera: Lyonetiidae)...

Documents

Transcript of Impact of natural enemies on coffee leafminer Leucoptera coffeella (Lepidoptera: Lyonetiidae)...