Indigenous Natural Enemies Associated with Phyllocnistis citrella (Lepidoptera: Gracillariidae) in Eastern Spain

Biological Control 18, 199 –207 (2000) doi:10.1006/bcon.2000.0830, available online at http://www.idealibrary.com on

Indigenous Natural Enemies Associated with Phyllocnistis citrella (Lepidoptera: Gracillariidae) in Eastern Spain ´ scar Toma´s, Antonio Garrido, and Josep-Anton Jacas 1 Alberto Urbaneja, Elena Lla´cer, O Departament de Proteccio´ Vegetal i Biotecnologia, Institut Valencia` d’Investigacions Agra`ries, Ctra. de Montcada a Na`quera km 5, E-46113 Montcada, Spain Received February 24, 1999; accepted March 1, 2000

The incidence of generalist indigenous natural enemies of the citrus leafminer (CLM), Phyllocnistis citrella Stainton (Lepidoptera; Gracillariidae), was monitored during three growing seasons at two different orchards located in the major citrus-growing area of Spain. Composition of the parasitoid complex changed during the study period. However, the eulophids Cirrospilus near lyncus Walker and Pnigalio pectinicornis L. were consistently the predominant species. Despite the varying composition of the parasitoid complex, oviposition, host feeding, and predatory preferences of the natural enemies of the CLM clearly centered on third instar larvae. Incidence of beneficial fauna increased as the season progressed, reaching maximal values up to 70% of susceptible leafminers (mature larvae) at the end of the summer. Parasitism was significantly related to relative host density. However, predation showed no relationship to host availability but did so to flushing in one of the orchards. Incidence of indigenous natural enemies of the CLM should not be ignored when planning any introduction of exotic parasitoids, and their conservation should be taken into account when planning any citrus IPM strategy. © 2000 Academic Press Key Words: Phyllocnistis citrella; indigenous natural enemies; Eulophidae; Spain.

The citrus leafminer (CLM), Phyllocnistis citrella Stainton (Lepidoptera: Gracillariidae), is a pest native to Southern Asia which has spread to almost all citrusgrowing areas worldwide within the past 5 years (CABI, 1995; Hoy and Nguyen, 1997). It was first detected in southern Spain in summer 1993 (Garijo and Garcı´a, 1994) and 1 year later it had colonized the whole country, except the Canary Islands. By 1995 it had spread throughout the Mediterranean Basin. Because of the extensive mining of young foliage caused 1

To whom correspondence should be addressed. Present address: Departament de Cie`ncies Experimentals, Universitat Jaume I, Campus de Riu Sec, E-12071 Castello´ de la Plana, Spain.

by this pest at that time, it was considered a serious threat to citriculture. Citrus leafminer was so intensively treated with pesticides that it was ranked fourth in the amount of pesticide sales for insect control in Spain (Puiggro`s, 1998). Currently, its damage to mature trees under typical Mediterranean conditions is considered only esthetical (Gonza´lez, 1997; Granda et al., 1997, 1999), but the CLM is regarded as an important pest within nurseries and on young plants and top-grafted trees. When the CLM arrived in Spain, it was rapidly accepted as a host by some indigenous parasitoids of other leafminers. Thus, in 1994, Garrido and Del Busto (1994) identified 4 different eulophid species on P. citrella: Pnigalio sp. (later identified as P. pectinicornis L.), Cirrospilus pictus Nees, C. vittatus Walker, and Sympiesis sandanis (Walker). By 1997, 7 more species were added to the list. These were the eulophids Cirrospilus near lyncus Walker, Sympiesis gregori Boucek, Diglyphus isaea Walker, Chrysocharis pentheus (Walker), Neochrysocaris formosa (Westwood), and Ratzeburgiola cristata (Ratzeburg) and a pteromalid species (Gonza´lez et al., 1996; Lacasa et al., 1998; Urbaneja et al., 1998a; Vercher et al., 1998). In the Mediterranean Basin as a whole, about 30 different species have been recorded since 1994 that parasitize the CLM (FAO, 1996; Schauff et al., 1998). Although extensive faunistic surveys were carried out throughout the Mediterranean Basin, the actual role of these natural enemies in regulation of P. citrella populations remains unknown, and biological control efforts have been directed toward introduction of exotic parasitoid species (Argov and Ro¨ssler, 1998, 1996; Lekchiri, 1996; Siscaro et al., 1997). In Spain, for instance, Cirrospilus ingenuus Gahan (Hymenoptera: Eulophidae), Semialacher petiolatus (Girault) (Hymenoptera: Eulophidae) and Ageniaspis citricola Longvinovskaya (Hymenoptera: Encyrtidae) were imported in 1995, followed by Quadrastichus sp. (Hymenoptera: Eulophidae) in 1996 and the eulophids Galeopsomyia fausta LaSalle and Citrostichus phyllocnistoides (Narayanan) in 1998. Preintroduction studies on indigenous natural enemy

199

1049-9644/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.

