In the previous chapters, the sources of pollution, the general responses of fish to natural and man-made changes in the aquatic environment, and to some extent their interactions, were considered in outline. This chapter, much of it based on Alabaster and Lloyd (1980), considers the important factors involved on an individual basis and in greater detail.
The following physico-chemical changes in the aquatic environment are those most frequently recorded as the primary causes of harm to fish in fish culture installations.
Fish are poikilothermic animals, that is, their body temperature is the same as, or 0.5 to 1°C above or below, the temperature of the water in which they live. The metabolic rate of fish is closely correlated to the water temperature: the higher the water temperature (i.e. the closer to the optimum values within the normal range), the greater the metabolism. This generalisation applies particularly to warm-water fish. Cold-water fish, e.g. salmonids, whitefish, or burbot, have a different type of metabolism: their metabolic rate can continue at comparatively low temperatures, whereas at high water temperatures, usually above 20°C, they become less active and consume less food. Water temperature also has a great influence on the initiation and course of a number of fish diseases. The immune system of the majority of fish species has an optimum performance at water temperatures of about 15°C.
In their natural environment, fish can easily tolerate the seasonal changes in temperature, e.g. a decrease to 0°C in winter and increase to 20–30°C (depending on species) in summer under Central European conditions. However, these changes should not be abrupt; temperature shock occurs if the fish are put into a new environment where the temperature is 12°C colder or warmer (8°C in the case of salmonids) than the original water. Under these conditions fish may die, showing symptoms of paralysis of the respiratory and cardiac muscles. With young fry, problems may arise even where the difference in temperature is as low as 1.5–3°C. If fish are fed and then abruptly transferred to water colder by 8°C or more, their digestive processes will slow down or stop. The food will remain undigested or half-digested in the digestive tract and the gases produced can cause the fish to become bloated, lose balance, and finally die. If carp are given a high-nitrogen feed (e.g. natural food or high-protein pellets), abrupt transfer to much colder water will considerably increase the level of ammonia nitrogen in the blood serum because the decrease in metabolic rate reduces the diffusion of ammonia from the gills. This can lead to ammonia autointoxication and death.
Considerable progress has been made recently in warm water fish culture. Techniques for water temperature control enables optimal condition to be maintained, so that the fish can fully utilize their growth potential to achieve maximum weight gains.
The optimal pH range for fish is from 6.5 to 8.5. Alkaline pH values above 9.2 and acidity below 4.8 can damage and kill salmonids (e.g. brown and rainbow trout); and pH values above 10.8 and below 5.0 may be rapidly fatal to cyprinids (especially carp and tench). Thus salmonids, in comparison with cyprinids, are more vulnerable to high pH and more resistant to low pH. The American char is especially resistant to acid waters and can tolerate pH levels as low as 4.5–5.0.
Low water pH most frequently occurs during the spring, especially when acidified snow melts, and in water draining peat bogs. High alkaline pH can occur in eutrophic reservoirs (ponds) where the green plants (the blue-green algae, green algae and higher aquatic plants) take up considerable amounts of CO2 during the day for intensive photosynthetic activity. This affects the buffering capacity of the water and the pH can rise to 9.0–10.0 or even higher if bicarbonate is adsorbed from waters of medium alkalinity. Water pH can also be changed when mineral acids and hydroxides, or other acidic or alkaline substances, are discharged or leach into water courses, ponds or lakes.
As a defence against the effect of a low or high water pH, fish can produce an increased amount of mucus on the skin and on the inner side of the gill covers. Extremely high or low pH values cause damage to fish tissues, especially the gills, and haemorrhages may occur in the gills and on the lower part of the body. Excess amounts of mucus, often containing blood, can be seen in post mortem examination of the skin and gills. The mucus is dull-coloured and watery.
Water pH also has a significant influence on the toxic action of a number of other substances (e.g. ammonia, hydrogen sulphide, cyanides, and heavy metals) on fish.
Oxygen diffuses into the water from the air especially where the surface is turbulent and also from the photosynthesis of aquatic plants. On the other hand, oxygen is removed by the aerobic degradation of organic substances by bacteria and by the respiration of all the organisms present in the water, as mentioned earlier. The concentration of oxygen dissolved in water can be expressed as mg per litre or as percentage of air saturation value. Water temperature, atmospheric pressure and contents of salts dissolved in water have to be taken into account when the values in mg per litre are converted to % saturation or vice versa.
