Toxoptera citricida
(black citrus aphid)

T. citricida is believed to be native to Asia where citrus originated. Since the first half of the twentieth century, the aphid has been known to be widely distributed on citrus in Asia, India, New Zealand, Australia, Pacific Islands (including Hawaii), Africa south of the Sahara, Madagascar, Indian Ocean Islands and South America. This distribution is attributed to movement of infested leaves or propagation material. Areas where citrus was established by seed or aphid-free propagation material have remained uninfested (e.g. North and Central America, Caribbean Basin) until recently (Yokomi et al., 1994). Because ancestral citrus (old line) is known to contain many viruses and virus-like agents, many countries prevent entry of citrus propagation material from abroad. This, undoubtedly, has restricted the aphid's hitchhiking potential. On the other hand, there is now more intercontinental movement of people and commerce than ever before and the threat of introduction by this route remains at its highest level.

Although the aphid is tropical/subtropical in origin, the presence of a sexual stage and overwintering as eggs in Japan (Komazaki, 1982) suggests that T. citricida can adapt to different climates. Due to the restricted host range of the aphid to citrus and its relatives, the most favourable citrus environs for T. citricida occur when weather is warm and humid resulting in frequent stimulation of new growth cycles. Similarly, desert/semi-arid and cooler regions provide conditions favourable for T. citricida only seasonally. Populations typically increase rapidly following colony initiation resulting in crowding, a decline in host suitability, and production of winged (alate) aphids. Winged morph production could also be triggered by the physiology of the host. However, a key requirement for spread of T. citricida is that the alata must alight on citrus with new shoot growth to successfully establish a new colony.

A record for Iran in CABI/EPPO (1998) is an error; the species reported in the literature was T. aurantii and not T. citricida.

Primary hosts of T. citricida are citrus and citrus relatives (Rutaceae) (Order Geraniales, Suborder Geraniineae, mostly in the Subfamily Aurantiodeae, Tribe Citreae). Typically, Aurantioideae are trees or shrubs with evergreen leaves. Flowers are usually white and often fragrant. Many genera bear subglobose fruit with a green, yellow, or orange peel with numerous oil glands that result in a nice aroma when handled. Most commercial citrus varieties and rootstocks are good hosts of T. citricida. In addition, relatives such as calamondin (x Citrofortunella microcarpa) and orange jessamine (Murraya paniculata), rough lemon (Citrus jambhiri), sour orange (C. auranticum), box orange (Severiana buxifolia) and lime berry (Triphasia trifolia) can support T. citricida. There are reports that T. citricida has been collected on many non-citrus plants (Essig, 1949) and Barbados cherry (Malpighia glabra) (Ponte et al., 1998), however, there is no verification that these are reproductive hosts capable of sustaining a population of the aphid. These reports may have resulted from misidentification of aphids. Records from potato and sweet peppers may be from plants growing near Citrus.

The aphid may be able to survive on some non-rutaceous hosts temporarily as they migrate away from a crowded food source (Yokomi et al., 1994). Athough Rutaceae are the preferred hosts, large colonies of apterae are often found on very different plants including Pyracantha (Rosaceae) in Malawi and Zimbabwe; Cudramia (Moraceae) in China and Australia; tea in the Seychelles; and Maclura (Moraceae) in Java.

