Home-immediately access 800+ free online publications. Download CD3WD (680 Megabytes) and distribute it to the 3rd World. CD3WD is a 3rd World Development private-sector initiative, mastered by Software Developer Alex Weir and hosted by GNUveau_Networks (From globally distributed organizations, to supercomputers, to a small home server, if it's Linux, we know it.)ar.cn.de.en.es.fr.id.it.ph.po.ru.sw

CLOSE THIS BOOKRice - Fish Culture in China (IDRC, 1995, 240 p.)
Part III: Interactions
VIEW THE DOCUMENTMaterial Cycles and Economic Returns in a Rice-Fish Ecosystem
VIEW THE DOCUMENTFish Culture in Ricefields: Rice-Fish Symbiosis
VIEW THE DOCUMENTEcological Effects of Rice-Fish Culture
VIEW THE DOCUMENTEcological Mechanisms for Increasing Rice and Fish Production
VIEW THE DOCUMENTRice-Azolla-Fish Cropping System
VIEW THE DOCUMENTEffect of Fish on the Growth and Development of Rice
VIEW THE DOCUMENTThe Role of Fish in Controlling Mosquitoes in Ricefields
VIEW THE DOCUMENTDeveloping Rice-Fish Culture in Shallow Waters of Lakes
VIEW THE DOCUMENTAbility of Fish to Control Rice Diseases, Pests, and Weeds
VIEW THE DOCUMENTDistribution and Residue of Methamidophos in a Rice-Azolla-Fish Ecosystem
VIEW THE DOCUMENTResidue and Application of Fenitrothion in a Rice-Fish Culture System

Rice - Fish Culture in China (IDRC, 1995, 240 p.)

Part III: Interactions

Material Cycles and Economic Returns in a Rice-Fish Ecosystem

Ni Dashu and Wang Jianguo

In the rice-fish ecosystem materials move in a benign cycle and the energy flows in the direction favourable to both rice and fish. The ricefields nourish the fish, and the fish nourish the rice. Like other theories, the theory of rice-fish mutualism has only been adopted slowly. The meaning of the word mutualism has, in recently years, been extended from its original meaning in classical ecology. It has now taken on the meaning of functional mutualism in addition to the original organization sense. A mutual relationship is one in which two different species live together and promote and accelerate their growth.

Although rice and fish are mutually beneficial, they are not totally dependent on each other. Their coming together is based on scientific principles and the anticipation of greater economic returns. As the system is further developed, and rice-fish culture is widely recognized as the best way to increase yields, their mutualism will become more of a necessity.

Chinese ecologist Ma Shijun said it is necessary to simulate mutualism of different species of plants and animals according to one’s needs. The theory of rice-fish mutualism was founded on both conceptional and practical principles.

Rice-Fish Ecosystem

Ecology, in a direct sense, is a branch of science that studies habitat. It is, indeed, very closely related with the development of the national economy. In nature, animals, plants, and microorganisms come together to form a unified entity, or ecological system. The close relationship among animals, plants, and microorganisms and between these organisms and the environment, is made possible by the flow of energy and the circulation of material.

Ecological systems are both large and small. The biosphere is a large system; a ricefield or pond is a small ecosystem. In addition to these natural ecological systems, there are other ecological systems, such as the rice-fish system. All agricultural systems are, in fact, anthropogenic.

The nonbiological factors in the rice-fish ecosystem include light, water, water temperature, pH, carbon dioxide, oxygen, and some inorganic matter. The biological factors in the ecosystem include producers, consumers, and decomposers. The main producers are plants with roots and large and small phytoplankton. In another words, there are three categories of producers in ricefields: rice plants, weeds, and algae. They are all involved in the circulation of carbon through photosynthesis and respiration, and they provide organic matter to consumers and decomposers.

There are also many consumers. They include zooplankton (protozoa, rotifers, and crustaceans); benthos (nematodes, molluscs, annelids, and water insects); fish reared in ricefields (common carp, crucian carp, bighead carp, nile tilapia, and grass carp); mosquito larvae, insects, and worms harmful to rice; natural enemies of harmful insects and worms (spiders and parasitic wasps); and the natural enemies of fry (chilopods, scorpions, dragonflies, frogs, otters, water rats, eels, loach, water snakes, sandpipers, ducks, kingfishers, gulls, and egrets). Many animals are both primary consumers and secondary or tertiary consumers. For example, water snakes feeds on frog, frogs feed on fry, and fish feed on plankton. Many animals are harmful to rice but useful to fish, and vice versa. For example, although frogs feed on fry, they also feed on many of other insect and worms that are harmful to rice. The composition of the producers, consumers, and decomposers in the rice-fish ecosystem is complicated and merits further investigation.

Cycling of Material and Energy Exchange

To create a rice-fish ecosystem in a ricefield, it is necessary to pay attention to the appropriate time and size of the system to ensure that the rice and fish are truly mutually beneficial. Material must be made to circulate in a benign cycle and the energy flow must be in a direction favourable to both rice and fish (Figure 1). The rice-fish ecosystem is created by adding fish fry or fingerlings to the ricefield. In this system, the cycling of matter and the movement and storage of energy becomes more rational. The difference, compared with a natural ecosystem, is that the rice-fish system is controlled and adjusted by the farmer. Of course, to perfect such an ecological system, many improvements are needed.

In the rice-fish ecosystem, rice is the dominant biological community. It absorbs large quantities of light, carbon dioxide, water, and inorganic elements and manufactures organic matter by photosynthesis. The large quantities of weeds, plankton, and photosynthetic bacteria in the ricefield undertake the same processes. However, they do not provide useful products. On the contrary, they compete for fertilizer, space, area, and sunshine with the rice and in some cases are the intermediate hosts of rice pests. Of course, weeds and plankton are all primary producers that help fix and store energy. The primary consumers are mainly zooplankton, herbivorous animals, and plant pests. The secondary consumers are mainly carnivorous animals.

Fish in the rice-fish system can be primary consumers, secondary consumers, or tertiary consumers. This creates the problem of which fish to raise to make the system more efficient. It is also the leading factor that affects the density of other biological species and communities. Repeated experiments and comparisons have demonstrated that grass carp are the best fish to use for rice-fish culture.

In ricefields, grass carp eat a large quantity of weeds. There are more than 30 kinds of common weeds in ricefields. Eleocharis yokoscensis, Hydrilla sp., Potamogeton crispus, Vallisneria spiralis, Najas marina, Potamogeton distinctus, and Lemna minor are eaten by grass carp. In general, weeds can reduce rice yields by 10-30%. Therefore, if weeds could be totally eliminated, rice yields should increase by over 10%. Our experiments have indicated that early ricefield without fish have 13-15 times more weeds than fields with fish. When the fish are harvested there are about 33-435 kg/ha of weeds, but in ricefields without fish there are 450-6520 kg/ha of weeds when the rice is harvested, even when weeding is done three times. The weeds eliminated by grass carp (coefficient of feed 1:80) provide about 5 kg of fish output. Furthermore, fewer weeds are available to compete for fertilizer. This stimulates increased rice output, purifies the water, and improves the environment.

By eating the plankton, weeds, and benthos in the ricefield, the grass carp grow quickly. The more they eat, the more excrement they discharge. A grass carp (6.5-13 cm) is estimated to eat 52% of its own weight and to excrete 72% of the amount of grass it eats. If 400 grass carp are reared for 110 days, fish excrement amounts to about 26475 kg/ha. This excrement is rich in nitrogen and sulphide and therefore increases the fertility of the field.

In most cultivation systems, most of the weeds in the ricefield are pulled out and discarded. This causes a large loss of soil fertility and wastes the solar energy captured by the weeds. In addition, much of the bacteria, plankton, zooplankton, and part of the benthos, are usually discharged with the water. This accounts, either directly or indirectly, for loss of soil fertility and solar energy. From the point of view of circulation of matter and energy, this is a natural phenomenon that is unavoidable. However, from the point of view of maximizing bioproductivity, it is obviously a waste of matter and energy. The raising of fish in ricefields captures part of the matter and energy that would otherwise be wasted and transforms them into fish products. At the same time, the fish stimulate rice output. This is a very economical practice. It is desirable to continue to seek ways to improve the system and to strive for the highest possible yield using the least amount of energy and matter to produce the maximum economic returns.

The introduction of grass carp into the ricefield changes the composition of the biological species and communities and their mutual relationships. Grass carp and rice become codominant factors in the system.

In the rice-fish ecosystem, nonbiological and biological factors are important. Growth and development of rice requires light, heat, carbon dioxide, water, and nutrients. Of these factors, air, water, and nutrients undergo the most dynamic changes and exert an extremely large influence on the growth of the rice plants. For example, carbon dioxide is an indispensable raw material for photosynthesis. During the day, the amount of carbon dioxide in a ricefield with fish is higher than in a ricefield without fish. The fish respire carbon dioxide and feed on plankton that compete with the rice for nutrients. Generally, there is an increase of 1.5-8.2 mg/L (average 5.1 mg/L) in dissolved oxygen in ricefields with fish. The minimum level of dissolved oxygen at night is also tolerable for grass carp. Furthermore the fish tend to raise the dissolved oxygen content level because they stir up the water and increase the contact between water and air. The activities of the fish can also make the distribution of oxygen more uniform. Because they move the soil, the fish also improve the oxygen supply to the soil, which favours the breakdown of organic matter and reduced material in the soil. This is why many rice-fish fields that are not exposed to the sun and are not weeded still yield 10% more than fields in which fish are not reared.

Economic Returns

The first and foremost objective of raising fish in ricefields is to increase rice output while reducing the labour required for weeding. Rice yields are increased (by about 10%) in rice-fish fields. In addition, grass carp make full use of the water and feed provided by the ricefield, harmful insects and other rice pests are reduced, and the system retains and creates more fertilizer.

This new model of a rice-fish ecosystem is becoming increasingly popular. From 1980 to 1983, Hubei and Hunan Provinces devoted about 33 300 ha of ricefields to this new model. If the area devoted to raising fingerlings is assumed to be 2000 ha, the area for growing food fish 2670 ha, the average output of fingerlings 4875/ha, and the output of food fish 525 kg/ha, the two provinces produced 9.75 million fingerlings and 1.4 million kg of food fish with an output value of CNY 3.2 million. If the increase in rice output is assumed to be 10%, the added output would be worth CNY 1 million. The total value of the rice and fish would be CNY 4.2 million.

In 1984, China had nearly 0.7 million ha of ricefields devoted to rice-fish farming. This was an increase of more than 80% from 1983. The increase in rice output was estimated at 285 million kg and the output of fish at 47 000 tonnes. Sichuan Province, which leads the country, devoted 0.3 million ha to rice-fish farming, and Chongqing City alone reserved 77 330 ha for rice-fish farming. Hunan Province had 0.2 million ha for rice-fish in 1984, 33.7% more than in 1983. In addition, many households have reported collecting 7500 kg of rice and 750 kg of fish from 1 ha ricefield.

If these improved methods of raising fish in ricefields could be applied to 6.7 million ha over the next 3-5 years, the rice output could be increased by 2 billion kg and the catch of fingerlings would amount to 20-50 billion, an abundant source of supply to raise adult fish. This would help China reach the goal of producing 4-5 million tonnes of fresh fish each year.

Ni Dashu and Wang Jianguo are with the Institute of Hydrobiology, Academia Sinica, Wuhan, Hubei Province.

Fish Culture in Ricefields: Rice-Fish Symbiosis

Xiao Fan

Fish culture in ricefields originated from natural symbiosis. The accidental discovery of wild fish in ricefields, and the subsequent catch of both adult fish and fry, induced people to make use of ricefields for fish rearing. Although the traditional rice-fish rearing was a modification of the natural system, modern rice-fish culture has undergone significant development in recent years. The improved systems have already surpassed traditional methods with respect to structure, carrying capacity, energy conversion and exchange, and use of materials, and produced ecological, financial, and social benefits.

Within the rice-fish ecosystem, the plants and animals complement and interact with each other. The organized food chains produce various materials (living and nonliving, organic and inorganic, molecular and ionic) that interact with the other biological, chemical, and physical activities in the ricefield. As a result, the food chains in the ricefield are rebuilt, soil in the fields is made fertile, the structure of the water and soil is improved, insect pests are diminished, diseases are controlled, and mosquitoes and weeds in the fields are reduced.

Rice-Fish Food Web

The long history of rice production in China has affected the natural ecosystem. In Jiangsu Province, the use of large quantities of poisonous insecticides and chemical fertilizers since the 1960s has killed rice pests in large areas. However, at the same time, many other useful organisms have also been destroyed. The use of these chemicals has changed the characteristics of the natural ecosystem, brought about an imbalance in the ecology, and caused the gradual disappearance of ecological advantages. The current rice-fish culture system has reestablished the food chain of fish eating insects and weeds and made it possible to use little or no chemical herbicides to kill weeds and no insecticides to control insect pests.

Microbes and mosquito larvae

When rice and fish are raised together in ricefields, the plants need fertilizers and the fish need rich food. In fields where fish are being raised, organic manure should be used as the basal fertilizer. Only chemical fertilizers that are not poisonous to fish should be used for supplementary applications. Ammonium bicarbonate can be used as a top dressing. The ammonium bicarbonate (15-20 kg) should be shaped into balls and placed under the soil in fields covered with 6-7 mm of water.