Different fish species have different requirements for the concentration of oxygen dissolved in water. Salmonids have the more demanding requirements for oxygen in the water; their optimum concentration is 8–10 mg per litre, and if the level declines below 3 mg per litre they begin to show signs of suffocation. Cyprinids are less demanding; they can thrive in water containing 6–8 mg per litre and show signs of suffocation only, when the oxygen concentration falls to 1.5–2.0 mg per litre.
The oxygen requirements of fish also depend on a number of other factors, including the temperature, pH, and CO2 level of the water, and the metabolic rate of the fish. The major criteria for the oxygen requirement of fish include temperature, and the average individual weight and the total weight of fish per unit volume of water. Oxygen requirements increase at a higher temperature (e.g. an increase in water temperature from 10 to 20°C at least doubles the oxygen demand); a higher total weight of fish per unit volume of water can lead to increased activity and thus increased respiration as a result of overcrowding.
Oxygen requirements per unit weight of fish significantly decline with increasing individual weight. In carp this reduction may be expressed by the following ratios: yearling = 1, two-year-old carp = 0.5–0.7, marketable carp = 0.3–0.4. Significant differences in oxygen demand are also found for different species. Using a coefficient of 1 to express the oxygen requirement of common carp, the comparative values for some other species are as follows: trout 2.83, peled 2.20, pike perch 1.76, roach 1.51, sturgeon 1.50, perch 1.46, bream 1.41, pike 1.10, eel 0.83, and tench 0.83.
As stated earlier, the factor most frequently responsible for a significant reduction in the oxygen concentration of the water (oxygen deficiency1) is pollution by biodegradable organic substances (including waste waters from agriculture, the food industry, and public sewage). These substances are decomposed by bacteria which use oxygen from the water for this process. A few chemicals may be oxidized in the absence of bacteria. The concentration of organic substances in water in terms of their capacity for taking oxygen from the water can be measured by means of the chemical oxygen demand (COD, which represents a theoretical maximum) and the biochemical oxygen demand within five days (BOD5, which represents the potential for bacterial degradation). The upper limit of COD, as determined by the Kubela method, for the optimal range for cyprinids in pond or river waters, is 20–30 mg O2 per litre and the corresponding BOD5 limit for cyprinids is 8–15 mg O2 per litre, depending on the intensity of the culture and the rates of reaeration. For salmonids the corresponding levels are up to 10 mg O2 per litre for COD and up to 5 mg O2 per litre for BOD.
In winter, fish are commonly killed by suffocation in polluted storage ponds and in summer this often happens in polluted water courses with high temperatures and low flow rates. In severely eutrophicated ponds, oxygen deficiency often occurs during the summer early in morning as a result of the night time oxygen consumption by bacteria for the decomposition of organic substances and the respiration of aquatic plants. In heavily fertilized ponds (e.g. those used for the treatment of sewage) with a constant inflow of degradable organic substances, oxygen deficiency can also be caused by an excessive development of zooplankton; the zooplankton itself requires oxygen for respiration and, in addition, its feeding pressure reduces the phytoplankton population which produces oxygen during the day.
Even in ponds where the oxygen levels have been satisfactory during the summer, when plant growth was vigorous, severe oxygen deficiencies can occur in the autumn when the plants begin to die and decompose. This deficiency can be more pronounced if the sky is heavily overcast during the day, so that the limited oxygen production by photosynthesis is further reduced. In these cases, the maximum oxygen deficiency occurs just before daybreak.
In summary, the oxygen levels in water depend on the balance between the inputs from the air and plants, and the consumption by all forms of life. Inputs from the air depend on the turbulence of the air-water interface, and the oxygen deficiency of the water. Inputs from plants depend on photosynthetic activity which increases with temperature and sunlight; excess oxygen can be lost to the atmosphere. Oxygen consumption depends on the respiration of aquatic organisms, including plants, and the aerobic decomposition of organic material by bacteria; these rates also increase with temperature.
This balance needs to be clearly understood; a satisfactory oxygen level recorded during the day is no guarantee that the levels will be maintained during the night. Moderate levels recorded in calm eutrophic waters on a warm, sunny afternoon will almost always indicate that severe oxygen deficiencies will occur during the night. Also, lower than expected daytime pH values due to high levels of CO2 may indicate high levels of bacterial respiration which could lead to low night-time oxygen levels.
