Cookies on Plantwise Knowledge Bank

Like most websites we use cookies. This is to ensure that we give you the best experience possible.


Continuing to use means you agree to our use of cookies. If you would like to, you can learn more about the cookies we use.

Plantwise Knowledge Bank

Your search results

Species Page

cotton aphid

Aphis gossypii
This information is part of a full datasheet available in the Crop Protection Compendium (CPC). Find out more information on how to access the CPC.
©CAB International. Published under a CC-BY-NC-SA 4.0 licence.


You can pan and zoom the map
Save map
Select a dataset
Map Legends
  • CABI Summary Records
Map Filters
Third party data sources:

Host plants / species affected

Main hosts

show all species affected
Carica papaya (pawpaw)
Cucumis melo (melon)
Cucumis sativus (cucumber)
Cucurbita maxima (giant pumpkin)
Cucurbita moschata (pumpkin)
Cucurbita pepo (marrow)
Cucurbitaceae (cucurbits)
Gossypium (cotton)
Gossypium hirsutum (Bourbon cotton)
Prunus salicina (Japanese plum)
Solanum lycopersicum (tomato)

List of symptoms / signs

Growing point - distortion
Inflorescence - honeydew or sooty mould
Inflorescence - twisting and distortion
Leaves - abnormal colours
Leaves - abnormal forms
Leaves - honeydew or sooty mould
Leaves - honeydew or sooty mould
Leaves - honeydew or sooty mould
Leaves - wilting
Stems - stunting or rosetting
Whole plant - dwarfing
Whole plant - external feeding


The initial symptom of A. gossypii attack is a yellowing of the leaves. As aphids become more numerous, leaves become puckered, and curled. As populations continue to rise, aphids move to younger leaves, stems and flowers (sepals mostly). Plants become covered with a black sooty mould which grows on the honeydew excreted by the aphid. Plants also become stunted and (particularly in the cucurbits) the stems become twisted. At very high densities, A. gossypii is able to kill its host. These symptoms are not unique to A. gossypii.

A. gossypii will attack most parts of the plant if population density is high enough. Exceptions include direct feeding on mature reproductive structures (fruits, berries, nuts) and feeding on roots. However, even though there is no direct damage to these structures, a general decline in plant health will affect the proper development of these tissues.

Prevention and control

Biological Control

Despite a number of experimental studies on the natural enemies of A. gossypii (see Natural Enemies), there are few records available of biological control being practised in the field. One successful study using biological control was in Egypt. Two field releases of Chrysoperla carnea at a ratio of 1:5 (predator:aphid) eliminated the aphid in 12 days, whereas it took a single release of Coccinella undecimpunctata at a ratio of 1:50 to get 99.7% control in okra (Zaki et al., 1999).

Cultural Control

Potts and Gunadi (1991) reported a decrease in A. gossypii populations in potatoes that are intercropped with Allium cepa or Allium sativum. To achieve the reduction, the onions had to be planted within 0.75 m of potato plants. However, intercropping poses a problem when the minor crop harbours a disease of the primary crop. Such a system has been documented in Taiwan where bananas were interplanted with cucumbers (an alternative host for Banana mosaic virus) (Tsai et al., 1986). A similar effect also occurs when alternative hosts (of aphid and virus) are in neighbouring fields (Tsai et al., 1986).

In cotton, an unusual approach was to top the plants after boll opening. This removed the top leaves where aphids fed, and thereby reduced contamination of bolls below these leaves. Topping was done by hand, using a pruning knife to remove the terminal spray of each plant (Deguine et al., 2000).

Host-Plant Resistance

Glabrate cotton supported fewer aphids than more pubescent cotton (Dunnam and Clark, 1938; Weathersbee et al., 1994). However, pubescence has the opposite effect on A. gossypii feeding on muskmelons (Kennedy et al., 1978; Ebert and Cartwright, 1997). Greater trichome density resulted in fewer aphids and less virus disease in ashgourd, Benincasa hispida (Khan et al., 2000).