200

URBANEJA ET AL.

impact is recognized as one of the requisites in any classical biological control program (Barbosa and Segarra-Carmona, 1993) and, although generalist indigenous natural enemies can result in a satisfactory control of an introduced pest (DeBach, 1974), classical biological control programs developed against the CLM in the Mediterranean Basin have consistently neglected this step. To provide insights into the role played by indigenous natural enemies of the CLM, the objective of the present study was to monitor incidence and to quantify the impact of these natural enemies at two different locations in the Valencia region, the major citrus-growing area of Spain. MATERIALS AND METHODS

Flushing patterns and incidence of both the CLM and its parasitoids were followed from February 1997 to November 1999 at 15-day intervals at two different locations: Elx (38.3° N 0.7° W, 86 m altitude) and Montcada (39.6° N 0.4° W, 33 m altitude). The orchard at Elx was 0.6 ha and consisted of 4-year-old Navelina sweet orange trees, Citrus sinensis (L.) Osbeck, surrounded by other citrus orchards (predominantly lemon trees, C. limon L.) and date palm (Phoenix dactylifera L.) and pomegranate (Punica granatum L.) groves. This orchard had received no pesticide treatments during 1996, but a foliar spray of fenthion (Lebaycid, Bayer) was applied on 5 November 1997 against the Mediterranean fruit fly, Ceratitis capitata Wiedemann (Diptera: Tephritidae). The orchard at Montcada was 0.4 ha and consisted of 9-year-old Verna lemon trees, surrounded by other citrus orchards. This orchard had not received any chemical treatment since 1994 and it remained untreated during the study period. Both orchards were drip-irrigated on bare soil and were previously monitored in 1996 (Urbaneja et al., 1998a) Densities of young foliage canopies were estimated because of dependence of CLM damage on citrus flushing patterns (Jacas et al., 1997). Twenty trees were randomly chosen at each sampling date. A ring, 40 cm in diameter, was thrown onto the tree four times per tree, and the number of total flushes, egg-receptive flushes (exhibiting leaves less than 5 mm long, Garrido and Gasco´n, 1995), and flushes housing any stage from larva to pupa of the CLM found within the ring were counted. Additional flushes, or suckers where no flushes were available, were also collected. These were put into a plastic bag and refrigerated until reaching the laboratory, where 100 infested leaves were observed under a stereoscopic binocular microscope for the presence of mines (either occupied or abandoned), CLM larvae and pupae, and parasitoid immature stages. Dead citrus leafminer stages exhibiting black spots were considered products of host feeding by para-

sitoids. Predation was assumed to be the reason for incomplete mines or mines where mutilated P. citrella larvae were found. Because indigenous idiobiont parasitoids caused permanent paralysis when stinging the host, the instar of parasitized hosts was recorded to establish ovipositional and feeding preferences of beneficial fauna. Percentage parasitism, feeding punctures, and predation were calculated based on number of susceptible CLM stages as determined during this study. From 1996 to 1998, additional infested leaves were enclosed in cages, where emerging parasitoids were periodically collected and prepared for taxonomic determination. Voucher specimens were deposited at the reference insect collection located at the Institut Valencia` d’Investigacions Agra`ries. Density dependence of natural enemies was studied by plotting the different components of mortality due to entomophages against the corresponding mean population densities, and corresponding regression equations were calculated using the Statgraphics software package (STSC, 1987). RESULTS

A total of 10,710 mines housing 6330 apparently healthy CLM larvae and pupae, 1038 parasitized specimens, 1208 punctured larvae and 2133 immatures exhibiting evidence of predation were found in 4200 leaves sampled in Elx. From 3700 leaves monitored in Montcada, 14,075 mines sheltering 9239 apparently healthy P. citrella larvae and pupae, 1151 parasitized individuals, 1305 pierced larvae, and 2375 immatures exhibiting evidence of predation were observed. In addition 6488 adult wasps were recovered from the samples taken in Elx and Montcada during the study period. Flushing Pattern Spring flush appeared in Elx from February to March (Fig. 1). The most abundant flush throughout the study period was registered on 17 February 1997 (352.6 ⫾ 42.3 flushes/m 2) and on 3 March 1999 (375.8⫾ 30.3 flushes/m 2). In May–June a new flush appeared at a much lower density and this situation was maintained quite homogeneously until October both in 1997 and 1998 (mean: 18.0 ⫾ 9.2 flushes/ m 2), providing a continuous supply of suitable leaves for the development of both the citrus leafminer and its natural enemies. In 1999, summer flushing was less regular and abundant (mean: 9.0 ⫾ 9.5 flushes/ m 2). Spring flush did not appear in Montcada until April– May (Fig. 2) but flushing patterns changed considerably from 1997 to 1998 and 1999. The spring flush was also the most abundant flush of the year, but it was considerably less abundant than that found in Elx (10 June 1997: 27.8 ⫾ 4.3 flushes/m 2) and summer and fall flushes were even more reduced and very irregular

