Leptocybe invasa
(blue gum chalcid)

Leptocybe invasa is believed to be native to Australia or to the native range of its host plants Eucalyptus - Australia, New Guinea, Indonesia and Philippines -, although it has only been detected in Australia in Queensland and New South Wales. During the last two decades, L. invasa has spread worldwide, invading all the continents where Eucalyptus has been imported to (Asia, Africa, Europe, South and North America). The invasive potential of L. invasa is very high. Its broad host range, polyvoltinism, overlapping generations, concealed life style (galls) and reproductive modalities allow the pest to quickly spread and exponentially grow from few individuals. L. invasa is considered thelytokous because sex ratio of the most widespread lineage is female biased, but a biparental lineage also exists.

L. invasa is believed to be native either only to Australia or to the whole native range of its host plants Eucalyptus, i.e. Australia, New Guinea, Indonesia and Philippines (Hill and Johnson, 2000; Nugnes et al., 2015). The species is closely related to its host plants and, in principle, could be found everywhere Eucalyptus is grown. At the moment, it seems to have spread from its native region into Africa, Asia, Europe, South and North America (Nugnes et al., 2015).

Globalisation and increasing trade promote the spread of the blue gum chalcid via its main pathways of introduction, including trade of nursery plants with leaves and shoots. Difficult detection of both adults and early stages of the galling process does not facilitate early interception of the pest, which can hence be introduced very easily. Furthermore, the species small dimensions can facilitate its local dispersal both by unintentional carriage and via wind currents. Strict quarantine measures can delay species spread, but almost certainly cannot stop invasion in all countries where eucalypts are grown.

In the invaded range, L. invasa is distributed in agricultural, forest and urban areas, and in rangelands where eucalypts are grown. Hence, it can be found in nurseries and in eucalyptus plantations growing as planted forest stands, as ornamental trees, as windbreaks around orchards or in groves growing under irrigation (Mendel et al., 2004; Nyeko, 2005; Protasov et al., 2008; Thu et al., 2009; Karunaratne et al., 2010; Pereira et al., 2014).

The majority of Eucalyptus species have been confirmed to be susceptible to L. invasa (Jorge et al., 2016). Among these, E. camaldulensis (var. camaldulensis and obtusa), E. grandis, E. robusta and E. tereticornis showed remarkably high susceptibility to the blue gum chalcid (Mendel et al., 2004; Thu et al., 2009; Nyeko et al., 2010). Studies carried out on several clonal hybrids or clones of these species showed that the incidence of L. invasa infestation could affect them differently (Nyeko et al., 2010; ICFR, 2011). E. gomphocephala and E. occidentalis did not show susceptibility to L. invasa, while E. erythrocorys exhibited “cork tissue” symptoms some days after deposition, but no further gall formation was observed (Mendel et al., 2004).

To date, the only susceptible species not belonging to the Eucalyptus genus is Corymbia polycarpa (Thu et al., 2009), with other species of this genus (C. citriodora, C. maculata, C. torelliana) apparently being tolerant (Mendel et al., 2004). 

Detailed information on the biology of L. invasa in different environments and hosts is available in many different publications (Mendel et al., 2004; Kavitha Kumari et al., 2010; Sangtongpraow et al., 2011; Eyidozehi et al., 2014; Udagedara and Karunaratne, 2014; Zhu et al., 2015).

Genetics

Few studies have been carried out on the genetic characterisation of L. invasa. Nugnes et al. (2015) found that L. invasa is represented by at least two genetically and biologically distinct lineages: a Chinese lineage present in East Asia and characterized by biparental populations, and a Western lineage (in Italy, Tunisia, Turkey and Argentina) that reproduces by thelytokous parthenogenesis. Each lineage was found to be infested by a respective endosymbiotic Rickettsia which, at least for the Western lineage, is considered responsible for the thelytoky (Nugnes et al., 2015).

Reproductive Biology

Studies on potential fecundity of L. invasa highlight that an increase in size can increase potential fecundity. Indeed, ovaries of small females contained an average of 90 eggs (range of 39-140), in contrast with large females that reached an average of 210 eggs (range of 141-298). As such, potential fecundity of the blue gum chalcid is, in average, 160 eggs/female (Sangtongpraow et al., 2011). Number of eggs is correlated with environmental temperature, with an increase in number of eggs per female being directly proportional to an increase in temperature from 20 to 29°C. Number of eggs/female was the highest at 29°C, with a sharp decrease at 32°C (Zhu et al., 2015).