Oxygen deficiency causes asphyxiation and fish will die, depending on the oxygen requirements of the species and to a lesser extent on their rate of adaptation. Fish exposed to oxygen deficient water do not take food, collect near the water surface, gasp for air (cyprinids), gather at the inflow to ponds where the oxygen levels are higher, become torpid, fail to react to irritation, lose their ability to escape capture and ultimately die. The major pathologico-anatomic changes include a very pale skin colour, congestion of the cyanotic blood in the gills, adherence of the gill lamellae, and small haemorrhages in the front of the ocular cavity and in the skin of the gill covers. In the majority of predatory fishes the mouth gapes spasmodically and the operculum over the gills remains loosely open.
Remedial action is to either reduce the input of degradable material, or to aerate the water. The latter is usually the best option; aeration can be with air or oxygen pumps, or by spraying the water into the air in the form of a fountain, or by increasing the input of aerated water. It must be remembered that these remedial actions are most important at night when the oxygen deficiency is likely to be at its greatest.
Damage caused to fish by too much oxygen dissolved in water is seldom encountered. However, it may happen, for example, when fish are transported in polythene bags with an oxygen-filled air space. The critical oxygen level of water is 250 to 300% of the air saturation value; fish may be injured at higher values. The gills of such affected fish have a conspicous light red colour and the ends of the gill lamellae fray. When such fish are used for stocking waters they may suffer from secondary fungus infections and some of them may die. It is possible that fish adapted to such high oxygen levels need to be progressively acclimatized to more normal concentrations. The condition described here should not be confused with the supersaturation of water with dissolved gas, which can cause gas bubble disease.
1 Oxygen deficiency is the appropriate saturation value minus the actual value in the water.
Supersaturation with dissolved gas occurs when the pressure of the dissolved gas exceeds the atmospheric pressure. It occurs when water is equilibrated with air under pressure, e.g. at the bottom of a lake or reservoir, in ground water, or if air is drawn into a centrifugal water pump. It can also occur if cold air-equilibrated water is warmed up without re-equilibration to the higher temperature. A bottle containing such water will show either minute bubbles forming as a cloudy suspension which will clear from the bottom upwards, or larger bubbles forming on the glass wall. This is analogous to that seen in an opened bottle of carbonated drinking water.
If fish are exposed (at a lower atmospheric pressure) to such water, their blood equilibrates with the excess pressure in the water. Bubbles form in the blood and these can block the capillaries; in sub-acute cases the dorsal and caudal fin can be affected, and bubbles may be visible between the fin rays. The epidermal tissue distal to the occlusions then becomes necrotic and cases are known where the dorsal fins of trout have become completely eroded. In severe cases, death occurs rapidly as a result of blockage of the major arteries, and large bubbles are clearly seen between the rays of all the fins. A similar effect of gas bubbles forming in the blood can be experienced by deep-sea divers when they return to the surface.
The remedy is either to remove the fish to normally equilibrated water or to provide vigorous aeration to strip out the excess gas.
Ammonia pollution of water courses, ponds and lakes may be of organic origin (domestic sewage, agricultural wastes, or the reduction of nitrates and nitrites by bacteria in anoxic waters) or of inorganic origin (industrial effluents from gas works, coking plants and power generator stations). In water or in biological fluids, ammonia is present in a molecular (nondissociated) form (NH3) and in the form of ammonia ion (dissociated) (NH4+). The ratio between these two forms depends on the pH and temperature of the water (Table 1). The cell walls of organisms are comparatively impermeable to the ammonia ion (NH4+), but molecular ammonia (NH3) can readily diffuse across the tissue barriers where a concentration gradient exists, and is therefore the potentially toxic form to fish. Also, under normal conditions there is an acid-base balance at the water-tissue interface. If this balance is altered, the side on which the pH is lower will attract additional molecular ammonia. This explains how molecular ammonia passes from water through the epithelium of the gills to the blood and also how it passes from the blood to the tissues. Ammonia has a particular toxic effect on the brain; this is why nervous symptoms are so pronounced in cases of ammonia toxicity to fish.
Water quality monitoring of water courses, lakes and fish culture facilities includes the measurement of total ammonia concentrations. To assess the potential toxicity of these concentrations it is important to know the amount of nondissociated ammonia (NH3) present. This is calculated from the measured values for total ammonia (NH4++NH3), temperature (T, °C) and water pH, using the formula:
Alternatively, the values can be interpolated from Table 1 compiled from calculations on the basis of this formula.