When manure is applied to the ricefield, benthos and plankton reproduce rapidly and provide the fish with sufficient food. However, as the fish grow, their need for food increases, but the availability of food in the fields decreases. A field investigation showed that the amounts of benthos with fish and without fish were, respectively, 4.3 g and 12.9 g in mid-June, 8.0 g and 25.1 g in early July, and 5.4 g and 10.2 g in mid-July. The Jiangxi Aquatic Research Institute compared fish and nonfish ricefields in 1984. There are fewer phytoplankton (625 500/mL or 2.4 mg/L of water) and fewer zooplankton (18 730 000/mL and 3.6 mg/L) in the fields with fish compared with fields with rice only.

Fish culture in ricefields differs greatly from pond culture because there is a lot of plankton in the fields and the amount of life decreases gradually. Fewer fish are stocked in ricefields (30 000/ha) than in fish ponds (3 million/ha). A study by the Hubei Aquatic Institute demonstrated that there was as much as 119 g/m³ of benthos in ricefields, but only 39 g/m³ in ponds. Moreover, the amount of potential food increased after the fry had been added to the fields and reached its peak 6 days later. As the fish grow and their appetites increase, the amount of food that is available begins to decrease. However, in ponds, the benthos began to decrease sharply only 3 days after the fry were introduced. After 5 days, the amount had dropped to less than 10 g/m³, which is far too little to meet the needs of the fish.

Fish culture in ricefields also helps eliminate mosquito larvae. In these areas, the density of larvae in the ricefields was reduced by 50-90%, and in residential areas the number was reduced by more than 50%. The Chendu Municipal Health Station, Sichuan Province, surveyed three different areas severely affected by malaria. The incidence of malaria was 0.01% in 1981 when there was no rice-fish culture, but in 1982, the incidence dropped to 0.002% after rice-fish culture was started. Jiadian County Health Station, Shanghai Municipality, monitored the rice-fish fields of the Chuanjin Aquatic Production Farm from 2 July to 18 October. No larvae were found in the 18 samples from rice-fish fields, but in the nonfish fields, larvae were found in every sample. The average per sample was 2.2 larvae (1.7 in July, 3.3 in August, and 2.1 in September). The use of organic insecticides has killed large numbers of mosquitoes, but has also given rise to insecticide-resistant strains. Rice-fish culture is an effective control method for all types of mosquitoes, including the insecticide-resistant strains.


Fish eat many of the weeds in ricefields. Some herbivorous fish loosen the soil by tilling and digging holes, which uproots the tender roots and stalks of the weeds. Weeding by fish is timely and frequent and superior to chemical weeding.

When the fry are first put into the ricefields, they feed on plankton. When the weeds begin to sent out sprouts, the small fish eat these sprouts as well as small insects. As the fish grow, their ability to eat weeds increases. If there are sufficient fish in the fields, they grow synchronously with the weeds, and control them. Even the withered rice leaves that fall into the water are eaten. A study by the Zhejiang Provincial Academy of Agronomical Sciences showed that on 22 August there were 8.0 kg of weeds in a rice-fish field, compared with 30.3 kg of weeds in fields without fish. On 11 November there were no weeds in the rice-fish field, but 3.07 million green duckweed in the field without fish.

Insect Pests

During the growth and development of rice, insects can be eliminated by fish if the proper measures are used and the habits of the insects are taken into consideration. For example, when rice planthoppers develop in the ricefields, they can be driven into the water if a rope is pulled over the rice plants. Because the planthoppers pretend to be dead when they fall from the rice plants, they are easily eaten by the fish.

During the incubation and developmental period for rice borers, a layer of water should be maintained in the ricefields. Because the rice borers transfer to a new rice plant after incubation, they will be forced to travel in the water where they can be eaten by the fish. If pest levels increase, the water level should be raised to drown part of the stalks and leaves and enable the fish to catch the insects. If the fish are close to the affected parts of the plant, they will jump to catch the insects.

Insect pests are normally not very serious and can be easily eliminated. Even during severe infestations, these methods can be used to reduce pest populations. According to materials published by the Rudong Botanic Protection Station in Jiangsu Province, in rice-fish fields there were 100 nest of rice planthopper with 984 eggs, compared with 4468 eggs in fields without fish. An investigation in Rugao County showed that in rice-fish fields there were 30% fewer eggs of the yellow stemborer, Tryporyza incertulas, the rate of white ears was 50% lower, there were 50% fewer rice planthoppers, the rate of rice leafrollers was 30% lower, the rate of white leaves was decreased by 30%, and the number of rice leafhoppers was 30% lower.

Furthermore, in rice-fish fields, the rice plants are usually very strong and have good resistance to diseases. The possibility of rice diseases was also reduced because of the fertile water and good environment, improvements in varieties, reduction in density, good ventilation, and sufficient light. A study in Chenxian County, Zhejiang Province, demonstrated that under the same cultivation conditions, the indices of sheath and culm blight of rice were 11.8, 10.7, and 7.8 in fields without fish, rice-fish fields, and idle rice-fish fields, respectively,

Improvements in Soil and Water Conditions

Soil Fertility

In rice-fish fields, the activities of the fish help mix the manure with the soil. The fish swallow, digest, and assimilate 30-40% of the organisms living in the fields. The rest of the organic matter is excreted into the fields and becomes manure. The fish faeces are a good quality manure that contains 42% phosphorus (a higher level than in pigs and cattle manure). Nutrient analysis has shown that are 1.2 times more phosphates in rice-fish fields than in fields without fish, and ammonia levels are 1.3-6.1 times higher. The Soil Fertilization Station in the suburbs of Yancheng County, Jiangsu Province, made a comparison of rice-fish fields with fields without fish and found that in rice-fish fields where fish had been raised for 2 years, the organic matter level was 1.8% in both fields and the nitrogen content was 0.12%. In ditches with fish, the organic matter content was 1.9% and the nitrogen content 0.142%, which was much higher than in ditches without fish.

Gases and Nutrients

Under normal conditions, the diffusion of oxygen in water is 10000 times slower than that in air. This often results in anaerobic conditions at the soil-water interface. The activities of the fish increase the contact area of the water with air and profoundly change the gas structure of the water and soil and improve their physical properties and chemical composition. A gas determination of the soil has shown that in the fields where early rice is planted and fish are raised, oxygen is present to a depth of 5 mm in the soil, but not to 10 mm. In fields where late rice is planted and fish are raised, aerobic condition extend to 8 mm because the fish are larger and more active. In fields without fish, the surface of the soil-water interface is normally anaerobic.

Rice-fish culture helps raise rice production by:

· Increasing the oxidation of the soil and decreasing the reducing agents (e.g., H2S, Fe++, and Mn++).

· Making it impossible for the medium matter (formed as a result of incomplete dissolution) to mineralize rapidly, to continuously release energy and produce various NH4+ and PHO4- ions, and to renew the humus in the soil.

· Allowing the highly concentrated nutrients to spread to the roots of the rice plants because of the activities of the fish.

Water Temperature and Oxygen Concentration

The water temperature and the conditions for oxygen solution in the ricefields are better than in fish ponds. The thin layer of water in ricefields puts large areas of water in contact with the air. There are also 100 times fewer fish in ricefields than in ponds. This is why fish do not come to the water surface as often in ricefields as in ponds.

Fish Diseases

The water in ricefields is usually shallow and fresh and is replenished frequently. Rice plants absorb fertilizers and purify the water in the fields and, as a result, the water is continuously fresh and clear (much better than the water in ponds). The absolute number of pathogenic bacteria in pond water is 2.6 times higher than in ricefields. The number of bacteria in the water has a direct effect on the number of bacteria in fish gills. The number of bacteria on one side of the gills of fish in ponds was 160 x 106; whereas in fish in ricefields there were only 18.5 x 106. The change in the bacteria content of the water in ricefields clearly reflects the incidence of fish diseases. From February to September, the number of bacteria in ricefields remains stable, and the incidence of fish diseases is low. July, the month during which fish diseases increase, is the time when the number of bacteria in ponds is highest.

Because fish live in the water, it is difficult to make any accurate diagnoses of diseases. Generally, sick fish have no appetite, and medicine cannot be applied or mixed with fish food. The significance of rice-fish culture is low fish density and a health environment, which promote normal growth of the fish and prevent stress.

Biological Control of Rice Pests

Insecticides and herbicides are normally used to prevent and control insect pests and weeds. Part of the insecticide is absorbed by the rice and the rest drops into the water and soil. In fact, some insecticides are directly applied to water or soil and consequently contaminate them. For example, in the early 1980s, 9.6 kg/ha of BHC were used. Some of the BHC entered the soil and water, but most was dissolved and flowed away. The part that was absorbed as a residue in the crops was consumed by humans in the rice, and the by-products (bran and straw) were used as fodder for livestock or fish, whose eggs, meat, and milk were eaten by people. The residue of the insecticide is being transferred from one organism into another and in the end accumulates and concentrates in the human body. The Scientific Experiment Base, Taihu Lake Area, determined that insecticide residues are highest in rice stalks (4.3 mg) and leaves (5.1 mg) followed by rice husks and roots (both 3.8 mg). In the rice grain, the highest concentration was found in the husks and rice bran (3.4 mg). In crude rice, the level is 0.7 mg; in refined rice 0.3 mg.

Tests were carried out in 13 counties of Jiangsu Province in 1983 on the residues of organochloride insecticides in rice. In samples of middle rice (which accounts for 31% of the total rice output of the province), the BHC content ranged from 0.01 to 1.06 mg/kg (average of 0.16 mg/kg). Of the samples, 99.1% contained BHC and 13.7% exceeded the allowable limit. The highest content (1.06 mg/kg) was 2.5 times more than the allowable limit. In samples of late rice (which accounts for 49% of total rice production), the BHC content ranged from 0.07 to 1.21 mg/kg (average 0.34 mg/kg). All samples contained residues, and 54% exceeded the allowable limit.

The human body absorbs 34.4% of the insecticide residues in the grain. An investigation by the Scientific Experiment Base, Taihu Lake Area, found that the amount of BHC residue a person absorbs each day through grains, edible oil, meat, fish, and vegetables was 0.58 mg, which was 15 times higher than the maximum allowable limit suggested by the World Health Organization (0.039 mg per 65-kg person or 0.0006 mg/kg body weight). The insecticide remains in the fatty tissues and other organs and causes damage to human health.

Rice-fish culture reestablishes a symbiotic ecosystem, prevents environmental pollution, and preserves an ecological balance in agriculture. Farmers have made great achievements in irrigation, rice-strain selection, planting techniques, and fish culture. Consequently, rice-fish symbiosis has been further developed. At least, three advantages have been confirmed.

· The ecological benefits of rice-fish symbiosis are becoming more obvious. The elimination of insects and weeds by the fish directly protects large quantities of living organisms from pesticide use. Therefore, other useful organisms, natural enemies of insect pest in particular, survive and reproduce. This extends the possibility of biological pest control and consolidates the ecological benefits of rice-fish symbiosis.

· Rice-fish culture ensures production of fine-quality fish strains and market-size fish and increases the income of farmers. Despite the decrease of planting area, rice unit output increases. The income from rice-fish culture has increased, and in some cases doubled, the income from traditional ricefields. This fact can be used to popularize rice-fish culture.

· Rice-fish culture has also reduced soil and water pollution. Polluted areas become less contaminated or completely unpolluted through the process of self-purification.

Xiao Fan is with the Crop Cultivation Technical Station, Department of Agriculture and Forestry, Nanjing, Jiangsu Province.

Ecological Effects of Rice-Fish Culture

Pan Yinhe

Rice-fish culture is a traditional farming system. Since 1978, the area devoted to rice-fish culture has been expanded several fold, fish production has increased rapidly, and fish-farming technology has been improved. In many areas, good harvests of both rice and fish have been achieved (7500 kg of rice and 750 kg of fish per hectare).

Ni Dashu developed the theory of rice-fish mutualism, in which ricefields are used for fish culture and fish farming increases rice production. This paper discusses the ecological effects of rice-fish culture and its economic, social, and ecological efficiencies.

Effects on the Ecosystem

Abiotic factors (e.g., water, soil, light, heat, and air) and biotic factors (e.g., crops, animals, and microorganisms) are closely interrelated and interdependent and form an ecosystem in the ricefield. In this ecosystem, the biotic community is transfers and cycles energy and materials.

The ricefield is a typical anthropogenic ecosystem in which rice production is the main activity. The rice absorbs solar energy, carbon dioxide (CO2), water, and various nutrients and through photosynthesis produces organic matter and energy, which are stored and converted into rice and straw. At the same time, wild grasses and other weeds, phytoplankton, and some photosynthetic bacteria grow in the ricefields. However, these products are not as useful and complete with the rice. In the ricefield, zooplankton, herbivorous animals, some insects, and pathogenic bacteria are the primary consumers. The carnivorous animals are the secondary consumers, and both bacteria and fungi in the soil decompose organic matter into inorganic matter.

In ricefields without fish, farmers must carry out regular and labour-intensive weeding. As a result, there is a heavy loss of soil fertility and solar energy and an increase of production cost. Because most of the bacteria, phytoplankton, and aquatic animals in the ricefield cannot be used by the rice, they are lost with the irrigation water. Moreover, insects, pests, and mosquitoes can reproduce rapidly and adversely affect both rice and human health.

