Eucalyptus spp. (Mytales: Myrtaceae) originated in Australia and Indonesia, and they have outstanding importance in forestry for use as timber, pulp, and as a potential bioenergy feed-stock. Eucalyptus spp. have been cultivated and used around the world, and they bring economic, social, and environmental advantages because they reduce the deforestation pressure on native forests (Zheng et al. 2014).

The gall wasp, Leptocybe invasa Fisher & LaSalle (Hymenoptera: Eulophidae: Tetrastichinae), is of Australian origin and prevalent in many regions of the world (Jhala et al. 2010). Leptocybe invasa was recently introduced into Brazil, where it exhibits the behavior of endophytic oviposition, inducing the formation of galls on leaves and petioles of eucalyptus trees, with increased frequency of leaf distortion, drying, and defoliation of new branches (Rinaldi et al. 2013). The development of galls occurs due to a physiological disorder caused by an Intrinsic and specific relationship between plant and insect by high dependency and specificity to small changes in morphological and phenological characteristics of plants, determining degrees of susceptibility and coevolution (Stone & Schönrogge 2003; Zheng et al. 2014). This high degree of coevolution can provide advantages for the management of the wasp because it allows the separation of species and hybrids resistant to gall formation and development of L. invasa (Dungey et al. 2000).

There are great differences in the susceptibility of eucalyptus to wasp attack (Mendel et al. 2004). The objective of the present study was to evaluate the susceptibility of Eucalyptus spp. and clones to L. invasa in the field. The research was carried out in Umuarama City, Paraná (PR), Brazil (23°47′21.3″S, 53°15′29.2″W, 402.57 m asl) at the Universidade Estadual de Maringa, Campus of Umuarama. Leptocybe invasa was evaluated in a 2.0 acre (0.86 ha) field of 1-year-old Eucalyptus spp. clones with an average height of 2 m, spaced 3.0 × 2.0 m. To study the susceptibility and preference of the plant parts attacked by L. invasa, we used 6 clones (‘urograndis (seminal)’, ‘urograndis H13’, ‘I144’, ‘I224’, ‘Grancam COP 1277’, and ‘Urocam VM1’) and 2 eucalyptus species (E. grandis W. Hill ex Maiden and Corymbia citriodora [Hook.] K.D. Hill & L. A. S. Johnson [Myrtales: Myrtaceae]).

The experiment was a randomized block in split plot design with 5 replications. The preference of occurrence of L. invasa was evaluated in the main plots, and the preferred location for the occurrence on the plants was evaluated in the subplots. Accordingly, the trees were divided into 3 strata (upper, middle, and lower). Each plot had an area of 216 m², where 4 biweekly evaluations were performed from 05/10/2013. In each evaluation, 9 branches per tree (3 branches/stratum) were collected from 7 trees per plot.

The numbers of galls per leaf and petiole of each branch were recorded. Subsequently, the percentages of leaves and petioles with galls were determined according to the genetic material and each part of the plant collected. The susceptibilities of Eucalyptus spp. and clones to L. invasa were determined by symptoms observed in branches with oviposition of the genetic material collected, and the data were subjected to analysis of variance (F test). The mean numbers of insects per sample unit in each stratum of the tree as well as their occurrence in leaves and petioles were compared using Tukey's test at a significance level of 0.05. We used the statistical software Assistat 7.6 Beta, and the data were transformed with log(x + 5).

All the samples of ‘Grancam COP 1277’ showed galls on both leaves and petioles, with the mean values of the percentages in the first and last assessments being 19.4 and 25.0% and 8.3 and 10.5% for leaves and petioles, respectively (Fig. 1). Corymbia citriodora had the fewest number of galls, and the percentages of leaves and petioles of the plants attacked remained below 1% in all evaluations. According to Bailey et al. (2006) resistance or susceptibility of plants to pests is predominantly inherited and can be expressed both in pure species and in hybrids. The susceptibility of Eucalyptus spp. observed can explain the results obtained for preference for E. grandis and the hybrids ‘Grancam COP 1277’, ‘urograndis (seminal)’ and ‘Urocam VM1’, which are derived from crosses between the species E. grandis × E. urofila, E. grandis × E. camaldulensis, and E. camaldulensis × E. urofila, respectively.

The differences in the susceptibility of eucalyptus materials to gallforming insects may indicate that they possess genetic factors for both their attractiveness to L. invasa for oviposition and for their suitability for larval development of L. invasa. Another factor prevalent in most cases of susceptibility is the synchronization of the gall-forming species with the phenology of the host plant. Therefore, coevolution demonstrated between eucalyptus plants and L. invasa may provide greater ease in controlling this pest through the development and deployment of resistant and productive genetic materials than through chemical means (Dungey et al. 2000). In this context, the clones ‘urograndis H13’, ‘urograndis I144’ and ‘urograndis I224’ in all samples presented a lower percentage of leaves with galls than ‘grancan COP 1277’ (the most attacked), thus indicating the possibility that the 3 clones possess resistance to attack by L. invasa.

Fig. 1.

Percentage of leaves and stems with galls caused by Leptocybe invasa in approx. 2-m-tall Eucalyptus spp. and clones on each sampling date at Umuarama, PR, Brazil, in 2013.

Fig. 2.

Percentage of leaves and stems with galls caused by Leptocybe invasa in plants in each stratum (lower, middle, upper) of approx. 2-m-tall Eucalyptus spp. and clones at Umuarama, PR, Brazil, in 2013.

With regard to the occurrence of galls on leaves and stems in the various tree strata of the Eucalyptus spp. and clones, the averages were statistically different only for leaves in the second evaluation. Furthermore, the highest percentage was obtained for the upper stratum, with 9.3% of leaves with galls (Fig. 2). In the evaluation of the preferred sites of attack by L. invasion on the tree (leaves and petioles), we noticed that regardless of the species or clone evaluated or from which stratum the samples were collected, wasp attack occurred in the same proportion of leaves and petioles (Table 1). In most studies on oviposition preference of L. invasion in eucalyptus, this pest species occurred on leaves, shoots, and stems of the eucalyptus (Mendel et al. 2004; Garlet et al. 2013; Rinaldi et al. 2013).

Regardless of the sampling period and the tree stratum evaluated, the average numbers of galls per leaf and per petiole were significantly higher for ‘Grancam COP 1277’ than for any other eucalyptus materials. ‘Grancam COP 1277’ averaged 14.46 and 14.25 galls on the leaves and petioles, respectively (Table 1). The clones ‘urograndis H13’, ‘urograndis I144’, and ‘urograndis I224’ as well as C. citriodora had the lowest occurrence of galls, demonstrating low susceptibility to L. invasa.

The results obtained in the present study with eucalyptus confirm previously observed percentages of leaves attacked by L. invasa. The occurrence of galls observed are within the range of those reported previously by Mendel et al. (2004), who found that numbers of galls caused by L. invasa ranged from 1 to 65 per leaf. According to Rinaldi et al. (2013), the highest occurrence of galls per leaf and petiole had been observed for plants undergoing rapid growth, which was confirmed in the present study for the most-attacked hybrids.

Table 1.

Average values of galls transformed by log(x+5) (original averages in parentheses), caused by Leptocybe invasa in leaves and petioles of approx. 2-m-tall Eucalyptus spp. and clones, regardless of when collected and from which stratum. The data were collected at Umuarama, Paraná (PR), Brazil, in 2013.

We express heartfelt thanks to “Instituto Agronômico do Paraná (IAPAR)” and to Dr. Alex Carneiro Leal for the eucalyptus species and clones.