Table 1: The NH3 content (as % of total ammonia) of water at different pH values and temperature
| pH | T °C | |||||
| 0 | 5 | 10 | 15 | 20 | 25 | |
| 7.0 | 0.082 | 0.12 | 0.175 | 0.26 | 0.37 | 0.55 |
| 7.2 | 0.13 | 0.19 | 0.28 | 0.41 | 0.59 | 0.86 |
| 7.4 | 0.21 | 0.30 | 0.44 | 0.64 | 0.94 | 1.36 |
| 7.6 | 0.33 | 0.48 | 0.69 | 1.01 | 1.47 | 2.14 |
| 7.8 | 0.52 | 0.75 | 1.09 | 1.60 | 2.32 | 3.35 |
| 8.0 | 0.82 | 1.19 | 1.73 | 2.51 | 3.62 | 5.21 |
| 8.2 | 1.29 | 1.87 | 2.71 | 3.91 | 5.62 | 8.01 |
| 8.4 | 2.02 | 2.93 | 4.23 | 6.06 | 8.63 | 12.13 |
| 8.6 | 3.17 | 4.57 | 6.54 | 9.28 | 13.02 | 17.95 |
| 8.8 | 4.93 | 7.05 | 9.98 | 13.95 | 19.17 | 25.75 |
| 9.0 | 7.60 | 10.73 | 14.95 | 20.45 | 27.32 | 35.46 |
| 9.2 | 11.53 | 16.00 | 21.79 | 28.95 | 37.33 | 46.55 |
| 9.4 | 17.12 | 23.19 | 30.36 | 39.23 | 48.56 | 57.99 |
| 9.6 | 24.66 | 32.37 | 41.17 | 50.58 | 59.94 | 68.63 |
| 9.8 | 34.16 | 43.14 | 52.59 | 61.86 | 70.34 | 77.62 |
| 10.0 | 45.12 | 54.59 | 63.74 | 71.99 | 78.98 | 84.60 |
| 10.2 | 56.58 | 65.58 | 73.59 | 80.29 | 85.63 | 89.70 |
| 10.4 | 67.38 | 75.12 | 81.54 | 86.59 | 90.42 | 93.24 |
| 11.0 | 89.16 | 92.32 | 94.62 | 96.26 | 97.41 | 98.21 |
Besides water temperature and pH, other factors that influence ammonia toxicity include the concentration of dissolved oxygen in water; the lower the oxygen concentration in water, the greater the toxicity of ammonia (Fig. 4).
To a lesser extent, the toxicity of ammonia is affected by the amount of free CO2 in the water. This is because the diffusion of respiratory CO2 at the gill surface reduces the pH of the water, thus reducing the proportion of nondissociated ammonia there. The extent of the reduction in pH depends on the amount of CO2 already present in the water.
In general, Table 1 shows that the toxicity of ammonia will be much greater in warm alkaline waters than in cold acid waters.
Non-dissociated ammonia is highly toxic to fish. The LC50 values, determined in acute toxicity tests, are in the range of 1.0 to 1.5 mg NH3 per litre for cyprinid fish and 0.5 to 0.8 mg NH3 per litre for salmonids. The maximum admissible ammonia (NH3) concentration is 0.05 mg per litre for cyprinids and 0.0125 mg per litre for salmonids.
It should be emphasized here that these standards apply to ammonia as a toxic substance. Other standards for total ammonia are applied to control eutrophication of waters and prevent excessive algal and plant growth that can cause physical problems and affect the oxygen balance.
Fig. 4: With low levels of oxygen in the water, lower concentrations non-dissociated ammonia can kill fish:
• fatal cases;
+ waters where no cases of injury to fish occurred in March to April;
hatched area = lethal boundary of non-dissociated ammonia
(Vámos and Szöllözy, 1974)
The first signs of ammonia toxicity include a slight restlessness, and increased respiration; the fish congregate close to the water surface. In later stages, cyprinids gasp for air, their restlessness increases with rapid movements and respiration becomes irregular; then follows a stage of intense activity. Finally, the fish react violently to outside stimuli; they lose their balance, leap out of the water, and their muscles twitch in spasms. Affected fish lie on their side and spasmodically open wide their mouths and gill opercula. Then follows a short period of apparent recovery. The fish return to normal swimming and appear slightly restless. This stage is then replaced by another period of high activity; the body surface becomes pale and the fish die.
The skin of ammonia poisoned fish is light in colour, and covered with a thick or excessive layer of mucus. In some cases small haemorrhages occur, mainly at the base of the pectoral fins and in the anterior part of the ocular cavity. The gills are heavily congested and contain a considerable amount of mucus; fish exposed to high ammonia concentrations may have slight to severe bleeding of the gills. Intense mucus production can be observed on the inner side of the gill opercula, mainly at the posterior end. The organs inside the body cavity are congested and parenchymatous, and show dystrophic changes.