T. citricida only feeds on newly developed terminals including unexpanded and young expanded leaves and flower buds of citrus and citrus relatives. New shoots generally last 10 days in summer in southern Florida, and 10-16 days in autumn and spring. Once the tissue becomes unfavourable for feeding, the colonies produce alate adults for dispersal, and the remaining nymphs either die or leave the trees searching for other branches or trees. The dispersal of nymphs from one tree to another by crawling a distance of up to 8-12 m is often observed. T. citricida is a colonizer species and is not a strong flyer; long distance disperal is mostly aided by wind current. Alate adults could, therefore, play an important role in the epidemiology of Citrus tristeza virus. However, there is no information available on the percentage of viruliferous alates in a given population during the dispersal period. In countries such as Argentina, Australia, Brazil, Kenya, Taiwan and Japan, two distinct population peaks per year are observed in the spring and autumn. A 2-year field survey conducted in a citrus grove in southern Florida indicated that there were two major peaks of T. citricida populations per year. The peaks did not fall on the same months each year, they were solely dependent on the rainfall of the preceding months (Tsai, 1999). T. citricida is anholocyclic and thelytokous throughout most of its range, preferring warm climates. It can, however, tolerate colder areas such as southern Japan by developing a holocyclic stage and overwintering as eggs (Komazaki, 1993). Development time is temperature dependent. At 20°C, T. citricida has a nymphal development time of 6-8 days with an average pre-reproductive period of 8.1 days, longevity is 28.4 days. Fecundity is 58.5 offspring/female with an intrinsic rate of natural increase of 0.36, net reproductive rate of 56.2, and mean generation time of 11.2 days. Its thermal threshold is 8.4°C and it required 125 degree days for development (Komazaki, 1982). Takanashi (1989) reported slightly longer generation time under similar conditions and differentiated between alata and aptera development time. The developmental periods of immature stages of T. citricida in an insectary ranged from 63.1 days at 8°C to 5.5 days at 30°C when eight constant temperatures (8, 10, 15, 20, 25, 28, 30 and 32°C) were evaluated. The lower developmental threshold for the T. citricida immature was estimated at 6.27°C. An upper temperature threshold for the development of the nymph (31.17°C) was determined from a nonlinear biophysical model. The percentage survival of the immature stages varied from 81 to 97% within the temperature range 8-30°C but survival was reduced to 29% at 32°C. The average longevity of adult females ranged from 60 days at 10°C to 6.5 days at 32°C. The average progeny per female was 52.5 at 20°C and 7.5 at 32°C. The largest intrinsic rate of increase (rm)(0.3765) occurred at 28°C. Populations reared at 10 and 32°C had the lowest rm values of 0.0588 and 0.0960, respectively. The mean generation time of the population ranged from 51 days at 10°C to 8 days at 32°C. The optimal temperature range for population growth of T. citricida was 20-30°C (Tsai and Wang, 1999). Winged morphs develop when populations become crowded and/or food source declines in quality and disperse in search of new hosts to begin new colonies. A spring and a fall flight peak of T. citricida occur in South Africa (Schwartz, 1965), Australia (Carver, 1978) and Brazil (Nickel et al., 1984). In Japan, T. citricida populations peak three times per year but can be found on citrus in all seasons, except when overwintering (Komazaki, 1993). Because the host range of the aphid is restricted to citrus and its relatives (all relatively non-cold hardy), it is unlikely that the aphid can exist outside citrus-growing areas or climates. Host plants have an effect on the development, survival, longevity and reproduction of T. citricida. The percentages of immatures surviving at 25±1°C on rough lemon (Citrus jambhiri), sour orange (C. aurantium), grapefruit (C. paradisi), key lime (C. aurantifolia), box orange (Severinia buxifolia), calamondin (x Citrofortunella microcarpa), lime berry (Triphasia trifolia) and orange jassamine (Murraya paniculata) were 82.0, 93.5, 93.3, 88.3, 53.1, 86.5, 41.6 and 62.8%, respectively. Developmental times for the immature stages among populations were similar (5.9-6.2 days) on rough lemon, sour orange, grapefruit and key lime. Longer developmental periods (6.5-7.2 days) occurred on box orange, calamondin, lime berry and orange jassamine. The average number of nymphs reproduced per female were 58.8, 43.0, 34.1, 42.5, 32.7, 17.7, 20.8 and 23.0 on sour orange, grapefruit, key lime, rough lemon, calamondin, box orange, lime berry and orange jassamine, respectively. Female adults lived an average of 22.1, 19.5, 17.5, 18.0, 22.8, 16.3, 22.6 and 14.6 days on these hosts. The intrinsic rate of natural increase (rm) for T. citricida was highest on grapefruit. Estimates of rm varied from 0.381 on grapefruit to 0.183 on lime berry. The mean population generation time on these hosts ranged from 9.7 to 12.2 days (Tsai, 1998).

The major impact of T. citricida is due to its efficient transmission of Citrus tristeza virus (CTV) (Costa and Grant, 1951; Yokomi et al., 1994), a phloem-limited closterovirus (Bar-Joseph and Lee, 1989). Two types of CTV strains are economically important: those that cause decline of citrus budded onto sour orange (Citrus aurantium) rootstock; and those that cause stem pitting of grapefruit and sweet orange regardless of rootstock. Both are readily transmissible by T. citricida.