When fish are introduced into ricefield ecosystem, the population and composition of aquatic organisms, and the relationships among them, change. Population numbers change. Fish, the largest consumers, eat weeds, phytoplankton, zooplankton, aquatic insects, and other animals. Fish have the greatest effect on population density and mortality. Because they are primary consumers, grass carp, common carp, and crucian carp feed heavily on weeds. In China, more than 100 varieties of weeds grow in ricefields. Of these, Hydrilla verticillata, Ptamogeton crispus, Vallisneria spiralis, Potamogeton matainus, and Lemna spp. are considered to be good feed for grass carp.

The Biological Department of Southwest Teachers College, Chongqing, Sichuan Province, stocked fish in ricefields at a rate of 3000 fish/ha (grass carp 30%, common carp or crucian carp 60%, and silver carp 10%). After 75 days, the fish had consumed 12 465 kg/ha of weeds and only 360 kg/ha remained. If 50% of the weeds growing in ricefields were consumed by the fish, this would produce 78 kg/ha of grass carp based on a food conversion rate of 1:80. Therefore, rice-fish culture can effectively eradicate weeds and control the loss of energy from ricefields.

Rice-fish culture can change the direction of energy flow in the ecosystem. In the ricefield, the stocked fish transform stagnant energy (e.g., weeds) and possibly lost energy (e.g., phytoplankton, zooplankton, and aquatic insects) into useable products (fish and rice). Rice-fish culture also coordinates the interrelationship between the biotic and abiotic environments. In ricefield ecosystem, rice requires light, heat, air, water, and nutrients for its growth. Air, water, and nutrients have the greatest impact on rice production. Because the ricefield is usually flooded, the normal water requirements of rice can be ensured. However, an inundated field does not favour root development of the rice. Under inundated conditions, dissolved oxygen (DO) from the surface water can only be supplied to soil through diffusion and transpiration. In general, the level of dissolved oxygen in the surface water varies diurnally with algal photosynthesis during the day. Dissolved oxygen usually reaches a maximum (12-14 mg/L) when light is adequate. However, more than 95% of the DO is taken up by various organisms in the surface water and little of the DO diffuses and permeates into the soil.

Under these circumstances, as temperature rises, soil reduction increases and reducing substances (e.g., methane, organic acid, and hydrogen sulphate) increase and decay rice roots. This problem is normally solved by sun-drying the ricefield. However, as fish move about in the ricefield, they increase contact between the air and water. This increases oxygen content throughout the field. In addition, the fish disturb the soil, which accelerates decomposition of organic matter and reduces the concentration of reducing substances.

Although sun-drying and weeding are sometimes not practiced in rice-fish fields, rice production is higher than in fields without fish culture. From the viewpoint of aquaculture, the total dissolved oxygen level is low in rice-fish fields (less than 4 mg/L in the early morning). However, fish mortality due to the oxygen depletion has not been reported.

Generally, the ricefield has a pH of about 7.0, which is optimal not only for the growth of rice and fish, but also for the reproduction of natural food organisms. Fish also have a positive effect on soil fertility because of the accumulation of fish excreta, which has a high nutritive value (Table 1). Silver carp excreta was the best, grass carp and common carp excreta second best, and crucian carp excreta the poorest. The concentrations of N and P in the fish excreta were higher than in pig and cow manure, similar to those of night soil and sheep manure, but lower than those of chicken and rabbit manure.

The daily manure production of one fish has been estimated to be about 2 g. If the average stocking density was 3000 fish/ha (stocking size about 100 g), 6000 g of fish manure would be produced every day. This would amount to 450 kg/ha of fish manure if the fish were reared for 75 days. The N content of the soil was reduced at the end of the production season by 1.1% in the ricefield with fish and 12% in the field without fish. The fish are able to transform the energy in the ricefield ecosystem and enrich the soil.

Fish can also minimize outbreaks of diseases and insect pests and reduce the application rate of pesticides, which can pollute water, soil, rice, and fish. When fish are cultured with rice, the main primary producer (rice) and consumer (fish) are combined to form a symbiotic rice-fish ecosystem.

In rice-fish fields, the rice reduces sudden changes in water temperature caused by sunlight, adjusts and stabilizes water temperature and quality, and, therefore, provides an environment that is conducive to the reproduction of natural organisms. Because the fish consume phytoplankton, zooplankton, and weeds that compete with rice, they play an important role in increasing and stabilizing soil fertility, eradicating harmful insects and pests, recovering lost energy, and adjusting energy flow. In the symbiotic rice-fish ecosystem, the mutualism between rice and fish is fully exploited to provide high-quality products and good environmental conditions.

Efficiency of Rice-Fish Culture

Economic Efficiency

Rice production is increased by 5-15% in rice-fish culture. Experiments in many locations have demonstrated that rice growth is improved in rice-fish fields. In particular, the rice developed evenly, tillering is improved, more rice grains are produced, ears are heavier, and the rate of false grains is lowered.

Rice-fish culture can also increase the production value of ricefields. Based on the collection of nation-wide information, net profit can be increased by CNY 300-750/ha. Profits can be even higher (CNY 1500-15 000/ha) if fry are reared in the ricefields. The economic efficiency is increased because the fish have a high value.

Fish can also help eradicate weeds, minimize the loss of fertilizer, and reduce outbreaks of insects and pests. Therefore, fertilizers, pesticides, and labour can be saved. In experiments in Taoyuan County, Hunan Province, the concentration of quick-acting N and P in rice-fish fields was increased by 10% and 124%, respectively, compared with fields without fish. Fish are able to reduce populations of rice hoppers and rice leafrollers 2-6 times. As a result, the application frequency and quantity of pesticides can be decreased. Moreover, based on investigations in Jiangxi, Guizhou, and other provinces, about 120-180 labour units per hectare can be saved with integrated fish culture. In some places, farmers do not plough the field when rice-fish culture is practiced. This further reduces the inputs needed for rice planting, and therefore, reduces production cost and increases the economic efficiency of rice cultivation.

Social Efficiency

Rice-fish culture expands the area for fish culture and produces more fish products. Rice-fish culture also produces (with less input) increased numbers of large-size fingerlings for the development of fisheries in ponds, reservoirs, and rivers. If the ricefield is used to culture food fish, average production is 300-750 kg/ha (maximum 750-2250 kg/ha). This practice is an effective way to increase fish production in hilly areas. At the same time, rice-fish culture effectively increases the income and living standard of farmers, particularly those living in hilly, rural areas.

Rice-fish culture also increases rice production. It makes multiple use of the ricefield to maximize the utilization of land and water resources. The proper combination of crop production and aquaculture will effectively promote the transformation of the structure of rice production.

Ecological Efficiency

In rice-fish culture, harmful insects and pests are greatly reduced. Therefore, pesticide application can be reduced or eliminated, and toxicity accumulation is minimized. This is beneficial to human health and the ecological balance of the environment. For example, the number of predators of rice pests is higher in rice-fish fields without pesticides than in fields without fish and with pesticides. Rice-fish culture also improves the environment and reduces infectious diseases of livestock and humans. In ricefields, mosquito larval, maggots, snails, and leeches, which are the intermediate host of malaria, encephalitis, dysentery, blood fluke, and filaria, reproduce rapidly. Fish, particularly common carp, crucian carp, tilapia, and other omnivorous fish, consume and eradicate these pathogenic parasites and minimize the infestation rate of human beings, thereby creating an improved living standard and a better level of health for the farmers.

Pan Yinhe is with the Freshwater Fisheries Research Centre, Chinese Academy of Fisheries Sciences, Wuxi, Jiangsu Province.

Ecological Mechanisms for Increasing Rice and Fish Production

Pan Shugen, Huang Zhechun, and Zheng Jicheng

Experiments plus production practices have demonstrated that rice and fish production can be increased by raising fish in ricefields. Mechanisms to increase rice and fish production have been developed based on experiments and investigations in Sanming Prefecture, Fujian Province.

Ecological Environment

Physical and Chemical Environment

In a double-cropped field in Ningua Sanming, Fujian, the mean water temperature from May to October was 27.5°C and the accumulated temperature was 5055.6°C. The highest temperature was 38°C on 27 July and the lowest was 17.5°C on 26 October. Compared with air temperature, the mean water temperature was 1.3°C higher and accumulated temperature was 231.1°C higher. Other observations in Yongan from 14 August to 30 November showed that the mean water temperature in a late ricefield was 25.5°C. Because the water was shallow, the temperature rose rapidly and sunshine reached the soil directly. Therefore, the water temperature was similar in the upper and lower layers, which favoured decomposition of organic matter.

In Ninghua, water levels in the fields used for rice-fish cultivation varied from 3 to 10 cm, and no water remained in the field after the field was drained field and the crop matured. The water level in the field for rice-fish rotation varied from 60 to 80 cm. The water demand for a field producing 7500 kg/ha of rice was 36 000 m³. Water for irrigation varied from 9000 to 12 000 m³. Because of water is shallow and there is a great exchange of water, the environment for fish was limited, and fertilizer and food flowed away easily.

The dissolved oxygen level was high in the fields because of the water exchange and the large amount of oxygen released by the rice and phytoplankton. In Ninhua, the average dissolved oxygen level varied from 3.9 to 5.6 mg/L (maximum 12 mg/L at noon on sunny days). Fish in the ricefields have a higher metabolism and a higher rate of food utilization than fish cultured in ponds.

Based on determination from many locations in Sanming, the pH of water in ricefields varies from 6.3 to 6.8. Most of the soils in the mountainous districts of Fujian are red, and the water in the fields is acidic.

In Ninghua, the rate of biological oxygen demand (BOD) in fields with fish was 33.4 mg/L; ammonia nitrogen 0.80 mg/L, nitrate nitrogen 0.68 mg/L, phosphate 0.06 mg/L, hardness 0.42 mg equivalent weight/L, and alkalinity 0.65 mg equivalent weight/L. These parameters are comparable with the levels found in rich water in reservoirs. This is why the fish cultured in ricefields have a high level of productivity. In Shangmin, there was less phosphate and lower hardness in the ricefields, and because of uneven fertilizer application and greater water exchange, the richness and stability of the fertilizer were decreased.


Sunlight is the most important energy source in ricefields. Based on data from meteorological observations, annual light duration ranged from 1708 to 1898 h and total irradiation from 91.3 to 106.4 kcal/cm². The rice used less than 1% of the light energy, therefore before the rice covered the field, most of energy was used by weeds and phytoplankton. This means that the light is also the major energy source that will eventually be used by the fish in the ricefields.

Rice roots, straw, flowers, and grain are the products of photosynthesis, and much of them remain in the field. In Ninghua, experiments indicated that in 1 ha of ricefield there was 48 210 kg of roots, 17 745 kg of stubble, and 14 385 kg of straw. Straw contains 9-13% cellulite, 1.6-3% potassium, and 35-40% cellulose, which favours growth of microorganisms and diatoms. Results from pollen-chamber studies show that 1400-1500 grains are produced per flower, and the blossoms drop to the field after the flowers are fertilized. In Ninghua, two seasons of rice blossoms amounted to 1559 kg/ha. Rice blossoms are rich in protein. There is a saying the more fragrant the rice blossoms, the fatter the carp. During harvest, about 3-5% of the grain drops into the field. Rice-plant trash was 81 000 kg/ha, or about one-quarter of the products produced by photosynthesis. These products provide organic matter and fertilizer for rice-fish culture.

In ricefields producing 7500 kg/ha of rice, the demand for nitrogen is 120-188 kg/ha, phosphorus 60-113 kg/ha, and potassium 135-270 kg/ha. In Shangmin, the amount of fertilizer applied to a field producing 7500 kg/ha of rice was 1125 kg/ha ammonium carbonate, 450 kg/ha calcium superphosphate, and 1200 kg/ha organic fertilizer. About one-third of the volatile section was absorbed by the rice and about half was drawn into the soil and dissolved in the water to become nutrients for food organisms. This is the major source of fertilizer for rice-fish culture.

Food Organisms

In the ecological environment of the ricefield, there are many organisms besides rice and fish. Investigation in Ninghua indicated that there were 25 families and 433 species of vascular plants in the ricefields. In the late ricefields without fish, the amount of biomass was determined at three times (before harvest, after the field was drained, and when the grain was filled). The amount of biomass averaged 6045 kg/ha, and most of the species were suitable food for fish.

Both the number of species and the quantity of plankton in ricefields were reduced compared with levels in fish ponds. Based on investigations in Jianning, Sanming Prefecture, there were 6 phyla and 61 genera of plankton, of which 20 genera were diatoms, 29 genera were green algae, 5 genera were blue-green algae, 1 genus was golden algae, and 1 genus was Dinophyceae. There were also 3 species of protozoa, 10 species of rotifer, 1 species of Cladocera, and 2 species of copepods.

The concentration of phytoplankton was 15-65/L; zooplankton was 900-2800/L. The recommended fertilizer rate and time of application also encouraged rapid plankton growth. Biomass reached 75-119 mg/L, which was 4-6 times higher than in fish ponds and easily satisfied the food requirements of fish fry.

Based on investigations in Jianning, there were 22 species of benthos, of which 17 species were insects, 3 species were gastropods, and 2 species were nematodes. Studies in Ninghua showed that the biomass of benthos can reach 109.3 kg/ha. All these species are good food for fish.

In fish stomachs, organic detritus was found most frequently. In Jianning, 42 carps were dissected and large amounts of organic detritus were found in all fish (1 g of organic detritus contains 450 bacteria, which weigh about 5% of organic detritus). Bacteria are rich in protein and are eaten by zooplankton and benthos animals as well as by fish. Bacteria play an important role in increasing rice and fish production.