In recent years, considerable losses among farmed carp have been caused by the so-called toxic necrosis of the gills. The factors responsible for the occurence of this disease include ammonia poisoning in which the ammonia level in the blood is considerably increased. As stated earlier, ammonia is the final product of nitrogen metabolism in carp (as it is in other species) and most of it is excreted via the gills into the water. If the diffusion rate is reduced for some reason or another (high water pH, oxygen deficit, damaged gills etc.), the ammonia level in the blood will steadily rise, causing a condition known as autointoxication, which may lead to toxic gill necrosis in carp.
A very interesting case of autointoxication among carp yearlings (C1) where extremely high ammonia N levels were found in the blood serum occurred after their transfer from a pond to well water in large aquarium tanks. Some of the fish caught and transferred during the morning exhibited typical symptoms of ammonia poisoning the following morning. These symptoms included considerable restlessness, increased respiration, leaping out of the water, uncoordinated activity, and tonic-clonic spasms of the muscles. The skin of the affected fish was light in colour; the gills were heavily congested, dark red and showed oedematous swellings (particularly severe on the edges of the gill filaments). It is known that ammonia toxicity is accompanied by an increase in the permeability of the fish epithelium to water, as measured by an increase in the flow of urine. If the kidneys cannot cope with the increased water influx, oedema is likely to occur. An increased water influx may also occur if the skin or the mucus coating of the fish is damaged by handling and during transport. The histopathological changes in the gills corresponded with what had been described for toxic necrosis of carp gills. The digestive tract of those fish with severe poisoning symptoms was filled with undigested food. On the other hand, fish that had cleared their gut (faeces found on the bottom of the tank, the gut almost empty), were free from symptoms of toxic damage. The average blood serum level of ammonia N in the fish with symptoms of poisoning was 3054 (2400–3600) μg per 100 ml of serum, whereas in the fish free of such symptoms the ammonia level was 825 (750–900) μg per 100 ml of serum. In the affected fish the autointoxication, associated with the considerable increase in the blood serum ammonia N level, was probably due to the persistence and absorption of the gut contents (natural food and high-protein feed pellets) of the carp exposed to environmental stress (confinement and reduced oxygen level during transport, and water temperature reduced by about 5°C).
On the basis of this case of ammonia poisoning of carp, some other unexplained incidents of rapid death among fish may be ascribed to a similar cause. Such events may occur mainly in carp farms where there is an intensive feeding with a high-nitrogen diet, if the fish are also exposed to other stresses caused by e.g. an abrupt oxygen deficit, or sudden changes in water temperature.
Toxic gill necrosis was diagnosed in fish from the Dřemliny pond on May 2, 1984; about 50% of two-and three-year-old carp died (Svobodová et al., 1987, Fig. 5). The clinical signs of toxic gill necrosis in carp included the congregation of the fish in the deeper and shaded part of the pond and subsequently, in the advanced stage of disease the body surface darkened and there was a reduced or total absence of the escape response. Respiration was laboured and the fish did not feed.
Pronounced hyperaemia, oedematic swelling and increased accumulation of mucus in the gills are typical features of the patho-anatomic picture. These are followed by a gill necrosis and separation of the epithelium from the gill lamellae. The pillar cells of the gill lamellae are completely exposed over the whole lamellar surface. In the later stages of the disease, necrotic gill lamellae become detached and the margins of the gills are distorted. Histological and pathological examination reveals venostasis, swelling, vacuolization and separation of the respiratory epithelial cells from basal membrane in the gills. Associated with these effects is an increase in the activity of chloride cells in the lamellar epithelium. Dystrophic and necrobiotic cells from the respiratory epithelium (including chloride cells) create a compact mass of debris in the interlamellar space of gills. Extensive effects are characterized by a total lysis and necrotic changes in the cell nucleus. A significant increase in the ammonia level of blood serum in fish is a specific feature of these effects. The normal physiological level of ammonia in the blood serum of carp ranges from 350 to 800 μg N in 100 ml. The ammonia level in the blood serum of carp with toxic gill necrosis fluctuates between 2000 and 4800 μg N per 100 ml, while 1000 to 2500 μg N in 100 ml serum are found in the early stages of toxic necrosis. In other gill diseases that cause necrosis, the following levels of ammonia in blood serum have been found (as N per 100 ml): bacterial diseases 420–900 μg, dactylogyrosis 500–730 μg, branchiomycosis 450–700 μg, and sphaerosporosis 450–960 μg.
