Greenidea psiidi van der Goot and Greenidea ficicola Takahashi (Hemiptera: Aphididae), are Asiatic species that feed on guava, Psidium guajava and Ficus spp.; both of these aphids were reported as exotic pests in Florida in 2002 and in Mexico in 2003. The present study characterized the population dynamics of both aphid species and their natural enemies on guava and ornamental figs in the Bajio region of Central Mexico. This report represents the first record of G. psiidi on Ficus sp. in Mexico and the first report of the presence of both species in the state of Guanajuato. Greenidea psiidi and G. ficicola were detected on guava in Mar 2007 and on fig trees during the same year in Apr near Irapauto, Guanajuato. Populations of both alate and apterous forms of G. psiidi in Apr were greater on guava than on fig trees (W = 119.0; P = 0.0122), which coincided with new vegetative growth after leaf loss in winter on guava. In Apr populations of apterous forms of both species were significantly greater than winged forms on both guava and figs. No correlation was found between temperature changes and population densities of aphids. The indigenous predators, Chrysoperla comanche Banks, Chrysoperla exotera (Navás) and Chrysoperla carnea (Stephens) (Neuroptera: Chrysopidae), fed readily on the aphids and were found on both guava and fig trees, although densities of all 3 species were in greater numbers on Ficus. The combined population densities of the 3 predators had a positive correlation with that of G. ficicola (r = 0.74), with a best fit found with a quadratic model of simple regression: y (densities of Chrysoperla spp.) = 1.2479x2 4.3073x 9.6493, and R2= 0.703. Nine species of coccinelid beetles (Coleoptera: Coccinellidae) were identified, the most common being of the genus Scymnus. Results suggest that non-deciduous ornamental fig trees may serve as reservoirs of beneficial insects for deciduous guava trees. Results from the present study provide basic biological data to aid in management of these 2 exotic species of Greenidea on guava in central Mexico.
Mexico contributes 25% of the world production of guava (Psidium guajava L.; Myrtaceae), of which the states of Michoacan, Aguascalientes, and Zacatecas of the central Altiplano region are the principal producers (González-Gaona et al. 2002). In 2008, the exotic aphid pests, Greenidea psidii van der Goot and Greenidea ficicola Takahashi were observed on guava and ornamental fig trees, Ficus benjamina (L.), in the central Altiplano state of Guanajuato (Salas-Araiza, unpublished data). There is concern among producers and agricultural researchers of central Mexico that these new pests may pose a threat to guava production in Mexico.
The genus Greenidea Schouteden belongs to the subfamily Greenideinae within the Aphididae and includes approximately 45 species (Pérez Hidalgo et al. 2009). The natural distribution of the genus is Asiatic and species are found to favor young foliage of plants of the families, Fagaceae, Betulaceae, Juglandaceae, Myrtaceae, Rosaceae, and Rubiaceae (Blackman & Eastop 1994). The presence of Greenidea in the New World may be the result of importation of infested ornamental fig trees which are widely commercialized throughout the world, and they have been implicated as vehicles for the introduction of exotic pests (O'Donnell & Parrella 2005).
Greenidea psidii was first discovered in the New World in 1916 on guava and its relative, Psidium cattleianum Sabine in Brazil (NoembergLazzari et al. 2006). Halbert (2004) reported that G. ficicola was collected in Florida in 2002, and together with G. psidii were found to feed on guava and ornamental fig trees (Ficus benjamina (L.)). Pérez Hidalgo et al. 2009 reported the presence of G. psidii in Costa Rica in 2009. The first reports of Greenidea in Mexico come from Peña-Martínez et al. (2003), who recorded G. psidii feeding on guava in the states of Hidalgo, Morelos, Guerrero, and the Federal District, and G. ficicola feeding on ornamental fig trees in the state of Guanajuato, Mexico. Trejo-Loyo et al. (2004) reported G. ficicola occurs on Myrtaceae in Cuernavaca, Morelos. Although further data is lacking, it is probable that both G. psidii and G. ficicola are widely distributed in Mexico.