Pests and Diseases

Fish diseases are rare in rice-fish culture because the water is clear and the oxygen content is high. In addition, the lower stocking density and rich natural food produce strong fish that are more disease resistant. Fish pathogens are rarely seen in ricefields. Based on investigations by Han Xianpu, there were 4100 bacteria per millilitre in ricefields compared with 8800/mL in fish ponds. Pathogenic bacteria were 1.6 times lower than in fish ponds. There were no significant differences in the number of bacteria on the fish bodies between the field and fish ponds. However, in ricefields, the number of bacteria in the fish gills was 1850/mL, compared with 16 000/mL in fish ponds (7.6 times lower). Therefore, rice-fish culture is an effective natural method of protecting fish (especially carp) from disease.

Because the water is shallow in the fields, it is difficult for fish to escape from predators (e.g., centipedes, leaches, birds, snakes, frogs, rats, and otters). In addition, the narrow levee of the ricefield can easily collapse if rats and eels dig holes in the bank. Fish can easily escape from ricefields if there is flooding; therefore, fish survival is not as high.

According to this analysis, organic matter and food organisms will provide more than 20 kg of natural fish productivity. In 1983, in Ninghua, a 21.9-ha field was used for rice-fish cultivation. Production averaged 316 kg/ha (without feeding). In Jianning, fish production from 20.6 ha of ricefields was 282 kg/ha (without feeding). In 1985, in Ninghua, Yongan, 133.3 ha of ricefields stocked with fish produced 720 kg/ha with the addition of some fertilizer and weeds.

These studies demonstrate that ricefields offer clean water, enriched food, less disease, and can provide over 300 kg/ha of natural fish productivity. However, there are still some problems to be solved (e.g., shallow water, unstable water quality, predators, and easy escape).

Increases in Rice Production

Controlling Weeds and Fertilizing Fields

Weeds compete with rice because they also need carbon dioxide, water, and nutrients for photosynthesis. Fish stocked in ricefields eat weeds continually and effectively control weed growth. Carp eat weeds at the rate of 30-50% of their body weight, and 1-year-old carp eat 25 g of weed seeds (about 4000 seeds) a day. Experiments during the late season in Ninghua showed that in ricefields stocked with fish, weed weight averaged 161.7 g/m³, compared with 604 g/m³ in the control field.

In Yongan, in a ricefield with rice-fish rotation, weeds numbered 23.3/m², compared with 137.1/m² in the control field (a decrease of 4.5 times). In Ninghua, weeds in the late-crop field stocked with fish decreased 4425 kg/ha compared with the control field. Using the value of 3.3% as the mean amount of nitrogen needed by vascular plants, these reductions would preserve 9.7 kg of total nitrogen. In addition, fish kill weeds continually and more efficiently than humans.

Accumulating Fertilizer and Increasing Fertility

Much of the food that fish consume is excreted and becomes a fertilizer for the rice. Only 30-40% of the weeds are digested by the fish, the remainder is excreted as feces and urine. If a fish excretes 2% per day of its body weight as feces, a field used for rice-fish culture that produces 315 kg/ha of fish without feeding for 180 days would produce 567 kg of feces.

A field used for rice-fish rotation that produces 1845 kg/ha (Yongan) for 120 days would produce 2205 kg of feces. Carp feces contain 1.1% nitrogen and 0.4% phosphorus. In rice-fish culture, fish feces contribute the equivalent of 31.5 kg of ammonium sulphate and 13.5 kg of calcium superphosphate. In rice-fish rotation, fish feces contribute the equivalent of 121.5 kg of ammonium sulphate and 52.5 kg of calcium superphosphate.

Soil tests in Yongan in 1982 showed that with rice-fish rotation, organic matter, total nitrogen, and total phosphorus increased by 0.6%, 0.03%, 0.001% respectively, compared with the control field. Soil tests in Ninghua in 1984 demonstrated that rice-fish cultivation increased organic matter by 0.09%, total nitrogen by 0.04%, total phosphorus by 0.38%, available nitrogen by 22 ppm, and available phosphorus by 2 ppm compared with the control field. Analysis of the water in the ricefield showed similar results. In fields stocked with fish, BOD increased by 7.49 mg/L, ammonium nitrogen by 0.14 mg/L, and phosphate by 0.032 mg/L compared with the control field.

Loosening of Soil and Promoting Fertility

When submerged in water, the soil of a ricefield experiences slow breakdown of organic matter, thorough decomposition of vegetation, stable fertility, and less loss of nutrients. If submerged for a long time, the soil has intensive reduction and anaerobic decomposition, which produce large amounts of organic acid. These acids are unfavourable for rice roots. When soil reduction is intense, methane and hydrogen sulphide are produced and damage rice roots. Therefore, the field must be tilled and dried during the middle growth stage to intensify soil oxidation and control the formation of reducing substances.

Fish move about in the ricefield looking for food. Consequently, they enhance contact between the water and air and increase the dissolved oxygen level. As the fish move, they stir the anaerobic layer of the soil. This accelerates the breakdown of organic matter and favours root growth. When rice roots draw nutrients from the soil, they decrease the concentration of nutrients around the root. The plants depend on infiltration of soil water that contains nutrients. This movement to the rice roots occurs very slowly. As the fish swim, they stir nutrient evenly and accelerate infiltration of nutrients into the soil. This helps the roots obtain nutrients more effectively.

Controlling Rice Pests

Fish eat many pests (e.g., rice plant hopper and leafhoppers) when they drop into the water. Some pests (e.g., rich borers, rice root worms, and snout beetles) damage rice after they travel through the water. While in the water, they are easily eaten by fish. Fish also help control bacterial diseases (e.g., spotted wilt disease) because they eat the cysts of the bacteria. An investigation in Ninghua in 1984 indicated that leafhoppers and rice planthoppers in ricefields stocked with fish were reduced by 16%, yellow rice borer by 17%, and spotted wilt disease by 52% compared with the control field. In Jianning in 1985, spotted wilt disease was reduced by 28-51%, withered paddy by 15-32%, and rice planthoppers by 70-84% in the field stocked with fish.

These factors combine to increase rice production in fields stocked with fish. In 1984 in Ninghua, the height of the rice increased by 2%, effective rice ears by 14%, number of grains by 9%, rate of fruit bearing by 2%, weight of 1000 grains by 4%, and yield by 7.1% in the field used for rice-fish culture. Similarly in Yongan in 1983, the height of the rice increased by 6%, effective rice ears by 12%, number of grains by 2%, the rate of fruit bearing by 0.4%, the weight of 1000 grains by 2%, and yield by 18% in rice-fish fields.


Ricefields provide an ecological environment that is suitable for both rice and fish. When ricefields are stocked with fish, the fish eat food organisms and organic detritus. Energy and material that used to be lost are captured and converted into fish protein. Fish kill pests and weeds and excrete feces to the field. In addition, fish movements promote air exchange and distribution of fertilizer.

There are contradictions between rice and fish when applying fertilizer and pesticides and draining fields. In recent years, these contradictions have been resolved by digging ditches and pools, applying fertilizer to all layers of the field, and applying pesticides that are highly efficient and have low toxicity.

In ricefields used for fish culture, there are still some problems (e.g., shallow water, excessive water exchange, unstable water quality, predators, and fish escape). Therefore, further improvements are still required.

Pan Shugen is with the Jimei Fisheries School, Jimei, Fujian Province; Huang Zhechun is with the Ninghua Popularization Centre of Fisheries Techniques, Ninghua, Sanming Prefecture, Fujian Province; and Zheng Jicheng is with the Yong’an Popularization Centre of Fisheries Techniques, Yong’an, Fujian Province.

Rice-Azolla-Fish Cropping System

Liu Chung Chu

China has a long history of raising fish in ricefields. However, fish yields are low because of difficulties in applying feed to the large areas of fish-raising fields. Azolla is a small aquatic plant that contains abundant nutrients because it can fix atmospheric nitrogen, carry out photosynthesis, and uptake nutrients from its surrounding environment through its root system. It is also an excellent feed for fish. Azolla is rich in the amino acid arginine, which may play an important role in fish growth (Tables 1 and 2). Azolla grow quickly, produce high yields, are a suitable size for fish grazing, do not require harvesting or chopping, and can grow in the ricefield. To increase its ecological and economic benefits, a rice-azolla-fish cropping system was established in 1981. These experiments have indicated the potential of this cropping system.

The Role of Azolla

Fish Feed

Both grass-feeding and omnivorous fish eat azolla. Grass carp (Ctenopharyngodon idella) and nile tilapia (Oreochromis niloticus) consume the equivalent to more than 50-60% of their body weight in azolla each day. The amount of azolla consumed by the common carp (Cyprinus carpio) increases with increased size.

Feeding experiments with four fish species were conducted by the Soil and Fertilizer Institute, Hunan Academy of Agricultural Sciences. The feed conversion coefficient of azolla was 49.0 for grass carp, 52.1 for tilapia, 31.2 for Hunan crucian carp, and almost zero for lotus carp. The weight gain for these species was 174 g, 134 g, 36 g, and 5 g, respectively. There were high levels of 15N-labelled azolla in the internal organs of nile tilapia and low 15N-labelled azolla levels in the external organs at the start of the experiments. However, the level of 15N-labelled azolla in the internal organs gradually decreased, whereas the level of 15N-labelled azolla in the muscles greatly increased (Table 3).

After uptake of 15N-labelled azolla for 96 h, 15N recovery in the intestine, stomach, and liver decreased from 10.3% to 1.0%, 1.6% to 0.2%, and 2.4% to 0.7%, respectively. A similar trend was found in other internal organs. In contrast, 15N recovery in muscles was 6.3% at 18 h and increased to 10.1% at 96 h. Metabolism balance estimates were obtained from a 4-day nile tilapia experiment. The amount of nitrogen accumulated by the tilapia represented 30% of total azolla N. Tracer techniques (using 15N) were used to obtain a better understanding of nutrient abundance in fish feces. During the 96-h excretion period, the highest determined 15N level was 3.8%, the lowest was 2.1%. This was much lower than the level of 15N in the azolla fed to the fish. This reduction was probably due to the dilution of 15N from the azolla by the other nitrogenous matter excreted from the alimentary canal of the fish (which includes digestive juice, sloughed cells from the stomach, and azolla). These results demonstrate that azolla-N accounts for 30% of N in fish feces. Because another 30% of total azolla-N accumulates in the fish body, it can be estimated that azolla-N is about 60% digested by the fish (N may also be excreted into the water in the form of urine, as excretions from the body surface, as falling scales, or as matter exchanged by the gills). The utilization of 15N from azolla in the rice-azolla-fish system is increased to 67.8% (Table 4); whereas, the rice-azolla treatment using 15N-labelled azolla as a top dressing at the maximum tillering stage had a utilization rate of 46.1%.

Fish Yields

In the traditional rice-fish system, fish grow slowly because there is insufficient feed. This problem can be solved by introducing azolla. Experiments over 3 years demonstrated that the rice-azolla-fish system will produce fish yields of 1000 kg/ha. As well, yields can be further increased by using some other techniques (e.g., the polyculture of grass carp and nile tilapia). Fish yields were almost doubled compared with the traditional system for silver carp (Table 5). The yield of edible fish is also raised. The rice-azolla-fish system increased farm income by about CNY 1954/ha.

Effect on Rice Yield

The rice-azolla-fish system provides an excellent growing environment for rice, fish, and azolla. Because of the high amount of organic fertilizer provided by fish, the rice grows well (Table 6), and because the fish eat azolla, rice pests, and weeds, the use of chemical pesticides can be reduced. However, the environment created by the rice-azolla-fish system is also conducive to the survival of the natural enemies of rice pests (e.g., spiders and black ants). This further decreases pesticide requirements. For example, during a plant leafhopper outbreak in Fuqing Country, Fujian Province, in 1984, four applications of pesticides were required in traditional rice-growing systems and provided incomplete control. In contrast, only one application was required under the rice-azolla-fish system. Observations from 1983 to 1986 indicate that the rice-azolla-fish system effectively suppresses weeds and rice pests (Table 7).

Waterweed species




Floating azolla species


Waterweed species



- -






Soil Fertility

In the rice-azolla-fish system, plant nutrients are provided by decomposition of azolla and by excretion of fish feces. Improvements in fertility were greater in the ditches than on the field surfaces (Table 8). This can be attributed to the effect of the fish in the system, especially the role played by fish feces in improving soil fertility. The rapid increase of available potassium is also apparent, which demonstrates the capability of azolla to enrich potassium levels. Although the rice yields from this system are similar to traditional systems, an extra 375-600 kg/ha of fish are harvested. The fish decrease the amount of mineral fertilizer required by the rice plants, maintain or improve soil fertility, and create an excellent ecological environment.

Implementation of the Rice-Azolla-Fish System

Field Design

Two forms of field design can be considered for the introduction of the rice-azolla-fish system. The first method involves digging pits and ditches in a traditional ricefield and transplanting rice seedlings in accordance with normal spacing practices. In the second method, rice seedlings are transplanted on ridges and fish are raised in the ditches between the ridges. The selection of fields is particularly important for both designs. In both cases, the field must have sufficient water and have good controlled of irrigation and drainage. In most cases, a rectangular pit(s) that occupies 5% of the total ricefield area will suffice. In all cases, pit depth should be between 1 and 1.5 m. Ditches are 30-50 cm deep, 40-50 cm wide, and occupy 3-5% of the total ricefield area. The field is designed according to the desired yields of rice and fish. In another words, to harvest more fish, pits and ditches should occupy more area, field ridges should be wide and thick to prevent fish escape, and drainage openings should provide for good irrigation.