Generally, realized fecundity is determined taking into account the number of eggs laid on/in a host over the life time of a female. L. invasa females lay eggs in lined groups, at a distance of about 0.3-0.5 mm of each other, in the midribs or petioles of young leaves of eucalyptus (about 0.5-5 cm in length). Due to this oviposition behaviour, it is difficult to detect the real number of eggs laid. Hence, to determine L. invasa’s realized fecundity, the number of emerging progenies over the life time of a female was taken into account (Sangtongpraow et al., 2011).

Dissections of newly emerging females of the blue gum chalcid showed the presence of mature eggs in ovaries, indicating L. invasa to be a pro-ovigenic species (Sangtongpraow et al., 2011; Nugnes et al., 2015). Soon after their emergence females start to lay eggs in host tissues, with a trend for these numbers to decrease with time. Indeed, during the first day, a female could lay about 30 eggs, whereas this number gradually declines, reaching 0.2 eggs/female on the sixth day (Sangtongpraow et al., 2011).

Physiology and Phenology

L. invasa develops two or three overlapping generations per year in most countries where it occurs, depending on the ecological conditions (Mendel et al., 2004), although in Guangxi, China, the species can complete five or six overlapping generations per year (Wu et al., 2009).

Mean development time from eggs to adults has been recorded to last about 138 days in the field (Hesami et al., 2005). In greenhouse, average development time varies greatly according to different studies: 132 days (Mendel et al., 2004), 126 days (Hesami et al., 2005), 60 days (Kavitha Kumari et al., 2010) or 46 days (Sangtongpraow et al., 2011). Duration of each life stage has been estimated as about 11 days for the egg stage, 19 and 30 days for the young and mature larval stages, respectively, and 38 days for the pupal stage. However, temperature affects development time of L. invasa, with development time decreasing linearly as temperature increases (Zhu et al., 2015). The temperature that allowed the shortest development time from egg to adult was 29°C, likely the species optimum developmental temperature. Temperature is also positively correlated to survival rate of the immature stage, with the highest percentage survival at 29°C, and a sudden decrease in survival at 32°C (Zhu et al., 2015). Tests carried out in 2011 in Guangdong, China, reported very low thresholds (0°C) for the development of several stages of the blue gum chalcid (egg, larval, pupal and adult stages). Furthermore, different requirements were estimated for cumulative temperatures (degree-days) of each life stage: 146.6 (eggs), 1228 (larvae), 161.7 (pupae) and 205.2 (adults) (Qiu et al., 2011). The developmental zero temperature (DZT) for the complete life cycle (egg-adult) of L. invasa in Dongfang (Hainan Island, China) was estimated to be about 19°C, while DZTs of egg, larval and pupal stages were 13, 19.7 and 17 °C, respectively. The same study calculated the cumulative temperatures needed for each life stage as 128°C for eggs, 284°C for larvae and 201°C for pupae, totalising a cumulative temperature of 563°C needed for the development period from egg to adult (Zhu et al., 2015).

Longevity

Adult longevity tests have been performed in climatic chamber (at 25°C and 80% relative humidity) under different feed treatments (Mendel et al., 2004; Sangtongpraow et al., 2011). Adult females of L. invasa fed with honey solution (honey:water, 1:1) showed the longest mean life span (6-8 days). Adults lived for 2-5 days if fed with fresh flowers of the host plant (E. camaldulensis), from 1.5 to 3.7 days if fed only with distilled water and 3.5 days if only young fresh foliage of the host plant was available. A diet made up of a combination of fresh foliage and honey solution did not extend life span over the 6 days observed for the “honey solution diet”. Adults with no food and no water lived for about 1.33 days. Male longevity was shorter than for females, with males exposed to the same food sources surviving for 5-6 days (honey solution), 1.22 days (flowers), 1.11 days (water), 1.56 days (water+flowers), 5.22 days (honey solution+flowers) and 1 day (no food and no water) (Sangtongpraow et al., 2011).

Temperature also influences adult longevity, with female longevity being the highest between 23 and 29°C (about 8-9 days), and just 5.8 days at 32°C. Male longevity reached the highest values of 12-13 days at 20 and 23°C and a proportional decline was observed with an increase in temperature, reaching 6 days at 32°C (Zhu et al., 2015).

Activity Patterns

In the field, L. invasa adults are active throughout the day, although their activity increases in sunny hours.