Many crops have some level of physiological resistance to A. gossypii that can be classified into one of three categories: tolerance, antixenosis or antibiosis. The causes for resistance were examined extensively in muskmelons and cucumbers (Ebert and Cartwright, 1997). Resistance has also been documented in okra (Uthamasamy et al., 1976; Gunathilagaraj et al., 1977); Gossypium hirsutum and Gossypium arboreum (Chakravarthy and Sidhu, 1986) (Reed et al., 1999); Antigastra catalunalis (Muralidharan et al., 1977); Citrullus lanatus (MacCarter and Habeck, 1973); Solanum melongena (Sambandam and Chelliah, 1970); Dendranthema morifolium x D. indicum (Storer and van Emden 1995) and Colocasia esculenta (Palaniswami et al., 1980).

While host plant resistance is easily classified into three categories, the cause-effect relationships are not often clear because studies focus on one part (toxicity) and ignore other aspects that may be correlated (genetically linked, or correlated within the study plants but without a genetic basis). Susceptibility to A. gossypii was attributed to high protein and high amino acid content of cucumber cultivars (Ahmed, 1994). Differences in trichome density and differences in toxins were not measured. Likewise, in studies of host-plant resistance, nutritional suitability was not usually measured.

In melons, resistance is conferred by the Vat gene. The cause of the resistance appears to be due to a modified phloem sealing physiology that reduces the quantity of sap an aphid can extract from each feeding site. Furthermore, the phloem they do get has reduced total protein (Chen et al., 1997).

A novel approach to pest management has been the idea of eliciting natural defences in plants using mechanical wounding, infection, or sprays of elicitors. Jasmonic acid applied to cotton plants in California, USA, reduced survival and decreased the number of progeny per leaf (Omer et al., 2001). A similar experiment was done in China, where cotton plants were physically wounded, and some wounds were infected with a bacterium (Pseudomona gladioli D-2251). Wounded plants had fewer aphids and it was less likely that wounded plants would be infested. Infestation frequency and aphid abundance on infested plants was further reduced if the bacterium was present (Li et al., 1998). The phenomenon is widespread, and occurs naturally. A similar study was conducted on cucumber with infection by Cladosporium cucumerinum (Moran, 1998), but without physically damaging the plants. In this system A. gossypii was unaffected by the presence or absence of the fungus, whereas damage from another fungal pathogen was reduced. Wind induced mechanical stress can increase peroxidase activity in cucumber, and with at least 12 days of wind stress can result in detectable reductions in aphid populations (Moran and Cipollini, 1999). However, Moran also found that this mechanical stress can increase pathogen susceptibility.

The opposite of host plant defence would be induced susceptibility. For example, cotton plants fertilized with high nitrogen are better hosts for A. gossypii, and this results in greater damage (Cisneros and Godfrey, 2001; Nevo and Moshe, 2001).

Chemical Control

Due to the variable regulations around (de-)registration of pesticides, we are for the moment not including any specific chemical control recommendations. For further information, we recommend you visit the following resources:



A. gossypii is extremely polyphagous and very damaging to many economically important crops, including cotton, aubergine, citrus, coffee, melon, okra, peppers, potato, squash and sesame. It is a major pest of cotton and cucurbits. A. gossypii has a worldwide distribution, although in arctic regions it is mostly confined to glasshouses. It is particularly abundant in the tropics.

Economic damage due to A. gossypii is by direct feeding, the excretion of honeydew and virus transmission. Damage to cotton, okra and certain cucurbits occurs when large populations of aphids build up, feed on the crops and excrete honeydew. However, its biggest overall economic impact is as a vector of pathogenic plant viruses in over two dozen crops. There is little quantitative information on exact crop losses. In cotton, for example, A. gossypii is only one of many crop pests. Monetary losses to this pest are substantial and are a result of crop loss and crop quality reduction, and the expense of pesticides.

Direct Damage

A. gossypii causes direct feeding damage to cotton, okra and some other crops, by sucking the sap directly from the phloem, causing the removal of nutrients. The drain on plant nutritional resources can be considerable. Adverse physiological responses of plants to direct feeding can also occur. The undersides of young leaves are preferred, but the entire plant may be covered when populations are large. Infested leaves curl downwards and may appear wrinkled or reddened. Heavy infestations can result in wilting. Young plants often have reduced or stunted growth, and may sometimes be killed. A. gossypii is the principal aphid attacking cotton, on which it is an early through mid-season pest; although damaging late season infestations can occur, especially if broad-spectrum insecticides have reduced natural enemy populations (Matthews, 1989; Ebert and Cartwright, 1997).