Combinations of Fish Species

Fish species should be chosen in accordance with their feeding efficiency. For example, grass carp are unable to fully digest cell walls of plants because their alimentary canal lacks cellulase. Consequently, they excrete feed residues into the water along with fish feces. Nile tilapia excretions stimulate the propagation of plankton. Under these conditions, pure cultures of either species do not use azolla efficiently. However, this problem can be solved if a mixed culture (polyculture) of silver carp and common carp with grass carp, nile tilapia, and common carp (ratio of 100:300:100:7500 fingerlings/ha) is introduced after the rice seedlings are transplanted.

Growing Season for Azolla

The key technological problem in the rice-azolla-fish system is a healthy and sufficient azolla biomass. Two methods are recommended to increase the azolla biomass: increase the space between rice rows to give the azolla sufficient room to grow, and prolong the propagation period to ensure the fish have sufficient food to eat. Polyculture of different azolla species and other waterweeds can also be introduced. Various kinds of waterweeds (e.g., Lemna minor and Wolffia arrhiza) can be cultivated in the ricefield to supply fish feed in June and July (this method is called rotational cultivation of azolla).

Field Management

Water management is the most important factor in the rice-azolla-fish system. During the early stage of rice growth, fingerlings can swim freely in the shallow water, which is good for tillering of early rice. Later, the larger fish need deeper water. At this time, the water temperature can sometimes reaches 40°C, and it is necessary to keep the irrigation water at the depth of 8-10 cm.

Fertilizer should be applied principally as green manure supplemented with chemical fertilizers. Basal application is stressed and should account for 70% of the total amount of fertilizer used. Deep placement of granules of N fertilizer decreases the loss of N, which benefits both fish and rice. To prevent disease and pests, it may be necessary to apply some insecticides, but the type, application rate, and application methods must be suitable for fish. Biological control methods are preferred.

Liu Chung Chu is with the National Azolla Research Centre, Fujian Academy of Agricultural Sciences, Fuzhou, Fujian Province.

Effect of Fish on the Growth and Development of Rice

Li Duanfu, Wu Neng, and Zhou Tisansheng

A rice-fish system was investigated for 3 years to determine its effect on the growth and harvest of rice and the income to farmers. A ridge-ditch cropping system was used.


An experimental ricefield of average fertility was plowed and levelled. A ridge-ditch system was used. The ditches were 20 cm wide and 30 cm long and the ridges were 30 m long and 50 cm wide. The ditch was 25 cm deep (from the surface of the ridge to the bottom of ditch). Rice was planted on the ridges and fish were stocked in the ditches. Rice plants were spaced at 17 cm x 13 cm, with 4-5 plant per clump and three line of rice plants per ridge. The ridges and ditches were estimated to cover 84% and 16% of the field area, respectively. There were three replicates for each of three different treatments. Treatments were randomly arranged. Nine small (0.02 ha) experimental areas were established. The total experimental field was 0.18 ha, and there was a 0.04-ha protective area around the field. The different treatments were separated by a low bank that was covered with a 50-cm plastic membrane that prevented fish escape and leakage of fertilizer.

Carp (7500 fry/ha) and grass carp (450 fry/ha) were released immediately after the rice seedlings were transplanted. Supplemental feed (375-390 kg/ha) was given until the rice plants bloomed. Fertilization and management techniques were the same as used for ordinary ricefields.

Results and Discussion

Fertilization of Ricefield

When the ricefield was stocked with fish, the nitrogen, phosphorus, and potassium (NPK) contents of the soil and water were increased significantly. Total nitrogen was particularly high. Weeds and plankton in the ricefield normally compete with rice for fertilizer. However, they were eaten by the fish and converted to a fertilizer that could be used by the rice. The physical and chemical properties of the soil also became more suitable for growth and development of the rice.

Oxidation and Reduction Potential of the Soil

A rice-fish ecosystem benefits both crops. Fish movements in the shallow water break the surface membrane formed by the microorganisms covering the soil. This increases the dissolved oxygen level in the soil and elevates its oxidation and reduction potential during the period of rice growth. These changes improve the oxygen content and effectively increase the utilization rate of soil nutrients. The ridge-ditch system allows water to be drawn into the soil in the ridge without having a negative impact on the fish. The sun can also increase the temperature of cultivation layer, which helps increase rice yields, especially of late rice. The ridge-ditch system can allows for the use of direct seedling, ratooning, and zero-cultivation method of rice planting.

NPK Content of Rice Plants

The NPK contents of the leaves and culm of rice plants grown with fish were higher than in the control. These differences were correlated to the differences in NPK levels in the soils in the two plots.

Chlorophyll Content of Plants

The chlorophyll content of rice plants at every developmental stages were significantly higher in the experimental ricefield. The high chlorophyll content indicates that the process of photosynthesis was more efficiency, which would lead to the accumulation of more carbohydrates.

Surface Area of Leaves

The surface area of leaves has higher in the early developmental stages in the experimental ricefield. In the booting and mature stages the factors were 6.9 and 2.5, respectively. In the control fields, the corresponding figures were 5.6 and 1.4. The larger surface area of the leaves and the higher content of chlorophyll will increase the efficiency of photosynthesis, and therefore increase the number of effective ears, the number of grains per ear, and the weight of the grains.

Activity of the Root System

The activity of the root system is expressed by the volume of water that flows through a wounded stem per unit of time. Strong activity means that the root system can absorb more nutrients from the soil. The root systems of rice plants grown in the experimental field always had stronger activity than the roots of plants in the control field at all developmental stages.

Accumulation of Dry Matter

The NPK content, surface area of the leaves, chlorophyll content, and activity of the root system were all higher in the rice-fish system. These differences are also expressed in the accumulation of dry matter. The total dry weight of the whole rice plant in experimental ricefield was 17.1% higher than in the control field. This is a fundamental condition for an increase in rice production.

Effect on Tillering

Tillering of the rice plant during the early stages of development is crucial stage to the production of effective ears. The number of ears and the time of earing are closely related to fertilizer level. The rice plants grown in the experimental field had a greater rate of tillering per day and more effective ears per plant. Although both fields originally received the same amount of fertilizer, the fish in the experimental field promoted more efficient use and distribution of NPK. The fish reduced the loss of fertilizer and increased soil fertility.

Weed Growth

Carp are omnivorous and grass carp are herbivorous. However, grass carp fingerlings also eat aquatic insects. When these two species of fish are stocked together, weeds are can be controlled in the ricefield. In the experimental field, there were significantly fewer weeds throughout the growing period.

Economic Benefits

The rice-fish system creates a mutually beneficial ecosystem. In the ridge-ditch system, the production of fish can reach 642 kg/ha. At the same time, the fish add fertilizer and eliminate pests and weeds from the ricefield. Rice yields were increased by 14.4%. It has been estimated that the ridge-ditch system can double total earnings.

Li Duanfu is with the Guangxi Agricultural Institute and Wu Neng and Zhou Tisansheng are with the Guangxi Institute for Prevention and Cure of Parasitic Diseases, Nanning, Guangxi Zhuang Autonomous Region.

The Role of Fish in Controlling Mosquitoes in Ricefields

Wu Neng, Liao Guohou, Lou Yulin, and Zhong Gemei

Starting in 1983, investigations were made for 5 years on the effect of controlling mosquitoes after rearing fish in ricefields. In 1987, further financial support was obtained and the research began to have economic impact.

Experiment Sites

The test site was in Quangzhou County in northeast Guangxi. More than 26 700 ha of ricefields were suitable for fish culture (about 80% of the total cultivated land). Traditionally, farmers stock 6000-9000 common carp and 150-1500 grass carp per hectare of ricefield after the rice is transplanted. No additional feed or management was used, and fish yields were about 150 kg/ha. However, because weeds were decreased and rice yields were increased, this type of cropping system has expanded. By 1987, rice-fish culture was practiced on half of the ricefields that were suitable for fish-rearing.

An isolated village in Quangzhou County was selected as an experimental site to study changes in mosquito population density in ricefields. The village had 127 ha of ricefields, and 90% of these fields were used for fish. The fish fingerlings were stocked into middle-rice fields after the rice seedlings were transplanted. When rice was harvested, the fish had grown to about 100 g and could be used as food or grown longer in the pond.


Mosquito Density in Ricefields

The main species of mosquitoes in the district are Anopheles sinensis, the main vector of local malaria, and Culex tritaeniorhynchus, the vector of Japanese encephalitis. Density measurements were taken once a week for a month before and after the fish were stocked. Larva and adult mosquitoes were also examined in a control village with no fish.

Frequency of Mosquito Biting

Mosquitoes were attracted to a special mosquito net 0.5 h after sunset in the fields adjacent to both the experimental village and the control village.

Age-Class Distribution of Larva (Pupa)

Larva numbers of each age-class were recorded throughout the year at both the experimental site and the control site. The distribution of mosquitoes in each age-class was evaluated and the differences were calculated.

Incidence of Malaria

The spread of fish culture in ricefields over the past 10 years in Quangzhou County was traced. The incidence of endogenous malaria was recorded over the same period and compared with the whole district.

Results and Discussion

Density of Mosquitoes

Compared with the control, the density of larva and adult mosquitoes was remarkably lower when the fish were reared in the ricefields. A comparison of the frequency of mosquito biting in the two locations also showed that contact between humans and mosquitoes was greatly reduced in the village where a large area of the ricefields was used for fish culture.

Natural mortality of mosquito larva is density dependent. The degree of the effect depends on the stage at which mortality occurs. If natural predators consume mosquitoes during the early stages of growth, increased survival is likely to make up for early losses. If mortality takes place in later stages, it is impossible to make up for the loss. This will greatly affect the density of adult mosquitoes. In the fish-rice field, the ratio of old larva and pupa was much lower than in the control field. This suggests that because the fish feed on old larva and pupa, density-dependent survival has no effect. Therefore, it is reasonable to suggest that fish are an effective biological control method for mosquitoes.

Incidence of Malaria

One of the most important criteria for judging the control of mosquitoes is the incidence of diseases spread by the mosquitoes. Table 1 shows the increased area of rice-fish fields in Quangzhou County and the annual incidence of endogenous malaria within the county and within the whole district. As the area of rice-fish culture has increased in Quangzhou County, the annual incidence of malaria has decreased (correlation coefficient -0.9225). Although other measures were taken to prevent malaria in Quangzhou County (e.g., inspection and control of sources of infection), the relative number of cases was much lower than in other counties in the district.

Wu Neng, Liao Guohou, Lou Yulin, and Zhong Gemei are with the Guangxi Institute for Prevention and Cure of Parasitic Diseases, Nanning, Guangxi Zhuang Autonomous Region.

Developing Rice-Fish Culture in Shallow Waters of Lakes

Wan Qianlin, Li Kangmin, Li Peizhen, Gu Huiying, and Zhou Xin

Per-unit fish output from lakes, especially large ones, is low in China. Rice-fish culture increases the yields of both fish and rice; therefore, the potential exists to invigorate inland fisheries and to increase rice production. Deepwater rice is adaptable to different water depths and is a stable and natural adjusting and controlling factor that could help solve the difficult problems involved in the establishment of large-scale ecological agriculture. Deepwater rice might also prevent the proliferation of blue algae in East Taihu Lake and lessen the impact of water pollution and eutrophication caused by population growth and urban and rural development. Deepwater rice is usually grown in areas that hold water during floods. Deepwater rice is a major crop in Southeast Asia, but few data are available on deepwater rice-fish culture in lakes. The feasibility of deepwater rice-fish culture was studied to observe the growth of deepwater rice in deep ponds and in shallow lakes and to determine which species of fish might be suitable.

Experiments with Deepwater Rice

Deepwater Rice

Twenty-three varieties of deepwater rice (including two varieties of floating rice) developed by the International Rice Research Institute (IRRI) and quarantined in the Philippines were studied. Seeds were dried in the sun, soaked, and germinated on 18 May 1987. Seedlings were transplanted on 18 June, three to a hill, 25 cm apart and in rows that were 25 cm apart (variety no. 1 was planted two to a hill). Some of the seedlings were transplanted to a 60-m2 area in Huayuan Lake that was 10-40 m in depth. On 12 July, the water in the lake rose sharply by 1 m in 12 h, which submerged the rice plants. The water remained at that level for so long that all the rice plants drowned.


Five days after the seedlings were transplanted, 3200 crossbred fingerlings were released into a 0.07-ha experimental plot. The fingerlings were bred by the Freshwater Fisheries Research Centre of the Chinese Academy of Aquatic Products. There were 1000 crosses between Cyprinus carpio Wuyuanensis and C. carpio Yuankiang (averaging 2 cm in length) and between Oreochromis aurea and O. niloticus (averaging 1-2 cm in length). As well, crucian carp (crosses between Carassius auratus Gibelio (Bloch) and Carassius curierit, (averaging 4 cm in length) came from the Wuxi Aquatic Products Breeding Farm in Jiangsu Province. The fish were caught on 6-7 November. The length and weight of some of the fish were recorded. Others were released into an aluminum tub (without feed or changes in water) to test their physical ability to withstand adversity.