Population Size and Structure

Gall density can vary with tree age: in nursery seedlings and coppice, density was 36.99 galls/10 cm shoot, while on grown-up trees the average number of galls for a shoot of 10 cm only reached about 16 galls (Kavitha Kumari et al., 2010; Kulkarni, 2010). Furthermore, density of galls in different tissues is correlated to the Eucalyptus species host. Studies revealed that E. urophylla x E. camaldulensis hybrids showed a higher density of galls on midribs and twigs, while E. exserta had more galls on petioles (Zhu et al., 2012).

L. invasa populations exhibit different sex ratios in different areas. Males have never been recorded in Europe, Africa, Laos, Malaysia, Sri Lanka, Vietnam, Florida (USA), Argentina, Chile, Mexico, Paraguay and Australia (Nugnes et al., 2015), which might indicate that, in these countries, the species reproduces by thelytoky. The role of symbionts as being responsible for thelytoky has been supported by several facts: complete female-biased sex ratio, high-density localisation within the ovaries, and vertical transmission of a Rickettsia endosymbiont to the offspring (Nugnes et al., 2015). The presence of males has been recorded in India (Karnataka), Turkey, Thailand, Brazil and China (Doganlar, 2005; Chen et al., 2009; Gupta and Poorani, 2009; Kavitha Kumari et al., 2010; Liang et al., 2010; Tung and La Salle, 2010; Zheng et al., 2014; Sangtongpraow et al., 2011; De Souza, 2016). Particularly in China, male proportion ranged from 18 to 48% in some populations (Liang et al., 2010) although, in others, males only occurred occasionally (Zhu et al., 2015). Genetic characterisation of both the insect and the endosymbiotic Rickettsia corresponds to the different geographic sex ratios recorded in populations from China and Mediterranean and South American regions (Nugnes et al., 2015). Furthermore, the presence of males does not always correspond to sexual reproduction, with unmated females from Chinese and Thai populations having offspring of both sexes (Sangtongpraow et al., 2011).

Associations

L. invasa populations have been associated with the endosymbiotic bacteria Rickettsia (Nugnes et al., 2015). In samples from the Mediterranean region and South America, Rickettsia causes parthenogenetic reproduction, resulting in a female biased sex ratio. The association Leptocybe-Rickettsia improves the biotic potential of the infected populations because the whole progeny is feminine, hence potentially doubling the intrinsic rate of population increase.

Environmental Requirements

L. invasa has substantial environmental plasticity, being able to live where the host trees are present, under different climatic conditions. However, some studies have demonstrated a negative correlation between altitude and L. invasa attack to Eucalyptus spp. (Nyeko et al., 2009), with no attack having been observed on trees in plantations at altitudes higher than 1938 m above sea level.

L. invasa shows tolerance to low temperatures, with pupae having higher tolerance than larvae and adults. Indeed, the supercooling points of larvae, pupae and adults were about -25.5, -25 and -21°C respectively, while the freezing points of the same stages were about -24, -23 and -17.5°C, respectively. Male adults showed more resistance than females to lower temperatures (Han et al., 2012).

In the native range, natural enemies play a very important role in limiting populations of L. invasa, such that the species does not threaten either wild Eucalyptus or plantations. Surveys carried out in the native area have highlighted the potential of the species Quadrastichus mendeli, Selitrichodes kryceri and S. neseri (Eulophidae, Tetrastichinae) (Kim et al., 2008; Kelly et al., 2012).
In areas invaded by L. invasa, many studies have been carried out looking for natural enemies of the species. Native Megastigmus spp. (Torymidae) have been reared in South America (Argentina and Brazil), the Mediterranean area (Israel, Italy and Turkey), Africa (South Africa) and Asia (India, Sri Lanka and Thailand) (Viggiani et al., 2002; Protasov et al., 2008; Vastrad et al., 2009; Doganlar and Hassan, 2010; Zaché et al., 2012; Doganlar et al., 2013; Sangtongpraow and Charernsom, 2013; Udagedara and Karunaratne, 2014; Doganlar, 2015; Nugnes et al., 2016; Zheng et al., 2016). Furthermore, Aprostocetus sp. and Aprostocetus causalis (Eulophidae, Tetrastichinae), Parallelaptera sp. [Erythmelus sp.] (Mymaridae) and Telenomus sp. (Platygasteridae) have also been identified in India (Vastrad et al., 2009).