The biology of species of Greenidea on guava and other host plants is poorly known (Halbert 2004; Sousa-Silva et al. 2005; Noemberg-Lazzari et al. 2006). Northfield et al. (2008) noted that many insect pests feed on alternative host plants and that the understanding of the relationships of pest populations on wild hosts with those on cultivated hosts is crucial in the development of management strategies. The objective of the present study was to determine the population dynamics of Greenidea psidii, G. ficicola and their natural enemies on Psidium guajava and Ficus benjamina in the state of Guanajuato, Mexico. The data generated from this study will help evaluate the importance of these aphids within the pest complex attacking guava in the Bajío Region, as well as aid in the development of integrated pest management strategies.
MATERIALS AND METHODS
The study site was conducted at the experimental field station of the Division of Life Sciences of the University of Guanajuato at the Ex-Hacienda El Copal (101°01′01″N, 20°49′49″O) at 1,750 masl in the municipality of Irapuato, Guanajuato, Mexico. The region has a mean annual precipitation of 750 mm, a mean temperature of 19°C and mean relative humidity of 56% (INEGI 2009). Two orchards, 1 guava (P. guajava), and the other ornamental figs, (F. benjamina) were chosen for the study sites.
All trees were 10 years old and the 2 orchards were separated by approximately 1 km. Weekly samples were made of aphids (alate and apterous forms) and predators (adults of Coccinellidae and larvae of Chrysopidae) from 10 trees of each host species. For each tree, samples consisted of 20 beats of a 1-m wooden rod on the branches of the trees, from which insects fell onto a 1-m2 beating sheet. All insects on the sheet were collected and placed in a labeled vial with 70% alcohol. The sample period was from 23 Mar to 22 Jun 2007 which corresponded to the early spring growth period of leaves of guava and to the reproductive activity of aphids. Samples were not taken on subsequent dates because aphids had ceased reproductive growth and individuals were virtually undetectable. All material from each tree was preserved in individual vials containing 70% alcohol, with a corresponding label and brought to the laboratory.
The collected specimens were identified in the Entomology Laboratory of University of Guanajuato, with a compound and stereo microscope. Aphids were mounted for species determination following techniques given by Peña-Martínez (1995). The keys of Blackman & Eastop (1994, 2000) were used for species identification of aphids. For the neuropteran and coccinellid predators, the keys of López-Arroyo et al. (2008) and Gordon (1985) were used, respectively. Meteorological data was obtained from a weather station maintained at the study site located 200 m from the orchards. All identified specimens were deposited in the Entomological Collection “Leopoldo Tinoco Corona” of the University of Guanajuato.
Data were analyzed with the statistical software program SAS (SAS 1995). Abundance measures were calculated for G. psiidi and G. ficicola and the various predator species on guava and figs. Due to non-linearity of data determined by the Shapiro-Wilks test, mean comparisons of aphid and predator numbers between hosts and among samples dates were made with MannWhitney non-parametric procedures. In addition, correlation analyses were conducted between population numbers of aphids with those of the various predators and also with climatic variables (temperature and precipitation).
RESULTS AND DISCUSSION
Species of Greenidea in Guava and Figs
Both G. psidii and G. ficicola were present on guava and the ornamental fig, F. benjamina (Figs. 1, A, B, C). This is the first report of G. psiidi on the widely planted F. benjamina from Mexico, and for this species for the state of Guanajuato. Although the habitual host for G. ficicola is Ficus spp. (Noemberg-Lazzari et al. 2006), Halbert (2004) reports that this aphid species also occurs on guava, as confirmed in the present study. Although the apparent habitual hosts of G. psidii is guava, and that of G. ficicola is Ficus spp., it is unclear how populations on the habitual and other infrequent host plants interact and which of the infrequent hosts can maintain viable, reproductive populations in the absence of the habitual host plants.
Both species of Greenidea were aggregated on new shoots and leaves of their hosts, a feeding preference previously reported by Pérez Hidalgo et al. (2009). On guava, G. psiidi and G. ficicola were present on new leaf buds and on either side of young leaves, whereas on F. benjamina they were found principally on the underside of young leaves. The preference by aphids for new plant growth is a common behavior in aphids. Gould et al. (2007) state that certain stages of aphids have preferences for specific tissues of host plants. For example, Chaitophorus populicola Thomas, preferably feeds on new growth with diverse and high levels of amino acids. These authors also note that high sugar levels and low amino acid concentration in leaves increases the production of winged individuals.