Experimental Sites

Experiments were conducted simultaneously in Wuxi and in Anhui Province. In Wuxi, deep open pits were rebuilt with clay into two experimental plots (pH 8). An adjacent stream served as the water source. One plot was 0.07 ha; the other 0.03 ha. Each could hold water more than 1-m deep. Before the rice seedlings were transplanted, one pit was pumped dry, and 1000 kg of mud were removed and spread on the bottom of the experimental plot to increase fertility and stabilize the plants. Then 735 kg of barnyard manure (496 kg pig dung and 240 kg poultry dung) were added as base manure. After the seedlings were transplanted, 5 kg of urea were applied as manure to stimulate the rice seedlings to turn green and another 5 kg of urea, 3 kg of calcium superphosphate, and 5 kg of plant ashes were added as manure to stimulate booting. Before the autumn equinox, 50 kg of lime were spread to prevent and control fish diseases. Later, 60 ml of 25% sumithion dissolved in 60 L of water were sprayed to control stem borer, and 25 g of 50% thiophnate were applied to control green smut.

It was difficult to raise the water level steadily, but water depth and temperature were recorded daily, and the heights of 10 rice plants for each rice variety were measured at random every 10 days. The degree days during the initial booting stage for 10 varieties and the degree days during the earing stage for two mature varieties were determined from the mean daily temperature readings issued by the Wuxi City Weather Station.

On 5 November, before the harvest, varieties 1 and 2 ripened and birds began to eat the ears of rice. Twelve sample plants were collected at random to record effective tillering. The 1000-kernel weight and the length of each ear were recorded for every 20 ears. The earing of each variety was closely scrutinized to avoid underreporting. Plants were sampled randomly to measure the distribution of the degree of earing. The Huayuan Lake experimental zone in Anhui is situated where Fengyang, Jiashan, and Wuhe counties cross on the southern bank of the lower reaches of the Huai He River. The surface of the lake is 4000-8000 ha, the area where water rises and falls covers more then 1000 ha, and the depth of water is 1.6-4.9 m. Mean annual water temperature is 17.2°C; during May-September the water temperature is 18-30°C. Mean air temperature is 14.9°C and there are 212 frost-free days. The shallow areas of the lake are broad, with a slope of about 1:20, and the lake is rich in humic substances.


Growth of Fish

A total of 1264 fish were caught (average 300 kg/ha), and the overall survival rate was 39%.

Common carp (Cyprinus carpio). The first catch netted 246 fish weighing a total of 2.89 kg, with a survival rate of 24.6%. Body length ranged from 6.9 to 23 cm (a difference of more than three times) and weight ranged from 6.5 to 264 g (a difference of 40.6 times). The fish seemed physically strong; 30-40 days into the experiment, and they were still active.

Crucian carp (Carassius carassius). The first catch netted 682 fish weighing a total of 9.23 kg. Survival rate was 56%. Body length was 9.5-12.4 cm, and weight 10-24 g. The average weight ranged from 12 g to 15.3 g. Growth was uniform, and the fish were strong and healthy, except those grown without rice in the 0.03-ha pond. The smallest and the largest fish withstood adversity-resistance tests for 43 days and still swam actively (water temperature was 7.3 C).

Tilapia (Oreochromis spp.). The first catch netted 336 fish weighing a total of 7.85 kg. Survival rate was 33.6%. Body length was 9.8-12 cm and weight 14-28 g (70% were 10-11 cm in body length and 19.5-23.5 g in weight). Length growth was uniform, but because of infertility of the water, body weights were not uniform. The weight difference among those with 9-cm bodies was as much as 6 g; among those with 11.4-11.5-cm bodies, the difference was as much as 4.5 g. Some fish measured 9.5 cm in body length, but weighed only 23.5 g; others measured 9 cm in body length, but weighed 28 g.

Ripening of the Deepwater Rice

Two varieties ripened and bore fruit: IR40992-1-3-2-1-1-2 865020 (no. 1) and IR40992-1-3-2-1-3-3 865022 (no. 2). Their productive properties are shown in Table 1. Part of the grains of variety IR41125-7-3-2-2-3-3 865026 (no. 3) were at the milk stage and one-third of the plants showed a 100% heading rate. The varieties that approached the milk stage were IR23426-RR (no. 22) and IR41132-R-27-1-1 (no. 4). No. 22 was 50% better than no. 4 in terms of heading, and one-third reached the late heading period (>80%). No. 4 only approached the middle and late heading period. Eleven other varieties showed heads, but not on all plants. The other varieties either just entered the heading stage (e.g., no. 13, 7, and 9) or showed heads, which disappeared before the harvest (e.g., no. 14 and 20).

The height of 23 varieties (1.4-2.1 m) exceeded previous records. Water depth in the experimental ponds was usually 40-60 cm, but exceeded 60 cm for one-fourth to one-third of the days. Booting and earing occurred when the water was 65-80 cm deep. Seven varieties grew in water more than 2-m deep and only no. 15 did not show heads. Two varieties grew in water less than 1.5 m and all showed heads. In either case, the highest and the lowest, there were varieties that showed heads in large tracts: no. 4 plant was 2.0-m high, and no. 22 was 1.4-m high. The stage of heading in the 1.8-m high mature varieties varied: some of the plants of variety no. 3 were in the milk stage, no. 9 and 7 showed heads to varying degrees, and two varieties did not show heads. Ears in floating varieties (no. 8 and 17), which were 1.9-m and 1.7-m high, respectively, were rare.

Prevention and Control of Pests

Fish, rice, frogs, and spiders lived together in the ponds. The experiment explored ways of combining the use of pesticides with biological methods of pest control.

Pests. Rice plants infested by stem borer had the symptom of white ears. Although road lamps on the side of the ponds attracted some borers, it was difficult to kill all of them. Where the stem borer invaded, the ear stem was higher than the water surface. The depth of water (50-60 cm during the earing stage) was far from being completely used by the fish. Furthermore, pesticides were not very effective. It is advisable to apply pesticides before stem borers invade the ear stem.

Plant disease. Green smut infested the rice in spots, often on ripened rice, but the spread was limited.

Birds. There were no ricefields around the experimental plots and flocks of birds were spotted only in nearby woods. No birds were seen feeding on the rice during the ripening stage until three short-grained rice plants on the edge of the plot were eaten off.

Growth of Deepwater Rice

Growth and survival rates of rice in Huayan Lake in water 10-cm deep were better than in water over 25-cm deep. Seedling growth was not apparent during the first few days after transplanting. New shoots were not visible until a week later; then growth was 1.0-1.5 cm/day and, after the tenth day, about 2 cm/day. When the rice plants had grown to 45 cm, they were already submerged in water. Seedlings 3 cm above the water survived and grew well if the water did not submerge a third or half of the plant. Twelve hours after transplanting, one or two new roots were visible; these proliferated after a week. Although the seedlings were placed in a dry place for more than 30 h before transplanting, survival rate was 85%.


Of the 23 varieties improved by IRRI, only seven (30.4%) showed unripened ears and 16 (69.6%) showed apparent heading. Except for a few which showed booting, most had more than 50% of heads showing. The booting stage of 10 varieties lasted 39 days (21 August - 29 September). During the 32 days starting from 28 August, full heading was seen in some plants and large tracts of heads were seen in some varieties. But from 29 August to 3 September, only varieties no. 1 and 2 showed mature heads (hard doughed ears). The booting stage for no. 22, 3, and 4, which showed no milking or did not fully mature (hollow or immature ears), started in mid- and late-September. This was related to the low temperature (<20°C) at the time. Varieties no. 7 and 14 entered the booting stage a week earlier than no. 1 and 2, but mature ears were rare and heads were found on less than 80% of the plants. This may have been caused by high temperatures (>35°C) for 3 days in mid-August, which affected follow-up growth.

In other immature varieties, analysis of the distribution of the number of heads showing revealed that some were affected by degree days. For example, varieties no. 9, 7, 13, and 3 showed a large difference in degree days, with the range of difference decreasing from large to small (81%, 76%, and 39%). The plants needed more degree days. Some had immature ears. Variety no. 6 had a head showing of 65-97%; whereas, no. 22 had 53-86%. Variety no. 22 entered the booting stage 10 days later than no. 6, but it was able to reach the earing stage in large expanses and mature more quickly than no. 6. The booting period does not necessarily determine the maturity of ears. Varieties 6 and 13 entered the booting stage on the same day, but the head showing of no. 13 was no more than 39%, which was lower than the lowest (65%) of no. 6. If no. 6 had been transplanted earlier, it might have shown the same general maturity as no. 2.

Pool Experiment

On 1 July, 12 days after transplanting, and when the plants were less than 50-cm high, the water rose nearly 30 cm in one week. After that, the water level rose and fell alternately until 1 August when it was nearly 60-cm deep on two occasions. It dropped to about 40 cm 43 days after transplanting, and most plants were more than 100-cm high. Therefore, the plants could withstand the water when it rose later to more than 60 cm. During this period of nearly one month, plant growth experienced two cycles, one from fast to slow (50-60 days after transplanting and at 70-80 days), the other from slow to fast (60-70 days after transplanting and at 80-90 days), that corresponded to the rise and fall of the water level. This shows that the water level had some impact on the growth of the rice plants. When a plant is about 50-cm high, it can resist submergence; as the water rises, the plant is able to continue to grow. Before Huayuan Lake was flooded, the plants were 45-cm tall and had already acquired the ability to survive submergence. However, if the water had risen too fast and too high, the plants would have died. They would not have been able to grow quickly enough to rise with the water level.

The deepwater-rice seedlings showed new roots less than 12 h after transplanting. It may be possible to replace transplanting with a new planting method that exploits the rapid root development of seedlings. When the seedlings turned green, the root system began to proliferate. After 10 days, plant growth doubled. The fertile soil in the lake, the density of transplanted seedlings and their accelerated growth during the later stages, built up their capacity to resist flooding. Because the water was shallow, the plants could have drooped as they grew taller, but they withstood the sharp rise in flood water without damage. However, it is important to prevent the plants from being submerged until they have grown tall. A certain time is needed between the growth of the rice plants and flooding.

Although the common carp grew unevenly, those in the experimental plots had reached a fairly high level of growth. In the comparative ponds where fish grew in a natural environment, maximum weight was 169 g, with 19 cm body length and 23 cm total length. The corresponding common carp (18.5 cm, 22.6 cm) grown in experimental plots with deepwater rice, weighed 191 g. The largest was 264 g, with 23-cm body length and 27 cm head-to-tail length. There were apparent differences between the least-developed common carp in the two ponds (5.7-cm in body length and 6.8-8.0 cm in head-to-tail length). Before the fish were caught, living organisms (144/m²) at the bottom of the lake were collected (weight 96.2 g). These included annelids, mosquito larvae and silk earthworms, and tulip shells (data collected by Mr Chen Wenhai). Because rice was planted, the bottom organisms had a rich supply of food. But because the pond leaked, water had to be constantly replenished, which reduced the level of fertilization. This change corresponded to good growth of bottom-living common carp and crucian cap, and poor growth of tilapia.

Rice planted in a shallow lake increases the productive surface area, oxygen, and food. These factors promote the growth of common carp. Because the volume of water is greater than in rice paddies, it is suitable for raising marketable common carp. Crossbred crucian carp, like the silver crucian carp in Northeast China, are large (up to 3 kg each), and when combined with the planting of deepwater rice, will help raise the harvest. The planting of deepwater rice will help create conditions that will turn lakes into highly efficient fishery bases for a variety of fish species. The shallow waters of lakes provided good fertility, temperature, and sunshine. The open expanses of water are deep and can hold large numbers of fish. The development of such resources will help increase fish harvests from lakes.

It is not possible to establish truly comparative conditions in the experimental ponds; therefore, data are lacking to substantiate the benefits of growing rice and fish together. The experimental ponds contained more water and received 2-3.4 times more fingerlings than flooded ricefields. The depth of water could not be kept at 60 cm, and therefore the volume of water per fish was lower than in the flooded ricefields. Per-unit output remained at about 20 kg, the same level of production as fish culture in flooded ricefields. Furthermore, the fingerlings were released late, they were small in size (the survival rate of winter fish may be higher), and water quality was poor. The fish were not fed in July-August, their peak growing season. The water levels changed quickly and there were potential natural enemies. All these factors limited the fish catch and the benefits from rice-fish cultivation.


Two of the deepwater rice varieties developed by IRRI grew to maturity in the catchment area of the Yangtze River in the Taihu Lake area. Rice-fish culture in this area produced 3750 kg/ha of rice and 350 kg/ha of fish, which indicates there are prospects for developing rice-fish culture in the shallow waters of lakes. Although per-unit fish catch in the laboratory was the same as in ricefields, the fish were strong, and the crucian carp were uniform in size. Common carp were not uniform in size, but improved fertility at the lake bottom might produce common carp of a higher quality. The short-term trial production of rice-fish culture in the shallow waters of Huayuan Lake shows that the growth of rice plants and the proliferation of new roots is as good as in flooded ricefields. Despite incomplete results, the comparison of rice and fish production in both enclosed ponds and lake shallows suggests that rice-fish culture is feasible in shallow areas of lakes.

Wan Qianlin, Li Kangmin, Li Peizhen, Gu Huiying, and Zhou Xin are with the Freshwater Fisheries Research Centre, Chinese Academy of Fisheries Science, Wuxi, Jiangsu Province.