In the USA, A. gossypii caused more insect-related damage to cotton than any other pest in 1991. Of 13 million acres harvested, around 10 million acres were classified as infested with aphids, resulting in losses of over 360,000 bales (Head, 1992). Losses in Texas alone were around 333,000 bales, representing a yield loss of approximately 6%. Yield reductions of over 100 pounds of lint per acre are not uncommon (Price et al., 1983). In a study of late season aphids, also in Texas, one part of a field was treated to keep aphid populations below 50/leaf, while the other part had aphid densities greater than 50/leaf. In 1992, aphid populations that exceeded 50/leaf for 3 weeks, and 100/leaf for 2 weeks significantly reduced cotton plant height. The higher aphid levels reduced lint yield by 16% in 1992 and 24% in 1993, with a 14% and 25% reduction in seed yield, respectively. None of the cotton quality parameters measured were affected. Differences in gross returns between low aphid plots and high aphid plots in the 2 years were $65.32 and $99.69 per acre (Fuchs and Minzenmayer, 1995). In the USA, the movement onto cotton from noncultivated host plants was an important factor determining A. gossypii infestation; this movement was gradual and ongoing from the time cotton seedlings emerged (O'Brien et al., 1993).

In California, USA, seed yield was reduced by 0.21 lbs seed per aphid-day for cotton planted early in the season (Godfrey and Wood, 1998). However, there was no yield loss in cotton planted later, even though aphid infestation levels were similar. Yield losses were due to a 13.8% decrease in the number of bolls, and a 5.7% decrease in boll size.

The impact of a pest complex on upland cotton in India was calculated in four regions. Aphids, mainly A. gossypii, were more numerous on plants in treatment, compared with control, plots in all cases. Insecticides (soil treatment and regular spraying) resulted in yield (kg/ha) increases between 58.4 and 75.1%. In economic terms, taking into account the cost of plant protection and the net income from the cotton, an additional income of between 4340 and 6115 Rupees/ha was obtained from the pest control treatments (Sivaprakasam and Balasubramanian, 1981). In another Indian study, yield losses in cotton due to sucking pests were between 20.90 and 26.30%, with a glabrous hybrid giving consistently higher yields than a hairy hybrid (Kulkarni and Raodeo, 1986). Reductions in cotton yield due to sucking pests by 16.2 to 55.6% have been reported in studies from Russia and Brazil (Moskovetz, 1941; Vendramin and Nakano, 1981). In cotton in Zambia, A. gossypii damage caused up to 80% yield loss, while a survey of that country's farmers ranked it as the most serious cotton pest (Javaid et al., 1987).

In China, A. gossypii infestations are most serious in the seedling stage, particularly in the northern cotton zone. In one study, cotton seedlings (3 leaves or less) were more sensitive to infestations, while damage on older plants was lower, in part due to compensatory growth effects when precipitation was sufficient. Seedlings were stunted, with a decreased leaf area index and reduced root system development, while the time to squaring was delayed. A damage index, used to establish thresholds for spraying, resulted in adequate control with a 50% reduction in insecticides (Zhang et al., 1982). In Australia, a prototype IPM system maintained cotton yields, while reducing insecticide usage by around 40 to 50% (Hearn et al., 1981).

In Spain, A gossypii is capable of causing 55% yield loss in clementines (Citrus clementina) (Mendoza et al., 2001). Mendoza et al. (2001) provided an estimate of the economic injury level (EIL) and the environmental EIL. The EIL for A. gossypii was 271 aphids/m²; however, this figure will change with the cost of control, the sale price, and yield loss due to the aphid change.

Direct feeding damage by A. gossypii on cotton is related to plant growth stage and level of aphid infestation. Aphid populations increase rapidly with favourable climatic conditions and plant nutritional quality. Levels of damage are influenced by the presence of natural enemies and biological control, pesticide efficacy, the presence of pesticide resistance in aphids, and compensatory growth in plants (Zhang et al., 1982; Slosser et al., 1989). The planting date was the most important variable, in a multiple regression analysis, affecting aphid density in dryland cotton in Texas, USA (Slosser et al., 1989). Optimum temperature for population growth is around 20-25°C (Akey and Butler, 1993). Light intensity and daylength significantly influence rate of population increase, while heavy rain can directly reduce populations by washing them off leaves (Ebert and Cartwright, 1997).