NATURAL ENEMIES ASSOCIATED WITH CITRUS LEAFMINER

201

FIG. 1. Flushing patterns (number of flushes/m 2) and incidence of citrus leafminer and its parasitoids (mines and individuals per leaf, respectively). Receptive flush refers to flushes exhibiting leaves smaller than 5 mm, which are preferred over older leaves for oviposition. Navelina orchard in Elx. On 5 November 1997, fenthion was applied against the medfly. Date (day/month/year).

(mean: 3.5 ⫾ 3.0 flushes/m 2). In contrast, in 1998 and 1999 trees exhibited a much more abundant and regular flushing during the whole growing season (21 April 1999: 38.5 ⫾ 4.79 flushes/m 2; mean: 10.4 ⫾ 3.6 flushes/m 2). These differences were attributed to deficient irrigation during 1997. Incidence of the Citrus Leafminer P. citrella was not found during the first flush in either Elx or Montcada. Citrus leafminer populations slowly increased and summer densities ranged from

2.0 to 4.5 mines per leaf in Elx (Fig. 1) and from 1.3 to 5.3 mines per leaf in Montcada (Fig. 2). Mean numbers of CLM larvae and pupae per leaf were 3.2 and 5.1 in Elx and Montcada, respectively. High leafminer densities reported in Montcada when no flush was produced corresponded to samples taken from suckers. The pest population concentrated on the suckers during these periods and, as a consequence, the highest densities were detected on 16 October 1997 (6.6 mines/leaf), 26 October 1998 (10.0 mines/leaf), and 21 October 1999 (8.4 mines/leaf).

202

URBANEJA ET AL.

FIG. 2. Flushing patterns (number of flushes/m 2) and incidence of citrus leafminer and its parasitoids (mines and individuals per leaf, respectively). Receptive flush refers to flushes exhibiting leaves smaller than 5 mm, which are preferred over older leaves for oviposition. Lemon orchard in Montcada. Date (day/month/year).

In 1999, the percentage of CLM-infested leaves per shoot was estimated. Both in Elx and Montcada, dynamics were similar. About 1 month after detection of the CLM, percentages of infested leaves increased from 30% to around 90%. From that moment onward percentages recorded at Elx (86.7 ⫾ 8.6%) and Montcada (96.2 ⫾ 2.5%) were significantly different (F ⫽ 5.39; df ⫽ 1,15; P ⫽0.0358). Incidence of Parasitism and Predation Immature parasitoid densities were much lower than those recorded for the citrus leafminer. The num-

ber of parasitoids per leaf ranged between 0.01 and 0.9 in Elx (Fig. 1) and between 0.1 and 1.6 in Montcada (Fig. 2). The impact of parasitoids, expressed as actual parasitization and host feeding, and of predators, expressed in terms of dead leafminers, increased as the season progressed and reached a maximum value of around 70% at the end of the summer in Elx (Fig. 3) and of 50 – 60% in Montcada (Fig. 4). In general, the contribution of parasitism (excluding host feeding) to overall mortality was less than that from predation, especially at the beginning of the season, when parasitism was lowest. Predation levels recorded in Elx

NATURAL ENEMIES ASSOCIATED WITH CITRUS LEAFMINER

203

FIG. 3. Incidence of predation, feeding punctures, and parasitism on citrus leafminer immatures (% of susceptible leafminer stages). Navelina orchard in Elx. Date (day/month/year).

greatly increased in 1998, but changed very little in Montcada. CLM Stage Incidence: Oviposition, Host Feeding, and Predation Preferences of the CLM natural enemies recorded for either parasitism, host feeding, or predation (Table 1)

were similar at the two locations considered. Third instars were the preferred target in all cases, receiving about 60% of the eggs and experiencing around 70% of both the feeding punctures and the predation. Fourth instars followed in importance, accounting for about 25% of the ovipositon, 9% of the feeding punctures, and 15% of the predation. Feeding punctures were never

FIG. 4. Incidence of predation, feeding punctures, and parasitism on citrus leafminer immatures (% of susceptible leafminer stages). Lemon orchard in Montcada. Date (day/month/year).

204

URBANEJA ET AL.