CTV is semipersistently transmitted by citrus aphids (Raccah et al., 1976). Aphids acquire virus from an infected trees with feeding times as short as 5-10 min; transmission efficiency increases with feeding times up to 24 h. There is no latent period and the virus does not multiply or circulate in the aphid. The time required to inoculate a plant is the same as for acquisition. The aphid is capable of spreading the virus for 24-48 hours without reacquisition (Meneghini, 1948). T. citricida also transmits Citrus vein enation (woody gall) virus, a probable luteovirus (da Graça and Maharaj, 1991), Cowpea aphid-borne mosaic virus, a potyvirus affecting groundnut in Brazil (Pio-Reheiro et al., 2000) and Celosia mosaic virus, a potyvirus affecting Celosea argentea in Nigeria (Owolabi et al., 1998). Migrating populations of T. citricida are also associated with the spread of certain nonpersistently-transmitted viruses such as Chilli veinal mottle virus (Blackman and Eastop, 1984) and Soybean mosaic virus in China (Halbert et al., 1986).

Most of the reported natural enemies of T. citricida are predators. If predators are present in or adjacent to the citrus grove when T. citricida colonies are forming, they can be effective even when aphid levels are low. Recently in Puerto Rico, predators, especially coccinellids, were observed decimating small T. citricida colonies and eliminating or reducing and/or delaying winged aphid production (Michaud, 1996). Tao and Chiu (1971) point out that T. citricida is toxic to many predators.

The principal primary parasitoids of T. citricida are solitary endophagous Hymenoptera in the families Aphidiidae and Aphelinidae. The aphidiids are wasp-like in appearance and, at pupation, produce a mummy with a typical crusty golden, swollen appearance. They range in adult size from one to several millimetres. Adult Aphelinids are usually less than 1 mm, possess reduced wing venation and an abdomen which appears broadly attached to the thorax. They turn an aphid into a black mummy. Female adult aphelinids also feed on aphid haemolymph, a behaviour that is essential for completion of oogenesis. In Taiwan, T. citricida is parasitized by Lipolexis gracilis and L. scutellaris (Tao and Chiu, 1971), whereas in Australia it is parasitized by Aphelinus gossypii (Carver, 1978). In Japan, Lysiphlebia japonica is the principal parasitoid of T. citricida (Kato, 1970; Takanashi, 1990). In China, Lysiphlebia mirzai is a common parasitoid of T. citricida (Liu and Tsai, 2002).

Entomopathogenic fungi attack T. citricida and can decimate a population with dramatic speed. A critical requirement for efficacy of such fungi is high humidity. This is especially true when using Beauveria bassiana, Paecilomyces fumosoroseus and Metarhizium anisopliae against T. citricida (Poprawski et al., 1999). Verticillium lecanii has been reported to attack T. citricida in Venezuela (Rondón et al., 1980) and other fungi have been observed associated with the aphid in South Africa (Samways, 1984).

Field infestations of T. citricida can best be detected by periodic visual inspection of new shoot growth of citrus. Winged forms can be monitored by yellow traps or suction traps.

T. citricida can be confused with Toxoptera aurantii, the black citrus aphid, because of its presence on citrus, dark brown-black coloration, size and the presence of stridulatory apparatus on the abdomen. However, alata of these aphids can be readily differentiated using a hand lens. T. citricida has antennae III entirely black, forewing pterostigma light brown and media vein twice branched; T. aurantii has antennae III, IV, V, and VI banded at joints, forewing pterostigma conspicuously dark blackish-brown and media vein once-branched. Wingless adults and nymphs are more difficult to distinguish. The easiest character on apterae is the antennae. T. aurantii antennae have several banded joints; whereas T. citricida antennae have one prominent band near the middle. Setal length and patterns can be used to differentiate the aphids but require higher magnification. The cauda of T. citricida is bushy with 25-40 setae; whereas that of T. aurantii is less bushy with 8-19 setae. Another black aphid that occurs on citrus is the cowpea aphid (Aphis craccivora). It can be distinguished by its strikingly white legs (knees of hind leg may be dark) and 7 caudal setae. See Stroyan (1961), Stoezel (1994a), and Halbert and Brown (1996) for full descriptions and keys of citrus aphids.