Natural enemies are important in many areas in preventing secondary pest outbreaks. Reduction of natural enemies by insecticides therefore might exacerbate aphid attack by removing this natural control. In a Chinese study, for example, chemical control of the pest complex during the early season led to increased damage to cotton by A. gossypii during the mid and late season (Chen et al., 1991). In predator exclusion experiments in the USA, large decreases in cotton lint yield were observed in caged plots compared with uncaged plots where natural enemies could control aphid numbers. Fibre quality was also reduced by high aphid numbers in predator exclusion cages (Kidd and Rummell, 1997). The reduction in natural enemies resulting in greater aphid abundance was also found in Egyptian cotton fields (Abou-Elhagag, 1998).

The relationship between infestation, nitrogen application and cotton yield was investigated in a study in the Philippines. The more nitrogen applied, the greater was the infestation by A. gossypii, but this was outweighed by the increased yields obtained due to nitrogen fertilizer applications (Villamayor, 1976).

In okra, A. gossypii, along with a jassid bug (Amrasca biguttula biguttula), reduced total yield by 19% in a study in Bangalore, India (Srinivasan and Krishnakumar, 1983). A. gossypii causes most damage at the seedling stage in okra, with feeding reducing the vigour of plants (Pareek et al., 1987).

Indirect damage

Losses due to honeydew, excreted by feeding A. gossypii, can be considerable. Honeydew interferes with leaf transpiration, and acts as a substrate for fungi, including sooty moulds (Capnodium), which blacken leaves and reduce photosynthetic efficiency. Honeydew can also act as an attractant to other crop pests, and insects such as bees, wasps and ants, that may provide protection for the aphids from their natural enemies (Slosser et al., 1989). The presence of honeydew contaminates cotton lint, reducing its quality and economic value (Slosser et al., 1989). The presence of honeydew on fruit crops can significantly reduce their marketability.

Virus transmission
A. gossypii transmits over 50 plant viruses, including non-persistent viruses of beans and peas, crucifers, celery, cowpea, cucurbits, Dahlia, lettuce, onion, papaya, peppers, soyabeans, strawberry, sweet potato, tobacco and tulips (Blackman and Eastop, 2000). In cotton, it transmits Cotton anthocyanosis virus, Cotton curliness virus, cotton blue disease, Cotton leaf roll and purple wilt (Kennedy et al., 1962; Brown 1992). In addition to Cotton anthocyanosis, Lily rosette, Lily symptomless and Pea enation mosaic virus, are all transmitted in a persistent manner (Blackman and Eastop, 2000). On the majority of crops it attacks, the ability of A. gossypii to act as a vector of plant viruses can result in significant economic losses (Ebert and Cartwright, 1997).

A. gossypii is an important vector of Papaya ringspot virus, transmitting both the P (PRSV-P) and W (PRSV-W) strains. The former is a disease of papaya, whereas PRSV-W, also called Watermelon mosaic virus 1 (WMV-1), infects cucurbits and watermelon (Kessing and Mau, 2000). A. gossypii also transmits Watermelon mosaic virus 2 (WMV-2), Zucchini yellow mosaic virus (ZYMV) and Celery mosaic virus (CeMV). These potyviruses are transmitted in a non-persistent manner.

A. gossypii is the most important vector of Cucumber mosaic virus (CMV) in cucurbits. CMV has one of the widest host ranges of any plant virus. It can be acquired in 5-10 seconds and be transmitted in less than 1 minute. The ability of CMV to be transmitted declines after about 2 minutes and is usually lost within 2 hours (Francki et al., 1985). Citrus tristeza virus (CTV) is transmitted in a non- or semi-persistent manner by A. gossypii; the virus remains infectious for 24 hours, and is specific to plants in the Rutaceae (Brunt et al., 1996).

A. gossypii is the only known vector of the virus causing cotton blue disease in Africa. In field trials in the Central African Republic, yield losses due to A. gossypii and cotton blue disease were significantly reduced with insecticide treatments. A combination of organophosphorus and pyrethroid insecticides was recommended for control (Cauquil et al., 1978; Cauquil, 1981). In Chad, all cotton varieties were found to be susceptible to cotton blue disease, except one with a small degree of tolerance. Cotton blue disease decreased the length of fibres by about 2.5%, compared with uninfested controls, although other quality characteristics were unaffected (Dyck, 1979).