Population Dynamics of Greenidea psidii and Greenidea ficicola
Both G. psiidi and G. ficicola were first recorded on guava trees in late Mar 2007, although the abundance of G. psiidi was notably greater during Mar and early Apr. This appearance and population growth coincided with the emergence of new shoots and leaves on guava, the timing of which corresponds to that described by DamiánNava et al. (2004). Colonizing alate aphids are often attracted to specific volatiles of host plants (Chapman et al. 1981; Nottingham et al. 1991; Powell & Hardie 2001). Because G. psiidi and G. ficicola are found initially on young shoots and leaves, it is probable that initial alate colonizers are attracted to volatiles associated with new growth of guava plants.
It important to note that the 2 tree species studied have marked differences in leafing patterns and the appearance of new growth. Guava is deciduous, with complete leaf loss occurring generally in Nov with new growth beginning in Mar in the study area. This is in contrast to F. benjamina which is a non-deciduous tree that produces new growth apparently in response to environmental factors. However, both G. psiidi and G. ficicola appeared first on guava, and then on F. benjamina (Fig. 1, A, B, C), although foliage was available on the latter host throughout winter months. These data suggest that G. psiidi and G. ficicola first colonize and establish on guava and then later move to F. benjamina. This behavior was expected for G. psiidi for which guava is considered a habitual host, but not for G. ficicola which Ficus is considered the habitual host. Further study is needed to establish the initial colonization behavior of these 2 species and whether the behaviors are the result of greater attractiveness of guava in comparison to F. benjamina and/ or the leaf quality of F. benjamina is inadequate until late Apr.
The peak abundance of both species (alate and apterous forms) occurred in mid and late Apr (7.5 and 6.0 aphids/sample, for G. psiidi and G. ficicola, respectively) (Fig. 1, A, B). The densities of G. psiidi were greater on guava than on F. benjamina during the first 5 sample periods, and significantly so at peak densities during the third week of Apr (w = 119.0; P = 0.0122; Fig. 1, A).
By mid May, densities of G. psiidi and G. ficicola were barely detectable, and remained at very low densities through late Jun. These low densities are presumably the result of the maturation of leaves and the deterioration of the physical and nutritional requirements for these species (Fig. 2, A).
The appearance and abundance of alate forms in relation to apterous forms of G. psiidi and G. ficicola followed patterns expected from observations reported by Noemberg-Lazzari et al. (2006). Alates were found in very low numbers initially on both guava and F. benjamina and differences between the 2 forms were not significantly different until the third week of Apr (Fig. 2, B, C). At that time, the abundance of apterous forms increased considerably (16–18 fold) for both species. Significant differences between the abundance of alate and apterous forms of G. psiidi on guava and F. benjamina were found from mid Apr through mid May (Fig. 2, B). Alate forms were also most abundant during this period, and continued to be found into Jun. It is assumed that many of these alate individuals dispersed from guava and F. benjamina when leaf and shoot quality declined after Apr. No correlation was found between the mean ambient temperature nor mean precipitation with the population changes of G. psiidi and G. ficicola on guava and F. benjamina (r = 0.15 for temperature; r = -0.21 for precipitation for G. psidii; r =0.16 for temperature; and r = -0.18 for precipitation for G. ficicola), This was expected given that there was little change in either temperature or precipitation during the sample period (Fig. 3). Even with greater temperature variation, however, little correlation would be expected given that temperature is often not a principal factor in aphid population growth. For example, Desneux et al. (2006) reported that temperature was not positively correlated with population densities of Aphis glycines Matsumura on soybeans, and some species, such as Schizaphis graminum (Rondan) on wheat produce more nymphs at lower temperatures (Pendleton et al. 2009). Differences in precipitation also were probably not an important factor in the changes in populations that were observed because the only noteworthy change in precipitation occurred between the last two samples. Although precipitation has been observed to physically dislodge aphids from plants and increase infection by pathogens (Nielson & Barnes 1961), in our study precipitation occurred primarily in Jun, and thus well after populations had already declined in late May It could be argued that the life cycle of G. psiidi and G. ficicola is adapted to avoid the unfavorable effects of the seasonal rains by completing their life cycle during the dry season. However, Noemberg-Lazzari et al. (2006) reported that in Brazil both species are present year round (G. psiidi on guava and G. ficicola on Ficus sp.), suggesting that the life cycle of the 2 insects is facultative and probably dependant more on host quality than on specific seasonal variations in climate.