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.

Cultural Control

Management of virus inoculum is the most important control strategy (Garnsey et al., 1996) because spread of severe strains of Citrus tristeza virus (CTV) is the major problem associated with T. citricida. The first factor to consider is the prevalence of CTV and its strains in a particular area. If virulent stem pitting strains and T. citricida are endemic, citrus scion varieties tolerant to CTV should be planted. These include mandarins, pummelos, tangelos and tangor. Only CTV-tolerant or resistant rootstock should be used. Avoid planting grapefruit or Pera sweet orange unless they have been pre-infected with a cross-protecting CTV strain. If CTV strains are less virulent than the previous scenario, sweet oranges and grapefruit, preferably pre-inoculated with a mild CTV isolate, can be grown with consideration for the market targeted (e.g. fresh fruit, domestic, export, juice, etc.). When CTV problems are anticipated, closer plant spacing should be considered to maximize land use during the grove's early years. Trees that decline or become stunted can either be replaced or simply removed and neighbouring trees allowed to fill in.

Close plant spacing is becoming a common practice in new groves in the USA. Tree size is managed by mechanical hedgers that trim the sides and tops of trees. This practice produces conditions excellent for CTV spread and allows tree canopies to touch in the direction of the row. Pruning induces new shoot growth in which CTV multiplication is optimal as long as temperature and moisture are favourable. Citrus aphid migration, including that of T. citricida, peak in spring and autumn (Carver, 1978; Schwartz, 1965). Hence the uniform growth that results from pruning maximizes opportunities for CTV acquisition and inoculation.

If CTV incidence is undetectable or mild and T. citricida is not established in a particular region, citrus trees grafted on sour orange rootstock may still be acceptable (Garnsey et al., 1996). This decision depends on the risk of losses due to CTV versus the advantages gained by the use of sour orange (e.g. salinity, cold hardiness, Phytophthora, high soil pH, poor drainage). Several areas have managed CTV by eradication of infected trees (for example in Israel and California). This programme is cost effective if virus incidence is low and spread is slow (Garnsey et al., 1996).

Regardless of the present CTV/aphid vector situation, a citrus budwood certification programme is essential for a good citrus industry. CTV and all other citrus virus and virus-like agents are readily graft transmissible. Diagnostic methods are available for testing and detection of citrus pathogens in budwood sources. Recent developments in serology and molecular biology allow some rapid evaluation of pathogen virulence. Thermotherapy and shoot tip grafting are now standard methods to eliminate pathogens from budwood. If a cross-protective CTV isolates are available, they can be incorporated into the budwood certification programme.

Biological Control

Although natural enemies are important in regulating aphid populations, they alone may not be satisfactory for controlling plant virus diseases. Aphid populations on citrus are often too variable to provide sufficient natural enemies for effective vector control. One concept is to direct biological control activities to reduce migrant vector populations before they spread through susceptible crops (Mackauer, 1976). Given that alternative prey are available, natural enemies could reduce T. citricida populations to mitigate secondary spread of CTV (tree to tree within a field), especially if conservation and augmentation efforts are used. In Japan, Lysiphlebia japonica is the most important parasitoid of T. citricida (Takanashi, 1990). However, this parasite was imported and released in southern Florida with no success (Deng and Tsai, 1998). In China, Lysiphlebus mirzai is a common parasitoid of T. citricida (Liu and Tsai, 2002). In South America, various natural enemies have been observed attacking T. citricida but none has been used for augmentation in a biological control programme.

Lysiphlebus testaceipes was found attacking T. citricida in Puerto Rico, Cuba and southern Florida (Yokomi and Tang, 1996; Ravelo and Triana-Fernandez, 1997; Evans and Stange, 1997) but the rate of parasitism was low, as was previously observed in Australia (Carver, 1984). Murakami et al. (1984) did not find effective parasitoids of T. citricida in the Cerrados region of Brazil and suggested that L. japonica be imported and released against T. citricida. T. citricida was introduced in south Florida in the later half of 1995 and spread to all citrus-growing areas in just 3-4 years. Assuming that biological agents colonize new areas more slowly than their hosts, multiple augmentative releases of mass-reared parasitoids at various sites should be conducted (Wellings, 1994). A classical biological control effort undertaken using this strategy in Florida with the release of L. japonica and Aphelinus spiraecolae (Tang et al., 1996) has not been successful.