TABLE 1 Relative Oviposition, Feeding, and Predatory Preferences on the Citrus Leafminer (Percentages and Actual Counts) in a Navelina Orchard in Elx and a Lemon Orchard in Montcada Elx Oviposition

Montcada

Feeding

Predation

Oviposition

Feeding

Predation

Host stage

%

Count

%

Count

%

Count

%

Count

%

Count

%

Count

Second Third Fourth Pupa

1.5 53.8 28.3 16.4

16 562 296 171

16.4 74.0 9.6 —

198 894 116 —

10.1 64.3 17.6 8.0

215 1371 376 171

1.6 59.0 27.2 12.3

18 679 313 141

16.2 76.2 7.6 —

212 994 99 —

10.0 70.3 15.3 4.4

237 1669 364 105

observed on pupae, but this stage supported about 10% of the eggs and 5% of the predation. Finally, second instars received just about 1% of the eggs, but were considerably better accepted as prey, accounting for around 15% of the feeding punctures and 9% of the predation. Parasitic Complex Composition In contrast to the preferences presented above, composition of the parasitoid complex developing on the CLM was not identical at the locations studied and changed from 1996 to 1998 (Table 2). Pnigalio pectinicornis was the predominant species in Elx in 1997, representing 57.1% of the total, followed by Cirrospilus nr. lyncus (25.8%) and C. pictus (11.2%). Cirrospilus vittatus represented 0.8%, and the introduced species Ageniaspis citricola and Quadrastichus sp. represented 4.6 and 0.5%, respectively. These exotic parasitoids had been released in May 1997 in another citrus orchard about 200 m distant. Nevertheless, P. pectini-

cornis lost its predominance to C. nr. lyncus in 1998. Exotic parasitoids introduced in 1997 were not detected in 1998 and only one individual of the recently introduced Galeopsomyia fausta was recovered that year. In Montcada, C. nr. lyncus was clearly predominant in both 1997 and 1998 and represented around 80% of recovered parasitoids, whereas P. pectinicornis amounted to about 10%, almost the same amount as A. citricola, which had been introduced in 1998. Quadrastichus sp., which was also released in 1998, represented only 0.2% of the total. Also in 1998, Pnigalio soemius (Walker) appeared for the first time in our samples. Mortality, Parasitism, and Predation as Functions of CLM Larval Densities Relationships between mortality inflicted by natural enemies (percentage parasitism, host feeding, and predation) and relative host densities (No. mines per leaf)

TABLE 2 Composition of Citrus Leafminer Parasitoid Guilds (Percentages and Actual Counts) in a Navelina Orchard in Elx and a Lemon Orchard in Montcada Elx 1996

Montcada

1997

1998

1996

1997

1998

Parasitoid species

%

Count

%

Count

%

Count

%

Count

%

Count

%

Count

Cirrospilus sp. near lyncus Pnigalio pectinicornis Cirrospilus pictus Ageniaspis citricola Cirrospilus vittatus Sympiesis gregori Pteromalidae sp. Pnigalio soemius Quadrastichus sp. Chrysocharis pentheus Neochrysocharis formosa Galeopsomyia fausta

40.7 54.2

121 161

461 301 122

42.5 54.9

697 900

85.6 11.0 1.5

1398 179 24

15

296 654 128 53 9

50.6 33.0 13.4

5.1

25.8 57.1 11.2 4.6 0.8

0.7

6

2.1 0.4

34 7

2.3

21

0.7 1.0 0.3

11 16 5

72.7 10.4 0.6 12.1 1.2 1.7 0.2 1.0 0.2

1967 282 7 327 31 45 5 26 5

0.1

1 0.1

1

0.5

6

Note. Data from 1996 Obtained from Urbaneja et al. (1998a).

0.1

1

NATURAL ENEMIES ASSOCIATED WITH CITRUS LEAFMINER

were similar at the two orchards investigated. No relationship existed between these variables at Montcada (F ⫽ 2.290; df ⫽ 1,38; P ⫽ 0.139) or Elx (F ⫽ 1.520; df ⫽ 1, 37; P ⫽ 0.226). When mortality due to parasitism (percentages of oviposition and host feeding) was studied separately and related to relative host density, a significant relationship was found in both Montcada (y ⫽ 0.069 x ⫹ 3.203; r ⫽ 0.347; F ⫽ 5.05; df ⫽ 1,38; P ⫽ 0.031) and Elx (y ⫽ 0.026 x ⫺ 2.539; r ⫽ 0.419; F ⫽ 7.650; df ⫽ 1, 37; P ⫽ 0.009). Predation (%) was not related to relative host density in Montcada (F ⫽ 0.010; df ⫽ 1, 38; P ⫽ 0.940) or Elx (F ⫽ 0.360; df ⫽ 1, 37; P ⫽ 0.559) Mortality, Parasitism, and Predation as Functions of Flushing Abundance Mortality due to natural enemies was not related to densities of young receptive flushes (No. flushes/m 2) in Elx (F ⫽ 0.970; df ⫽ 1, 37; P ⫽ 0.330), but was in Montcada (y ⫽ 0.741 x ⫺ 0.082; r ⫽ ⫺0.829; F ⫽ 81.060; df ⫽ 1, 38; P ⬍ 0.0000). Mortality due to parasitism (percentages of oviposition and host feeding) showed no relationship to densities of young receptive flushes in Montcada (F ⫽ 1.150; df ⫽ 1, 38; P ⫽ 0.290) or Elx (F ⫽ 0.760; df ⫽ 1,37; P ⫽ 0.390). Predation (%) was not related to densities of young receptive flushes in Montcada (F ⫽ 0.410; df ⫽ 1,38; P ⫽ 0.525) or Elx (F ⫽ 0.360; df ⫽ 1,37; P ⫽ 0.551). DISCUSSION