The genus Leptocybe (where L. invasa is the unique species) did not exist until 2004, which caused some authors to misidenty L. invasa as Aprostocetus sp. based on the Tetrastichinae key (Viggiani et al., 2002). L. invasa is distinguished from the other Tetrastichinae by the presence of a weakened area on the vertex behind the ocelli. It can be distinguished from Aprostocetus by several characters that are in common with Baryscapus. Characters useful in distinguishing L. invasa from Baryscapus and Oncastichus (as reported in Mendel et al., 2004) are the following:

- L. invasa vs Baryscapus: propodeum with a raised lobe of the callus that partially overhangs the outer rim of the spiracle; spiracular depression open to anterior margin of propodeum; mesoscutum with a single row of 2-3 adnotaular short, weak setae.

- L. invasa vs Oncastichus: postmarginal vein less than 0.25 the length of the stigmal vein; scape not reaching upper margin of vertex.

First and second stages of L. invasa galls could be confused with early stages of Ophelimus maskelli galls, but while L. invasa makes bump-shaped galls on petioles, leaf midribs and young branches (Zheng et al., 2014), O. maskelli induces blister-like galls on the leaf’s surface (Burks et al., 2015). Similar blister-like galls are also induced by Epichrysocharis burwelli, but this species current distribution range includes the USA, Brazil and Portugal and its main host is Corymbia citriodora (Schauff and Garrison, 2000; Franco et al., 2016; Schnell e Schühli et al., 2016). Selitrichodes globulus also causes multilocular galls that hang from branches and twigs of the host (La Salle et al., 2009).

Due to the variable regulations around (de)registration of pesticides, your national list of registered pesticides or relevant authority should be consulted to determine which products are legally allowed for use in your country when considering chemical control. Pesticides should always be used in a lawful manner, consistent with the product's label.

Prevention

Because the blue gum chalcid has already spread to all continents, it is too late to implement preventive measures that can effectively avoid its introduction to new large areas. Nevertheless, introduction to new countries can perhaps be prevented by using detection protocols that provide accurate examination of leaves, petioles and twigs when searching for galls and by adopting quarantine measures that allow development of visible stages of symptoms and consequent detection of the pest. In addition, avoiding movement of infested plants or seedlings could prevent further spread from a plantation to another, especially in areas where L. invasa is not totally widespread.

Control

Physical/mechanical control

Physical control has been suggested during the early stage attack of the blue gum chalcid only. Indeed, removing and destroying infested material could decrease L. invasa populations. However, this method is not efficient for late and/or high infestations, because it is arduous, or even impossible, to remove all infested material. Furthermore, due to the overlapping generations of L. invasa in the field, removal of the attached material would be practically incessant (ICFR, 2011).

Biological control

Biological control seems to be the most promising solution for L. invasa’s control. Different species of parasitoids have been found and studied since the first finding of the blue gum chalcid.

Quadrastichus mendeli - The best performing parasitoid against L. invasa which, from 2007, has been released in Israel, Kenya and India as part of classical biological control programs (Kim et al., 2008; Shylesha, 2008; K.E. Mutitu, pers. comm.). In Israel, percentage of parasitism of L. invasa by Q. mendeli reached 73%, and its spread and establishment were confirmed by its retrieval in all sampled sites (Kim et al., 2008). In contrast, Q. mendeli did not express any control action against L. invasa in Kenya (Dittrich-Schröder et al., 2012) while, from preliminary studies, a reduction in gall infestation in nurseries has recently emerged in India (Jacob et al., 2015). In Italy, Q. mendeli was never officially released but, from 2013, specimens were collected from galls of L. invasa on Eucalyptus camaldulensis and subsequent surveys showed its spread in Central and Southern Italy (including in Sicily). In Italy, the highest mean percentage of parasitisation reached 50.5 ± 6.16%, while the lowest percentage was 30.2 ± 8.10%. However, its effective control action against the blue gum chalcid was shown both by the 100% parasitisation percentage reached in the Southeast coast and the complete disappearance of L. invasa galls from Portici (Southwest Italy) soon after the first record of the parasitoid (Nugnes et al., 2016).

Selitrichodes neseri - This eulophid wasp was originally collected in Queensland (Australia) in 2010 (Kelly et al., 2012). It has been reared in quarantine in South Africa, where its nominal parasitism rate ranged from 9.7 to 71.8% and where it was shown to be specific for L. invasa, when tested on 17 possible non-target hosts from South Africa (Dittrich-Schröder et al., 2014). In 2012, it was released in fields in South Africa and some adults emerged in subsequent sampling sessions (Hurley, 2012). Preliminary laboratory and field studies about the efficiency of S. neseri were carried out in Brazil, showing a higher index of population growth in the fields than in laboratory (De Souza, 2016).