As most predators are generalist feeders, presence of alternative prey provides stability to their contribution as biological control agents. The spirea aphid Aphis spiraecola is a cosmopolitan species with a wide host range and is quick to colonize new citrus shoots (Cole, 1925; Miller, 1929; Tsai and Wang, 2001). This aphid also develops large populations rapidly on citrus which attracts predators. If T. citricida arrive when these predators are present, they readily attack T. citricida. However, if predators discover a T. citricida population shortly after colonization, their probability of establishment is low without alternative prey as another food source (Michaud, 1996). Several predators including lacewings (Chrysoperla plorabunda), syrphid fly (Pseudodorus clavatus) and coccinellid beetles (Coelophora inaequalis, Coccinella septempunctata, Cycloneda sanguinea, Harmonia axyridis, Hippodamia convergens, Olla v-nigrum and Coleomegilla maculata fuscilabris) have been evaluated in the field and laboratory with varying degrees of success (Michaud, 1999, 2000, 2001; Wang and Tsai, 2001).

Host-Plant Resistance

There is limited information available on the host range of T. citricida and it is unknown if any citrus cultivars or citrus relatives are resistant to the aphid.

Chemical Control

Insecticidal control of T. citricida to slow spread of CTV is an unproven strategy. Although insecticides may not act quickly enough to prevent primary infection by viruliferous aphids, reduction of aphid populations would decrease secondary spread. Its effectiveness depends on longevity of suppression and extent of the treated area in relation to inoculum reservoir and migratory activity of the aphid (Knapp et al., 1996). It should be cautioned that use of foliar insecticides can interfere with biological control agents and, ultimately, their use to protect citrus (a perennial crop) is temporary. In the continental USA, most CTV spread occurs during spring and autumn when temperatures are mild. This is concomitant with when CTV titre (virus replication) in infected citrus trees is highest and when shoot growth and migration of T. citricida peak. Therefore, this time frame should be targeted if chemical control is attempted. Several systemic insecticides including acephate imidacloprid and acetamiprid have been used against T. citricida with various residual effects (Yamamoto et al., 2000; Tsai, 1999). A recent study showed that neem seed extract (azadirachtin) had a marked effect on the survival, longevity and fecundity of T. citricida but not its parasitoid Lysiphlebus testaceipes (Tang et al., 2002).

CTV is transmitted only by vectors that colonize citrus because it is phloem-limited. Thus, its epidemiology resembles persistently transmitted viruses more in this regard than nonpersistently transmitted viruses. As vector control has been shown to limit spread of some luteoviruses (Gourmet et al., 1994) it could be expected to have some impact on CTV spread. However, no data exist to recommend chemical control for CTV control.

Integrated Pest Management

It is not clear what level of vector control is necessary to reduce spread of CTV. The typical integrated pest management (IPM) approaches do not apply for CTV control. Economic thresholds are contingent both on T. citricida population and CTV inoculum pressure. Host-plant resistance to the aphid is not available. A unique management strategy must be practised for CTV in the presence of T. citricida. A strong regulatory component is necessary, covering both propagation and inoculum control (detection and removal of wild and possibly urban reservoirs of CTV) (Garnsey et al., 1996; Halbert and Brown, 1996). Management (conservation and/or augmentation) of biological control agents is feasible. Insecticidal control of vector populations may be useful in specific situations such as in a citrus nursery, or to protect budwood sources. Some value may result from the use of selective insecticides working in tandem with natural enemies. In the final analysis, vector management should be one component of a disease management strategy including other available elements such as mild strain cross-protection; tolerant rootstocks; regulatory measures, isolation or protection of nursery stock; and citrus scions with tolerance or resistance to CTV (Garnsey et al., 1996). In areas where T. citricida and severe (stem pitting) strain of CTV are endemic, the practice of using mild CTV strain to cross protect the subseqent infection of stem pitting strain has been carried out on a commercial basis in South Africa, Australia, Brazil, Peru, Reunion Island, India and Japan (Rocha-Pena et al., 1995; Bederski and Aubert, 1999).