Flushing patterns observed during this study followed the general trends of citrus trees grown along the Mediterranean coastal districts of Spain. Spring flush represents about 80 –90% of total leaves produced during the whole season, whereas successive flushes are much less abundant (Granda et al., 1999). These successive flushes, where the CLM could complete from 10 to 13 overlapping generations, were more copious and regular, except in Montcada in 1997, where flushing was poor and intermittent. P. citrella remained almost unnoticed during the spring flush and CLMblemished foliage could hardly be detected. However, almost every new flush appearing from late-May onward was infested. Despite the conspicuous presence of CLM, studies carried out recently have demonstrated that, under Mediterranean conditions, leafminer damage on adult trees usually lacks economic importance (Gonza´lez, 1997; Granda et al., 1997, 1999). Leafminer densities were very similar to those found during 1996 in the same orchards (e.g., about 4.0 mines per leaf and 3.0 citrus leafminer larvae per leaf) (Urbaneja et al., 1998a). In contrast, parasitoid counts were considerably higher than those found in 1996. In Elx, this density barely exceeded 0.5 individuals per leaf, the maximal value registered in 1996 (Urbaneja et al.,

205

1998a), whereas during the study period, this value was exceeded repeatedly during the summer and almost reached 1.0 individual per leaf at the end of August 1997. In Montcada, maximum density was registered in 1998 (7 October 1998: 1.6 parasitoids per leaf), and that was more than double the maximum rate found in 1997 (16 September 1997: 0.7 parasitoids per leaf). Composition of parasitoid fauna in Elx in 1997 was almost the same as that found in 1996, when P. pectinicornis and the complex C. pictus and C. nr. lyncus represented about 50% each (Urbaneja et al., 1998a). However, this changed in 1998 and approached proportions found in Montcada in both 1997 and 1998, when C. nr. lyncus was clearly predominant. In 1996, the composition of the parasitic complex in Montcada was very similar to that found in Elx in both 1996 (Urbaneja et al., 1998a) and 1997. Similar variability in CLM parasitoid fauna composition was reported for different locations in Spain (Lacasa et al., 1998; Urbaneja et al., 1998a; Vercher et al., 1998). Leafmining insects typically exhibit a larger parasitoid species load than both free-living insects and borers (Hawkins, 1988; Hawkins and Lawton, 1987). Both temporal and spatial variation in parasitoid guilds are common phenomena observed when studying leafminers (Hawkins and Mills, 1996), as reported for other leafmining species (Cossentine and Jensen, 1994; Johnson and Hara, 1987). It is worth noting that the predominant species recorded in this study were not found in a survey of native parasitoids of leafminers of horticultural crops carried out in the same region (Verdu´, 1991). This is in agreement with the general rule that leafminer parasitoid species are usually shared on the basis of plant host (Hespenheide, 1991), in this case, trees and shrubs versus herbs. C. nr. lyncus is a new record for Spanish fauna, and its original host range remains unknown. Contrarily, P. pectinicornis exhibits a wide host range, mainly lepidopterous and coleopterous leafminers on deciduous trees (Boucek and Askew, 1968) and has been studied in detail as a parasitoid of the olive fruit fly, Bactrocera oleae (Gmelin) (Diptera: Tephritidae) (Neuenschwander et al., 1986). Until now, exotic parasitoids appearing in the sampled orchards either have not showed any significant effect on leafminer populations (Quadrastichus sp., G. fausta) or have not successfully overwintered and have therefore not been detected the year following their introduction (A. citricola). A. citricola was mass reared and periodically introduced in citrus orchards all over Spain during the period 1996 –1998. Although it rapidly colonized the release sites, it could never be recovered the following year. The Canary Islands were the only exception. One single release of this species in 1997 resulted in satisfactory establishment of this species in both Gran Canaria and Tenerife. Subtropical climate providing a continuous supply of citrus shoots

206

URBANEJA ET AL.