Selitrichodes kryceri - This biparental parasitoid was introduced to Israel as part of a biological control program and its percentage of parasitism of L. invasa reached 52% (Kim et al., 2008). It has been released in India (Shylesha, 2008) and South Africa too (ICFR, 2011). Detailed descriptive and biological features of the species are available in Kim et al. (2008).

Aprostocetus causalis - Firstly identified as A. gala in India (Kavitha Kumari, 2009; Vastrad et al., 2009), but a subsequent identification clarified its species status (Yang et al., 2014). Its control activity was tested in 2010 in greenhouses, confirming its effectiveness (Vastrad and Ramanagouda, 2014).

Megastigmus dharwadicus - Although it was found in a survey in India in 2008 as the most dominant parasitoid on L. invasa galls (90.7%) (Vastrad et al., 2009), it was described for the first time in 2010 (Narendran et al., 2010).

Megastigmus thitipornae - Although its life cycle is shorter than that of the blue gum chalcid, a preliminary study carried out in Thailand demonstrated its scarce parasitism capacity due to its low potential and realized fecundities, and its male-biased sex ratio (Sangtongpraow and Charernsom, 2013).

Megastigmus leptocybus - Reared from galls of L. invasa on E. camaldulensis from Israel, Turkey and Italy (Doganlar and Hassan, 2010; Doganlar, 2015).

Megastigmus zebrinus - Reared from 2010 from galls of L. invasa in South Africa (Kelly et al., 2012).

Megastigmus zvimendeli - Found in 2010 in Queensland (Australia) and then introduced to Israel and Turkey (Doganlar, 2015).

Megastigmus viggianii - Reported for the first time in India on leaf galls of the blue gum chalcid (Gupta and Poorani, 2008), with a percentage of parasitisation almost reaching 25% in China (Zheng et al., 2016).

Megastigmus brasiliensis - Its behaviour is considered parasitic against the blue gum chalcid because of the findings of adults in stems and leaves of eucalypts infested by L. invasa (Doganlar et al., 2013).

Parallelaptera sp. [Erythmelus sp.] (Mymaridae) and Telenomus sp. (Platygastridae) - Found once in India, but their role in the control of the blue gum chalcid is not clarified yet (Vastrad et al., 2009). Based on species from the same families, these two parasitoids should be oophagous, and gall wasps are usually not included among the hosts of Telenomus spp. (Viggiani, 1997), while Erythmelus spp. are usually parasitoids of Hemiptera eggs (Triapitsyn et al., 2007).

Chemical control

Chemical control is probably not feasible both in large plantations and natural environments, but may be possible in nurseries and seedlings (EPPO, 2006). However, because L. invasa completes its life cycle inside a gall, chemical control could be ineffective (Maged, 2016) or effective only with systemic insecticides. The use of systemic insecticides has been tested in different studies and results are very variable depending on eucalypt species and environments (Basavana Goud et al., 2010; Javaregowda et al., 2010; Jhala et al., 2010; Kulkarni, 2010; Nyeko et al., 2010; Vastrad et al., 2011; Cegatta and Villegas, 2013; Chakrabarti, 2015). Furthermore, their use would be too expensive and dangerous towards its natural enemies (both of the blue gum chalcid and of other eucalypt pests), so it is not considered a practicable option for large plantations of Eucalyptus spp.

Host resistance

There seem to be differences in the efficiency of insecticides and of natural enemies, probably associated to environmental factors or to the existence of two L. invasa lineages. Numerous studies have highlighted the susceptibility, resistance or tolerance of eucalypt species (including hybrids and clones) to L. invasa attack (Nyeko, 2005; Thu et al., 2009; Basavana Goud et al., 2010; Javaregowda et al., 2010; Dittrich-Schröder et al., 2012; Zhang et al., 2012; Chen et al., 2015; Jacob et al., 2015). Differences in chemical, structural and nutritional characteristics could clarify the mechanisms ruling the resistance or susceptibility of the hosts (Nyeko et al., 2010; Oates et al., 2015).

18/01/17 Original text by:

Francesco Nugnes, Consiglio Nazionale delle Ricerche, Istituto per la Protezione Sostenibile delle Piante, Italy