Ability of Fish to Control Rice Diseases, Pests, and Weeds

Yu Shui Yan, Wu Wen Shang, Wei Hai Fu, Ke Dao An, Xu Jian Rong, and Wu Quing Zhai

From 1985 to 1987, a series of tests on rice-fish culture were conducted in Shangyu, Xiaoshan and Huangyan Counties, Zhejiang Province, to determine if grass carp, common carp, and nile tilapia could be used as biological control agents in ricefields.

Materials and Methods

Field Arrangements

A suitable ricefield was selected as the test plot. Before the field was ploughed, the border dikes were raised to 40 cm, and earth dams were built to separate the different test plots. Space for fish ditches and pits was left when the seedling were transplanted. After the rice was transplanted, fish ditches and pits were dug and a fish screen was installed at the outlet.

Selection of Fry and Breeding Fish

Fry that had a body length of 3-4 cm and were able to swim against the current were selected. Healthy, strong fish with shining body colour, no injuries, and a body weight of 20-50 g were chosen as breeding fish.

Rice Cultivation

Fish were stocked in the test plot 7 days after the early rice was transplanted. The fish were moved into the fish ponds before the early rice was harvested and returned to the test plot after the late rice was transplanted. Fish were harvested just before the late rice was harvested. In the test plot, water was kept at a depth of about 10 cm, and basal fertilizers and top dressing were used. No pesticides, seed treatments, or herbicides were used.

Fish-Farming Treatments

Six different treatments were carried out in Shangyu County in 1986, and each treatment was repeated three times. The treatments were: (1) grass carp (5250/ha), (2) common carp (5250/ha), (3) nile tilapia (5250/ha), (4) grass carp (600/ha) and common carp (3000/ha) together, (5) long-term deep-watering for fish-farming, and (6) normal water irrigation for fish-farming. In 1987, three different species of fish (grass carp, common carp, and nile tilapia) were raised. For polyculture, 1500 fish of each species were used per hectare. For pure cultures 4500 fish were raised per hectare. The control plot was the same as in 1986. No comparisons between fish and no-fish plots were made in Xiaoshan and Huangyan.

Survey Methods

Twenty clumps in the small plots and 50-100 clumps in the large plots were inspected for rice planthoppers, 500 clumps were examined for rice borer, 40 clumps in the small plots and 100 clumps in the large plots were examined for rice leafrollers, and 50 clumps in the small plots and 200 clumps in the large plots were examined for rice sheath and culm blight. Weeds were sampled at five locations (0.11 m² each) in the ricefield, and records were kept of variety and fresh weight.

Results and Analysis

Control of Diseases, Pests, and Weeds

Control of rice planthoppers. The raising of fry in early ricefields provided poor control of rice planthoppers; however, raising breeding fish provided good control. As the fish grew in size during the growing period of late rice, they provided good control of rice planthoppers. A survey on 10 July 1985 in Xiaoshan showed that there were 1900 rice planthoppers (third generation) per 100 clumps in early ricefields with fry, a decrease of 34.5% compared with 2900 rice planthoppers per 100 clumps in the early ricefield without fish. In ricefields with breeding fish there were 1030 rice planthoppers per 100 clumps in the early ricefields, a decrease of 64.5% compared with ricefields without fish. A survey on 20 September 1985 indicated that there were 2410 rice planthoppers (fifth generation) per 100 clumps in the late ricefield with fish, a decrease of 734.3% compared with fields without fish.

Different varieties of fish had different effects. Investigations from 1986 to 1987 at the test plot in Shangyu showed that there was a relatively large difference (Table 1) in the control of rice planthoppers during the peak season. The ricefields in which only grass carp were raised showed the best control of rice planthoppers. The fifth and sixth generation planthoppers were decreased by 40.5% and 74.2% in 1986 and by 57.3% and 59.4% in 1987, respectively, compared with the control field. Polyculture of the three varieties of fish was less effective than pure cultures of grass carp.

There were several reasons that the fish could effectively control rice planthoppers. First, the rice planthoppers normally oviposit on plant leaves near the bottom of the plant. The grass carp consume these outer leaves, which controls hatching of ova. Second, fish eat rice planthoppers that fall into the water, which directly reduces the number of insects in the field. Third, the deep water protects parts of the rice plant on which rice planthoppers oviposit and feed. In 1987, there was a severe attack on late rice by the brown-back rice planthopper (Nilaparvata lugens). Some plants were infected with sheath and culm blight in normal and deep-water ricefields without fish; however, in the ricefields with fish, the rice plants continued to have green stems late in the growing season, and the crop ripened without the use of any pesticides. The plots with grass carp produced the highest yields.

Control of rice stem-borer. Observations in Shangyu from 1986 to 1987 indicated that there were, on average, 1980 rice borers per hectare in ricefields with fish. This was a decrease of 51.1% compared with normal-water ricefield without fish, and a decrease of 47.2% compared with deep-water ricefield without fish. A survey on 1 July 1987 showed that the attacked rate per plant was 0.7% in ricefield with fish. This was a reduction of 44.3% compared with normal-water ricefields without fish and a reduction of 27.7% compared with deep-water ricefields.

There was a significant difference in the number of pests per hectare and the attacked rate per plant over the 2-year period in ricefields with fish compared with normal-water ricefields without fish. However, the differences were not significant compared with deep-water ricefields without fish. Observations in Xiaoshan showed that the attacked rate per plant was 0.7% in the ricefield with adult fish, a decrease of 80.6% compared with ricefields without fish. The main reasons that the fish were able to mitigate harm done by rice borer were that the pests were eaten when they fell into the water. In addition, grass carp eat the rice borers when they strip the lower leaves, which are often attacked by rice borers.

Control of rice leafrollers. A survey in Xiaoshan on 8 July 1986 showed that there were 90.5 rice leafrollers (second generation) per 100 clumps in the ricefields with fish. This was a 6.5-times increase compared with 12 rice leafrollers in the ricefield without fish. The survey on 22 September 1987 indicated that there were 15.4 rice leafrollers (fourth generation) per 100 clumps in the late ricefields with fish, an increase of 1.9 times compared with the ricefields without fish. In Shangyu, a survey on 10 September 1987 indicated that the highest number of larva of rice leafroller (fourth generation) per 100 clumps was observed in the late ricefields with grass carp. There were 234 larva with grass carp and 193 larva with polyculture. This was an increase of 57.2% and 29.5%, respectively, compared with the 149 larva in the control plot. The reasons that fish-farming increased the number of rice leafrollers in the ricefields were that the fish did not eat the larvae. As well, the great amount of fish waste, the tender, green rice plants, the deep water in the field, and high humidity in the microclimate all favoured oviposition, hatching, and feeding of larvae.

Control of rice sheath blight and clump blight. From 1986 to 1987 in Shangyu, plant morbidity and disease incidence were reduced in the early ricefields with fish. This was highly significant compared with normal-water ricefields without fish, but there was no significant difference compared with the deep-water control. Disease incidence declined by 9.9-19.6%. Plant morbidity in the late ricefields with fish was 13%, a decrease of 58.4% compared with the normal-water control and of 24.9% compared with the deep-water control (Table 2). There were several reasons why damage from rice sheath blight and culm blight were mitigated. First, the fish (mainly grass carp) stripped the diseased leaves near the bottom of the rice plant, which directly diminished sources for reinfection in the field. Second, after the bottom leaves of plants were ripped off, the microclimate in the field was unfavourable to infection because ventilation and light penetration were improved. Third, long-term deep-water conditions prevented germination of the spores and reinfection.

Control of weeds. Fish control weeds grown in ricefields. Grass carp eat 21 different species of weeds in 16 families (e.g., Echinochloa crusgalli, Eleocharis yokoscensis, Cyperus difformis, Rotala indica, Sagittaria pygmaea, Monochria vaginalis, and Marsilea quadrifolia). In addition, common carp eat young roots, buds, and underground stems of weeds in the ricefield. Observations in late ricefields in Xiaoxhan in 1987 showed that there were three different kinds of weeds in rice-fish fields without weeding. The fresh weight of the weeds was 117 kg/ha. This represented a decrease of one kind of weed and 29.7% in fresh compared with ricefields with manual weeding, and a decrease of six kinds of weeds and 97.2% in fresh weight compared with a ricefield without fish and without weeding.

Only Echinochloa crusgalli, Paspalum distichum, and Alternanthera philoxeroides survived in the rice-fish fields. Echinochloa crusgalli was transplanted in the field with the rice seedlings and the fish were unable to control it effectively. Paspalum distichum normally extended from border dikes into the ricefield. Young buds and stems of Alternanthera philoxeroides were eaten by the fish, but they were not well liked. The surface of the ricefields with fish was smooth and grass-free. Weed control was more effective than with either manual weeding or the use of herbicides.

Economic and Ecological Benefits

Economic benefits. In Shangyu in 1987, rice yields in both early and late ricefields with fish reached 11 093 kg/ha, an increase of 12.9% compared with the control (9821 kg/ha). With a fish yield of 600 kg/ha, the net increase in income was CNY 1986/ha. In Huangyan in 1986, early and late ricefields with fish produced 11 334 kg/ha of rice and 1778 kg/ha of fish with a net income of CNY 9230/ha. This was an increase of 25.7% in income compared with the control. The practice of rice-fish farming has been popularized to 16 650-20 000 ha in Zhejiang Province, and to a total of over 135 000 ha throughout China.

Ecological benefits. The interactions of fish and rice create changes in the ecology of the ricefield. The ricefields hold water all season because of the fish. The movements of the fish stir the soil, which plays a role in weeding the field and increases the dissolved oxygen content in the soil. Ventilation and light penetration are also improved. The fish eat weeds, damaged plant leaves, and some pests. In exchange, they discharge wastes that add organic manure to the ricefield. Experiments have shown that 3000 grass carp per hectare (6.3-11.2 cm in length) discharge 1440 kg of waste in a month. Therefore, they contribute a constant supply of fertilizer to the rice plants. Rice-fish fields require less farm chemicals. This diminishes problems related to pollution and toxic residues and to some extent, protects natural enemies of plant pests and increases their effectiveness for biological control.

Issues and Discussion

Rice-fish farming still faces some problems. First, animals pests (e.g., rats and snakes) eat fish in ricefields and frogs eat fry. Surveys in Shangyu in 1986 indicated that the harvest rates for grass carp, common carp, and nile tilapia were 78.9%, 82.1%, and 92.3%, respectively. Second, the fish do not control Echinochloa crusgalli; therefore, herbicides were required. Third, large grass carp stocked with late rice may damage the ricefield. Fourth, the proper ratio of fish pits and fish ditches in the ricefield must be determined to achieve bumper harvests for both rice and fish.

Yu Shui Yan and Wu Wen Shang are with the Agricultural Department of Zhejiang Province; Wei Hai Fu and Ke Dao An are with the Agricultural Bureau of Shangyu County; Xu Jian Rong is with the Agricultural Bureau of Huangyuan County; and Wu Quing Zhai is with the Agricultural Bureau of Xiaoshan County, Zhejiang Province.

Distribution and Residue of Methamidophos in a Rice-Azolla-Fish Ecosystem

Xu Yinliang, Xu Yong, and Chen Defu

Methamidophos (O,S-dimethyl phosphoramidothioate) is an organophosphorus insecticide that is used in great quantities in ricefields in China. Rice-fish culture is common in the southern parts of China as well as in many other rice-producing countries (e.g., Thailand, Malaysia, and the Philippines). It is necessary to apply pesticides to control rice pests in the rice-azolla-fish ecosystem. In this study, the distribution, degradation, and residual behaviour of methamidophos were measured in both simulated and natural rice-azolla-fish ecosystems.

Materials and Methods

Test Materials

Two rice varieties were used: Sifu 8512 (an early rice) and Xinshui 04 (a late rice). The soil was a silty loam with a pH of 7.1 and an organic matter content of 1.3%. The fish species used was nile tilapia (Oreochromis niloticus) that were 4-6 cm in length, and the azolla was Azolla caroliniana.

A 50% methamidophos emulsion was obtained from the Hangzhou Pesticide Factory. The Institute of Nuclear-Agricultural Sciences, Zhejiang Agricultural University, synthesized the 35S-methamidophos (specific activity 5227 dpm/mg, radio chemical purity 99%). The 50% 35S-methamidophos emulsion was formulated in 59% 35S-methamidophos, 47% methanol, and 3% emulsifier (PP2).

Trial Design

Simulated ecosystem. This part of the study was conducted under outdoor conditions. A glass aquarium (95 cm x 68 cm x 45 cm) held the soil (200 kg), rice plants (30 hills), fish (30), and azolla. Surface water was maintained at a depth of about 7-10 cm.

The trial was divided into two treatments. The first was one application on early rice and two applications on late rice, the second was two applications on early rice and three applications on late rice. The application rate was 0.75 kg/ha. The interval between the latest application and harvest was 35 days for early rice and 30 days for late rice. Samples of water, soil, rice plants, azolla, and fish were taken for analysis of 35S-methamidophos residue at different times. After the harvest of early and late rice, the residues of 35S-methamidophos in different parts of the ecosystem were analyzed.

Natural rice-azolla-fish ecosystem. The experiments were carried out in experimental plots that were each 430 m². The trial design was similar to the simulated rice-azolla-fish ecosystem (Table 1). All treatments were replicated three times.

Sample Preparation and Measurements

Prior to fortification with 35S-methamidophos to give concentrations of either 0.5 or 0.1 ppm, plant tissues were macerated, soil was placed in Buchner funnels and stripped of excess water by aspiration, and whole fish (one per sample) were weighted and cut into small pieces.