where the leafminer and its natural enemies could endure is likely the cause of this success. Competition between the predominant parasitoids, C. nr. lyncus and P. pectinicornis, may be indicated by similarities in relative incidence of parasitism among lifestages: third instars ⬎⬎ fourth instars ⬎ pupa ⬎ second instars (Table 1). Laboratory results with C. nr. lyncus confirmed these preferences (Urbaneja et al., 1999, 1998b). However, a field survey carried out by Vercher et al. (1998) established this pattern for C. nr. lyncus, but a different one for P. pectinicornis (fourth instars ⬎ third instars ⬎ pupa). Competition may also take place among parasitoids and predators when host feeding, because the pattern third instars ⬎⬎ second instars ⬇ fourth instars (Table 1) was observed in both Elx and Montcada. Although predators were not directly sampled during this study, Chrysoperla carnea Stephan (Neuroptera: Chrysopidae), Chiracanthium sp. (Araneae: Clubionidae), Orius sp. (Heteroptera: Anthocoridae), and different ant species (Hymenoptera: Formicidae) were frequently observed feeding upon immature leafminers, especially when aphids were also present in the orchard. These observations, supplemented with laboratory studies on C. nr. lyncus (Urbaneja et al., 1999, 1998b), provide further evidence that indigenous natural enemies of the leafminer can exert a substantial effect on mature larvae of the CLM, but eggs, young larvae and pupae remain largely ignored by both indigenous predators and parasitoids, whose feeding preferences were directed predominantly toward third instars (Table 1). For this reason, successful introduction of egg and young larval parasitoids, such as A. citricola (Edwards and Hoy, 1998), and pupal parasitoids, such as G. fausta (LaSalle and Pen˜a, 1997), would probably increase overall impact of beneficial fauna on the citrus leafminer. Impacts of the beneficial fauna on CLM populations (Figs. 3 and 4) increased as the season progressed. Regressions relating density-dependent mortality caused by natural enemies to CLM indicators were similar at the locations studied. Mortality was not related to relative host density and was related to flushing only in Montcada. When the two main components of mortality (parasitism and predation) considered in this study were examined separately, it was found that parasitism (including oviposition and host feeding) depended on host availability at both locations, whereas predation exhibited no relationship to any of the parameters considered. Therefore, it appeared that parasitoids, even if mostly generalist eulophids, rapidly accepted the CLM as a new host and thus responded to variations in its density, whereas polyphagous predators did not. Nevertheless, when the relationship between predation (%) and flushing (No. flushes/m 2) was studied in Elx only during the period 1997–1998 (years with regular summer–fall flushing) it was found that predation was significantly related to densities of

young receptive flushes (y ⫽ 0.609 x ⫹ 11.371; r ⫽ 0.430; F ⫽ 6.14; df ⫽ 1,27; P ⫽ 0.020). Therefore, it appeared that predators responded to changes in flushing when this was regular, and this may be related to continuous availability of preferential hosts, such as aphids. A study of aphid incidence in different citrus varieties in different areas of the Valencia region (Hermoso de Mendoza et al., 1986) demonstrated that Navelina varieties usually support higher aphid populations than lemon trees, and this difference can be as high as 10-fold. Observations made during this study are in agreement with these results. Aphids were especially abundant on Navelina trees, whereas they could hardly be detected on lemon. Therefore, generalist predators feeding on aphids, such as lacewings, and aphid symbionts, such as ants, which also feed on CLM, were probably responsible for the higher predation rates observed in 1998 in Elx. In summary, our results are indicative of a significant association between CLM natural enemies and the CLM. Although indigenous fauna consisted of generalist eulophids in addition to an undetermined number of polyphagous predators, the parasitoids’ effects appeared to be density dependent. According to the results presented, opportunistically recruited entomophagous insects can have a significant and practical effect on CLM populations (up to 70% reduction at the end of summer), and their contribution should not be ignored when planning importations of additional natural enemies. Opportunities for introduction exist, and conservation of indigenous natural enemies of CLM should be taken into account when planning any citrus IPM strategy. ACKNOWLEDGMENTS J. Pen˜a (TREC, University of Florida, Homestead, FL), E. Vin˜uela (Unidad de Proteccio´n de Cultivos, Universidad Polite´cnica de Madrid, Spain), L. Navarro (IVIA, Montcada), and two anonymous reviewers provided valuable comments on the manuscript. We are also grateful to M. J. Verdu´ (IVIA, Montcada) for her assistance in identifying parasitoids, to R. Hinarejos (IVIA, Montcada) for his help in the laboratory, to E. Carbonell (IVIA, Montcada) for advice on the statistical analyses, to J. LaSalle (International Institute of Entomology, London, UK) for useful discussions on leafminer parasitoids, and to the Escola de Capacitacio´ Agra`ria d’Elx for the use of their orchards. This work was partially funded by the Instituto Nacional de Investigacio´n y Tecnologı´a Agraria y Alimentaria, INIA, and the Conselleria d’Agricultura, Pesca i Alimentacio´ de la Generalitat Valenciana. E. L. and A. U. were recipients of a scolarship from IVIA.