Chrysopidae. Three species of Chrysopidae (Neuroptera) were identified on guava and F. benjamina: (1) Chrysoperla comanche (Banks), (2) Chrysoperla exotera (Návas), and (3) Chrysoperla carnea (Stephens). The most abundant species was C. comanche and the present work represents the first report of this species for the state of Guanajuato. This chrysopid is cited by Ramírez-Delgado et al. (2006) as the predominant species in pecans and grapes in the states of Coahuila and Durango, suggesting that it is one of the most important aphid predators of the central highlands of northern Mexico.
The combined populations of the 3 species of Chrysoperla were present from 20 Apr 2007 to 15 Jun 2007 in guava and ficus. Chrysoperla was first recorded on F. benjamina on 20 Apr of 2007 (Fig. 4, A). In both guava and F. benjamina, an increase of individuals of Chrysoperla spp. coincided with an apparent increase in the populations of the apterous forms for both aphid species on 20 Apr 2007 (Fig. 1, A, B, C). Populations of the combined densities of Chrysoperla spp. peaked first on F. benjamina on 27 Apr 2007 (0.32 individuals/sample) and later on guava on 11 May 2007 (0.15 individuals/sample). Densities of these predators gradually decreased from mid-May to mid Jun.
Combined populations of the 3 species of Chrysoperla were positively correlated with populations of G. ficicola (r = 0.74) on guava and F. benjamina and the relationship defined by y (densities of Chrysoperla spp.) = 1.2479x2 - 4.3073x + 9.6493, grade 2 polynomial with R2 = 0.703 (Fig. 5). This suggests that there was a numerical response in the populations of Chrysoperla species in conjunction with density changes in G. ficicola. There was no apparent correlation found between populations of G. psiidi with Chrysoperla on either guava or F. benjamina.
Coccinellidae. Species of Coccinellidae (Coleoptera) on guava and F. benjamina were: Hippodamia convergens Guérin-Méneville, Harmonia axyridis Pallas, Zoglobra spp., Coccinella spp., Olla v-nigrum (Mulsant), Cycloneda sanguinea L., Stethorus spp., Hyperaspis quadrioculata (Motschulsky), and Scymnus spp. On F. benjamina, Scymnus spp. were the most abundant with 167 total individuals found compared with 126 captured in guava. Individuals of Scymnus spp. first appeared in F. benjamina on 20 Apr 2007 and began increasing by 11 May 2007, reaching a maximum abundance on 1 Jun 2007 of approximately 6 individuals/sample and maintained densities of 2.5–4 individuals/sample until the end of the sample period. The overall population densities of Scymnus spp. was higher on F. benjamina than on guava (11.9 vs 9 individuals/sample, respectively), and on 5 of the 10 sample dates, and these predators had significantly higher densities on F. benjamina than on guava (Fig. 4, B). Results suggest that other prey present on F. benjamina were more important in maintaining Scymnus spp. than the Greenidea species because these predators appeared after major populations peaks of Greenidea, and increased after mid-May in the virtual absence of these aphids.
Aphid mummies were not observed in the present study and there are no reports of parasitism of Greenidea suggesting that predators are the principal insect biological control agents of Greenideae in the region. Although F. benjamina was a host plant of Greenidea, both aphid species were found first on guava, and then reached higher densities on guava than on F. benjamina. However, following aphid colonization, chrysopid and coccinelid predators developed greater densities on F. benjamina. These data suggest that F. benjamina is more important as a refuge for natural enemies of the pest aphids in the region, than as an alternative host plant from which Greenidea colonizes guava. The movement of predators from one host to another is well documented, in particular for aphidophagous generalists such as chrysopids and coccinellids (Sloggett et al. 2008).
The following study lays the groundwork for the management of Greenidea psidii and G. fisicola in guava in the region. Indigenous predators were found to readily feed on G. psiidi and G. ficicola and are apparently an important component in reducing densities of the pests. Further work is needed to study in detail the colonization behavior of G. psiidi and G. ficicola in relation to guava and other hosts, and to investigate methods to optimize the action of native predators on these aphids.