Extraction and clean-up. Water samples (20 mL) were extracted with ethyl acetate three times (30 mL, 20 mL, 20 mL). The Na2SO4 was added before the first extraction. The straw, azolla, and fish extracts were filtered through a layer of activated carbon (Celite 545, 1:4, W/W) in Buchner funnels. All the fractions from the sample were combined in 250-mL round-bottom flasks, evaporated in rotary evaporator, and subjected to radioanalysis (Table 2).

The soil (5 g) was extracted with water and methanol (3:1, V/V, 40 mL). The extract was filtered through a glass funnel. The crude extract (equivalent to 2.5 g of soil) was analyzed following the analytical method used for the water. Husk and brown rice (10 g) were also extracted with methanol and ethyl acetate (1:4, V/V, 50 mL) using the method of vibrating extraction. Extract, equivalent to 5 g of husk or brown rice, was evaporated in a rotary evaporator and subjected to radioanalysis.

Radioanalysis. The radioactivity in the organosoluble fractions from extraction of the water, soil, straw, fish, azolla, husks, and brown rice was directly determined using a liquid scintillation counter (Model LKB-1217). Recoveries of 35S-methamidophos from the seven substrates are listed in Table 3.

Gas chromatography analysis. The residues of methamidophos in the samples from the field trial were detected by gas chromatography (GC) using AFID (alkali flame ionization detection). The operation parameters were: gas chromatography Perkin-Elmer Sigma 2000, column 120 cm x 2 mm ID, 2% Reoplex 400 on the gas chromQ, air 18 psi, H2 18 psi, N2 42 mL/min, column temperature 170°C, and inlet and outlet temperature 210°C. Quantification of methamidophos was based on the average peak heights of external standards that were injected before and after sample analysis. The GC recoveries of methamidophos from the seven substrates are listed in Table 4.

Residue in the simulated rice-azolla-fish ecosystem. After application of 50% 35S-methamidophos emulsion to the rice plants in the rice-azolla-fish ecosystem at the rate of 0.75 kg/ha, samples of water, soil, straw, azolla, and fish were taken for analysis of methamidophos residue at different times from immediately after application to 14 days. The dynamics of 35S-methamidophos residue are listed in Table 5. The 35S-methamidophos disappeared exponentially from the simulated rice-azolla-fish ecosystem.

After 1 day of equilibrium, the residues of 35S-methamidophos in the azolla, water, and fish reached a peak (4.16, 0.30, and 0.61 ppm, respectively) then decreased exponentially. The half-lives of 35S-methamidophos in azolla, water, and fish were 1.5, 2.8, and 6.2 days, respectively. The half-life of 35S-methamidophos in the straw was 2.2 days. There was no residue detected in the soil.

Residue in the simulated rice-azolla-fish ecosystem. The 35S-methamidophos was applied to the early and late rice in the simulated ecosystem. The residues of 35S-methamidophos in the different parts of the ecosystem at harvest are listed in Table 6.

The interval between the last application and harvest was 35 days for early rice and 30 days for late rice. At harvest of early and late rice, there were trace residues of 35S-methamidophos in the water, but no residues in the soil of the azolla. The residues of 35S-methamidophos in fish were 0.41-0.55 ppm when harvested with early rice and 0.21-0.40 ppm when harvested with late rice. This decrease of concentration was related to the increase in the weight of the fish.

The distributions 35S-methamidophos residue in the parts of the rice plant differed. The residue was highest in the husk and lowest in the brown rice (e.g., for late rice the concentrations were 0.35 ppm in the husk, 0.12 ppm in the straw, and 0.04 ppm in the brown rice). With increases in application times, the concentrations of 35S-methamidophos residues in the rice plant increased. For example, in husks of early rice, the residue was 0.01 ppm for one application and 0.03 ppm for two applications. However, there was no linear correlation to the increase.

In the natural rice-azolla-fish ecosystem, there was only 0.09-0.10 ppm of methamidophos residue in the rice husks when the early rice was harvested. In the other parts of the ecosystem, no residues were detected.

Influence of polishing brown rice. Brown rice (20 g) that had a residue of 0.01 ppm was polished for 10 minutes. The residue of methamidophos in polished rice was decreased to 0.005 ppm (Table 7).

Residue in fish. When applied twice to early rice and three times to late rice, the distribution and residue of 35S-methamidophos in fish when late rice was harvested were determined (Table 8). The residue was highest in the viscera and lowest in the scales. The edible portion of the fish body contained 26.9% of the radioactivity and the nonedible portion contained 73.1%.


Methamidophos is not well degraded by animals and plants. In female houseflies with different doses of 32P-methamidophos, the parent compound has only been detected in the extract of the houseflies, not in their metabolites or degradation products. Experiments with white mice and houseflies have also suggested that the metabolic pathway of methamidophos may be through the release of CO2. However, the quantity of this metabolic product may be so low that it cannot be detected. When extracts from rice plants and fish were made, separated on thin-layer chromatography, and scanned with a radioscanner, only the parent compound was detected.

These experiments with the rice-azolla-fish ecosystem have demonstrate that:

· The disappearance of methamidophos in the rice-azolla-fish ecosystem was exponential. The half-lives of 35S-methamidophos in azolla, water, fish, and straw were 1.5, 2.8, 6.2, and 2.2 days, respectively.

· When 2 x 0.75 kg/ha of 35S-methamidophos emulsion were sprayed on early rice there was 0.01 ppm of methamidophos in the brown rice and 0.55 ppm in the fish. When sprayed at 3 x 0.75 kg/ha on late rice, there were 0.04 ppm in the brown rice and 0.40 ppm in the fish.

· When brown rice was polished the concentration of methamidophos residue was reduced by about 50%.

· Methamidophos is not well degraded by animals and plants. No metabolic and degradation products were detected in the extracts from rice plants and fish.

Xu Yinliang is with the Institute of Nuclear=Agricultural Sciences, Zhejiang Agricultural University, Xu Yong is with the Institute for the Control of Agrochemicals, and Chen Defu is with the Institute of Soil Sciences, Zhejiang Academy of Agricultural Sciences, Hangzhou, Zhejiang Province.

Residue and Application of Fenitrothion in a Rice-Fish Culture System

Lou Genlin, Zhang Zhongjun, Wu Gan, Gao Jin, Shen Yuejuan, Xie Zewan, and Deng Hongbing

Fenitrothion (o,o-dimethyl-o-3-methyl-4-nitrobenzene phosphnothiocester) has low mamalian toxicity (white rat oral LC50 490-700 g/kg) and effectively controls stem borers. Fenitrothion is one of twelve pesticides recommended by the Chinese government for priority research and increased production. Pest and disease control is one of the most important measures in rice production. Rice-fish culture is a major component of freshwater fisheries. Pesticides with high efficacy and low residual effects are needed to integrate production and increase yields of both rice and fish. The effects of fenitrothion in rice-fish culture was studied in Sichuan Province from 1984 to 1987.

Test Design

Fenitrothion Residue in Ricefields

Five 30-m² treatment plots were designed to study degradation. Every treatment was divided into subplots with three replicates. Water depth was maintained at 6.6 cm during the experiment. Samples were taken at 2 h, and 1, 3, 5, and 7 days after application of 750 g/ha (50% EC 800 dilution). The residual experiment (Table l) included: using different dosages in ricefields (750-1125 g/ha, 50% EC 500-800 dilution), spraying the pesticide at different times, and multiple applications of fenitrothion.

Fenitrothion Residue in Fish

The designs and treatments were identical to those used to study the effects in ricefields, but the sampling interval was from 2 to 148 h at regular intervals after application of the pesticide. Tests on residue distribution were made in tanks (50 L) of clean oxygenated water.

Each tank contained 200 fish that weighed approximately 0.5 g. The range of test concentrations was 0.05-2 ppm. Samples from internal organs were taken at regular intervals. Benzene was used to extract residual pesticide from the rice and fish. Dichlromethane and trichloromethane were used to extract residues from the soil and water.

Results and Discussion

Degradation of Fenitrothion in Aquatic Ecosystems

Tables 2-4 indicate the degradation rate (%) of fenitrothion in rice and in the environment. Fenitrothion broken down rapidly in rice. The degradation rate was up 50% when used at 750-1125 g/ha (50% EC 500-800 dilution). The residue of fenitrothion in rice was generally less than 0.1 ppm 5 days after application. Straw and rice bran contained less than 25% after only 1 day. Over 90% of the fenitrothion had been broken down after 7 days (Table 2). The residual curve formula was:

yrice = 2.8573e -0.3510x

which indicates that the half-life (HL50) in rice was 2 days. The rate of degradation was also fast in soils and water under similar conditions (Tables 3 and 4). The residue curve formulae were:

ysoil = 2.9404e -0.3470x and

ywater = 2.7897e -0.6053x

which indicated that the half-life in soil was 2 days and that the half-life in water was 1 day.

There was little contamination of the rice when it was sprayed from one to three times. Soil and water were not polluted under these conditions. Under similar dosages and number of sprays, residual levels were higher when the rice was sprayed closer to harvest time. The soil and water were not polluted at any frequency of spraying. There was more residue after application at 1125 g/ha than at 750 g/ha under similar conditions.

Accumulation of Fenitrothion in Fish

The level of fenitrothion in the fish increased as the level decreased in the water. The pesticide in the water was absorbed and concentrated by the fish (Table 6). When the fish were treated with different concentrations of the pesticide, the residual level of the pesticide in the fish increased as the amount of residue in the water increased. However, the coefficient did not increase without limit. After a certain concentration, the coefficient decreased. For example, when grass carp were kept in 0.05 ppm and 0.1 ppm, the coefficients after 2 h were 12.6 and 17.3, respectively. The coefficients increased to 38.3 and 61.4 after 8 h. When the concentration of the pesticide was higher (1 ppm and 2 ppm), the residue in the fish increased to 2.0 ppm and 3.2 ppm, but the coefficients were only 2.5 and 2.1.

When different fish species (Table 6) were exposed to the same concentration, the residue levels were different. Different species of fish have different shapes and feeding habits, fat contents, and distributions. In all species, fenitrothion was more concentrated in fish than in the water. The concentration coefficients were highest at low concentration. For example, in grass carp at application rates of 0.05 ppm, they were 12, 39, and 10 from 2 to 24 h; whereas, at 1-2 ppm the concentration coefficients ranged from 1 to 3. The concentrations were higher for common carp and crucian carp (5.4-7.5).

The amount of fenitrothion (Table 6) in water decreased from 2 to 8 h then increased by 24 h; whereas, in the fish, the amount increased from 2 to 8 h and then decreased. The amount in water probably reflects two processes: the natural breakdown of fenitrothion in water, and the accumulation (2-8 h) and elimination of the pesticide by the fish (24 h).

Table 7 indicates the elimination dynamics of fenitrothion from fish when used at 1125 g/ha (50% EC 800 dilution) and shows a similar pattern to Table 6. The results show that some of the pesticide in the water is absorbed by the fish after application. The fish absorb fenitrothion quickly during the earlier stages, and the concentration of the pesticide decreased in the fish 24 h after application. Other experiments showed no difference between fish raised in ricefields and fish raised indoors (Table 7).

The half-life of fenitrothion was 2 days and the residue curve formula was:

y = 1.6282e -0.3559x (r = 0.9648)

When fish are exposed to fenitrothion, they both absorb and accumulated and degrade and eliminate the chemical. Accumulation and elimination of fenitrothion are kept in a dynamic balance. In a polluted environment, toxic pollutants are concentrated and accumulated by fish. When the fish are in a nonpolluted environment, the chemical is eliminated. Fish are an important part of the food chain. Fenitrothion was accumulated in the polluted ricefield ecosystem. Different fish concentrated the fenitrothion at different rates.

Experiments were also carried out to determine the rate of fish purification. Table 8 shows the results of the purification experiments. Although the pesticide broke down quickly in the fish and the ricefields, these purification measures could speed up the elimination of pesticides from the fish.

Distribution of Fenitrothion

Residual fenitrothion was detected in fish in ricefields in 12 cities and counties in Sichuan Province. Middle-season rice is normally treated with 560-750 g/ha (50% EC 800 dilution) sprayed from one to three times through the season. Three hundred samples of rice and fish were obtained through out Sichuan. The residue level in rice grains was 0-0.027 ppm, the residue level was from trace to 0.077 ppm in rice bran, and the level was 0-0.037 ppm in fish. The concentration of fenitrothion was higher in the viscera than in the meat of the fish (Table 9).


Fenitrothion is a pesticide with high efficacy, low residues, and low toxicity. Contamination should not occur in the aquatic ecosystem of the ricefield when fenitrothion is used. Fenitrothion could be promoted to increase outputs of rice and fish.

The residual standard for fenitrothion in rice grains has been set at 0.2 ppm in China. There is no standard for the maximum residue in fish, but in Japan it is 0.05 ppm. If the pesticide is applied as recommended and not close to the harvest time for the rice and fish, the residues should be well below acceptable levels.

Fenitrothion should not produce lasting accumulations in rice and fish because its degradation rate is fast (LH50 2 days). If the concentration of toxic pollutants is controlled in the water, there should be no long-term contamination of fish.

Lou Genlin, Zhang Zhongjun, Wu Gan, and Gao Jin are with the Institute of Plant Protection, Sichuan Academy of Agricultural Sciences, Chengdu; Shen Yuejuan, Xie Zewan, and Deng Hongbing are with the Sichuan Provincial Bureau of Fisheries, Chengdu, Sichuan Province.