REFERENCES Argov, Y., and Ro¨ssler, Y. 1998. Rearing methods for the citrus leafminer Phyllocnistis citrella Stainton and its parasitoids in Israel. Biol. Control 11, 18 –21. Argov, Y., and Ro¨ssler, Y. 1996. Introduction, release and recovery of several exotic natural enemies for biological control of the citrus leafminer, Phyllocnistis citrella, in Israel. Phytoparasitica 24, 33– 38.

NATURAL ENEMIES ASSOCIATED WITH CITRUS LEAFMINER Barbosa P., and Segarra-Carmona, A. 1993. Criteria for the selection of pest arthropod species as candidates for biological control. In “Steps in Classical Arthropod Biological Control” (R. G. Van Driesche and T. S. Bellows, Jr., Eds.), pp. 5–23. Proceedings Thomas Say Publications in Entomology. Entomological Society of America, Lanham. Boucek, Z., and Askew, R. R. 1968. Hym. Chalcidoidea. Palearctic Eulophidae (excl. Tetrastichinae). In “Index of Entomophagous Insects”(V. Delucchi and G. Remaudie`re, Eds.), pp. 1–254. Le Franc¸ois, Paris. Commonwealth Agricultural Bureau International (CABI). 1995. “Distribution Maps of Pests.” Series A: Map No. 274 (2nd revision). Cossentine, J. E., and Jensen, L. B. The role of two eulophid parasitoids in populations of the leafminer, Phyllonorycter mespilella (Lepidoptera: Gracillariidae) in British Columbia. J. Entomol. Soc. Br. Columbia 91, 47–54. DeBach, P. 1974. “Biological Control by Natural Enemies.” Cambridge Univ. Press, London. Edwards, O. R., and Hoy, M. A. 1998. Biology of Ageniaspis citricola (Hymenoptera: Encyrtidae), a parasitoid of the leafminer Phyllocnistis citrella (Lepidoptera: Gracillariidae). Ann. Entomol. Soc. Am. 91, 654 – 660. FAO. 1996. “Report of the Workshop on Citrus Leafminer (Phyllocnistis citrella) and its Control in the Near East.” Safita (Syria), 30 September–3 October 1996. Garijo, C., and Garcı´a, E. 1994. Phyllocnistis citrella (Stainton, 1856) (Insecta: Lepidoptera: Gracillariidae: Phyllocnistinae) en los cultivos de Andalucı´a (Sur de Espan˜a): Biologı´a, Ecologı´a y Control de la Plaga. Bol. San. Veg. Plagas 20, 559 –571. Garrido, A., and Del Busto, T. 1994. Enemigos de Phyllocnistis citrella Stainton encontrados en Ma´laga. Invest. Agr. (fuera de serie) 2, 87–92. Garrido, A., and Gasco´n, I. 1995. Distribucio´n de fases inmaduras de Phyllocnistis citrella Stainton segu´n el taman˜o de la hoja. Bol. San. Veg. Plagas 21, 559 –571. Gonza´lez, L. 1997. Dan˜os causados por los ataques de Phyllocnistis citrella Stainton (Lepidoptera: Gracillariidae) y su repercusio´n sobre la produccio´n de a´rboles adultos de cı´tricos en el suroeste espan˜ol. Bol. San Veg. Plagas 23, 73–91. Gonza´lez, L., Bernabe´, P., and Castan˜o, M. 1996. Enemigos naturales de Phyllocnistis citrella Stainton (Lepidoptera, Gracillariidae, Phyllocnistinae) en la provincia de Huelva. Distribucio´n geogra´fica, evolucio´n estacional y tasas de parasitismo. Bol. San. Veg. Plagas 22, 741–760. Granda, C., Costa J., Vercher, R., Olmeda, T., Alonso, A., Rodrı´guez, J. M., and Garcı´a, F. 1997. Dan˜os producidos por el minador de hojas de cı´tricos Phyllocnistis citrella Stainton. In “Libro de Resu´menes de las VI Jornadas Cientı´ficas de la SEEA.” Sociedad Espan˜ola de Entomologı´a Aplicada, Madrid. Granda, C., Garcı´a, F., Juan. M., Medina, F., Agustı´, M., Zaragoza, S., Costa J., Olmeda, T., and Marzal, C. 1999. Dan˜os producidos por el minador de hojas de cı´tricos Phyllocnistis citrella Stainton en plantaciones adultas de naranjo. In “Libro de Resu´menes de las VII Jornadas Cientı´ficas de la SEEA.” Sociedad Espan˜ola de Entomologı´a Aplicada, Madrid. Hawkins, B. A. 1988. Species diversity in the third and fourth trophic levels: Patterns and mechanisms. J. Anim. Ecol. 57, 137–162. Hawkins, B. A., and Lawton, J. H. 1987. Species richness for parasitoids of British phytophagous insects. Nature 326, 788 –790. Hawkins, B. A., and Mills, N. J. 1996. Variability in parasitoid community structure. J. Anim. Ecol. 65, 501–516. Hermoso de Mendoza, A., Fuertes, C., and Serra, J. 1986. Proporciones relativas y gra´ficas de vuelo de pulgones (Homoptera,

207

Aphidinea) en los cı´tricos espan˜oles. Inv. Agrar.: Prod. Prot. Veg. 1, 393– 408. Hespenheide, H. A. 1991. Bionomics of leaf-mining insects. Annu. Rev. Entomol. 36, 535–560. Hoy, M. A., and Nguyen, R. 1997. Classical biological control of the citrus leafminer Phyllocnistis citrella Stainton (Lepidoptera: Gracillariidae): Theory, practice, art and Science. Trop. Lepid. 8(Suppl. 1), 1–19. Jacas, J. A., Garrido, A., Margaix, C., Forner, J., Alcaide A., and Pina, J. A. 1997. Screening of different citrus rootstocks and citrusrelated species for resistance to Phyllocnistis citrella (Lepidoptera: Gracillariidae). Crop Prot. 16, 701–705. Johnson, M. W., and Hara, A. H. 1987. Influence of host crop on parasitoids (Hymenoptera) of Liriomyza spp. (Diptera: Agromyzidae). Environ. Entomol. 16, 339 –344. Lacasa, A., Martı´nez, A., Oncina M., and Sa´nchez, J. A. 1997. Enemigos naturales de Phyllocnistis citrella Stainton (Lepidoptera: Gracillariidae) y su incidencia en los cı´tricos de Murcia. In “Libro de Resu´menes de las VI Jornadas Cientı´ficas de la SEEA.” Sociedad Espan˜ola de Entomologı´a Aplicada, Madrid. LaSalle, J., and Pen˜a, J. E. 1997. A new species of Galeopsomyia (Hymenoptera: Eulophidae: tetrastichinae): A fortuitous parasitoid of the citrus leafminer, Phyllocnistis citrella (Lepidoptera: Gracillariidae). Fla. Entomol. 80, 461– 470. Lekchiri, A. 1996. La mineuse des feuilles des agrumes. Lutte biologique: tous les espoirs sont permis. Maroc Fruits, Mars, 9. Neuenschwander, P., Michelakis, S., and Kapatos, E. 1986. Thephritidae. In “Traite´ d’Entomologie Ole´icole” (Y. Arambourg Ed.), pp. 115–159. Conseil Ole´icole International, Madrid. Puiggro`s, J. M. 1998. Insecticide resistance in Spain from the point of view of the Industry. In “Pesticide Resistance in Horticultural Crops” (I. Cuadrado and E. Vin˜uela, Eds.), pp. 63–76. Fundacio´n para la Investigacio´n Agraria en la Provincia de Almerı´a, Almerı´a, Spain. Schauff, M. E., LaSalle, J., and Wijesekara, G. A. 1998. The genera of chalcid parasitoids (Hymenoptera: Chalcidoidea) of citrus leafminer Phyllocnitstis citrella Stainton (Lepidoptera: Gracillariidae). J. Nat. Hist. 32, 1001–1056. Siscaro, G., Barbagallo, S., Longo S., and Patti, I. 1997. Prime acquisizioni sul controllo biologico ed integrato della minatrice serpentina degli agrumi in Italia. Informatore Fitopatologico 7– 8, 19 –26. STSC. 1987. Statgraphics User’s Guide, Version 5.0, Graphic Software System. STSC Inc., Rockville, MD. Urbaneja, A., Jacas, J. A., Verdu´, M., and Garrido, A. 1998a. Dina´mica e impacto de los parasitoides auto´ctonos de Phyllocnistis citrella Stainton en la Comunidad Valenciana. Invest. Agr. Prod. Prot. Veg. 13, 787–796. Urbaneja, A., Lla´cer, E., Jacas, J.,and Garrido, A. 1998b. Ciclo biolo´gico de Cirrospilus pro´ximo a lyncus, parasitoide auto´ctono del minador de las hojas de los cı´tricos. Bol. San Veg. Plagas 24, 707–714. Urbaneja, A., Lla´cer, E., Garrido, A, and Jacas, J.. 1999. Effect of temperature on development of Cirrospilus sp. near lyncus. Environ. Entomol. 28, 339 –344. Vercher, R., Castrillo´n, D., Costa, J., Marzal, C., and Garcı´a, F. 1998. Parasitoides auto´ctonos del minador de las hojas de los cı´tricos, Phyllocnistis citrella en las comarcas centrales valencianas. Levante Agric. 333, 305–312. Verdu´, M. 1991. Chalcidoidea (Hym. Apocrita, Terebrantia) en plantas hortı´colas de la Comunidad Valenciana. Boln. Asoc. Esp. Ent. 15, 245–255.