The diagnosis of toxic necrosis is based on a detailed examination of fish. The main specific effect in carp is the elevated ammonia level in the blood serum. However, because such toxic gill necrosis can be caused by other unfavourable conditions in the pond environment (Fig. 5), a detailed hydrochemical and hydrobiological analysis of pond water is necessary to provide a definitive diagnosis.
Preventive measures to control frequent outbreaks of gill necrosis in carp in highly eutrophic ponds are centred on optimizing of the hydrobiological and hydrochemical conditions and ensuring the healthy state of fish stock (e.g. by a proper control of the feeding of fish). Stocking the ponds with fish at the correct time in the spring, and preventing or oxygen deficiency, are among the most important preventive measures.
In this context a simple biological test has been developed to determine the optimum timing for the spring stocking of two-year-old carp into ponds with a history of toxic gill necrosis. This test is based on the ability of carp to eliminate ammonia (under the existing physical and chemical conditions of the pond water) given as an oral dose of 350 mg. 100 ml-1 in starch gel. If the ammonia level in the blood serum decreases to the original value within 6 hours of the dose being given, the fish can be stocked in pond. On the other hand, if the ammonia level in the blood serum remains at a threefold higher level than the original value, the stocking of fish must be postponed until the physical and chemical conditions of pond water allow the fish to eliminate the toxic ammonia.
Application of the pesticide Soldep at a rate 200 ml ha-1 (depth of pond l m on average) can ensure the survival of the fish stock when an overproduction of zooplankton, followed by an oxygen deficiency, is expected. Soldep is effective in controlling the daphnid zooplankton and should be applied when there is still a reasonable phytoplankton community in the pond. Both during and after the Soldep application to the pond, standardized safety regulations must be followed (Svobodová and Faina, 1984).
Nitrites as a rule are found together with nitrates and ammonia nitrogen in surface waters but their concentrations are usually low because of their instability. They are readily oxidized to nitrate or reduced to ammonia, both chemically and biochemically by bacteria. Nitrates are the final product of the aerobic decomposition of organic nitrogen compounds. They are present in low concentrations in all surface waters. There is almost no nitrate retention in soil, so it is readily leached to watercourses, ponds and lakes. The main sources of nitrate pollution of surface waters is the use of nitrogenous fertilizers and manures on arable land leading to diffuse inputs, and the discharge of sewage effluents from treatment works.
Nitrite can be associated with ammonia concentrations in the water. In normal aerobic conditions, ammonia is oxidized to nitrite and then to nitrate by two separate bacterial actions. If the second stage of oxidation is inhibited by bactericidal chemicals in the water, nitrite concentrations will increase. This may be important in small ponds or aquaria where water is recirculated through a purification filter; the ammonia-oxidizing bacteria need to become established for the filter to function, and they may be affected by the use of antibiotics to control fish diseases.
The toxic action of nitrite on fish is incompletely known; it depends on a number of internal and external factors (such as fish species and age, and general water quality). The importance and role of these factors have been frequently studied and reviewed. Different authors often come to contradictory conclusions, and usually fail to offer a definitive explanation of either the mechanism of nitrite toxic action on fish or the modifying effects of different environmental factors.
It is now clear that nitrite ions are taken up into the fish by the chloride cells of the gills. In the blood, nitrites become bound to haemoglobin, giving rise to methaemoglobin: this then reduces the oxygen transporting capacity of the blood. The increase in the amount of methaemoglobin can be seen as a brown colour of the blood and gills. If the amount of methaemoglobin in the blood does not exceed 50% of the total haemoglobin, the fish usually survive. If the fish have more methaemoglobin in their blood (70–80%) they become torpid, and with a further increase in the methaemoglobin level they lose their orientation and are unable to react to stimuli. Nevertheless, the fish may still be able to survive because the erythrocytes in their blood contain the enzyme reductase which can convert methaemoglobin to haemoglobin. This process can return the haemoglobin to its normal level within 24–48 hours, if the fish are put into nitrite-free water.
Fig. 5: Hydrobiological and hydrochemical conditions in the Dremliny Pond before and during the course of toxic gill necrosis in carp. Toxic necrosis was diagnosed on 2 May 1984.
Several authors have shown that nitrite toxicity to fish can be affected by certain water quality characteristics (e.g. Lewis and Morris, 1986). In this investigation, the 96h LC50 for rainbow trout ranged from 0.24 to 12.20 mg per litre, depending on the chloride content of the dilution water (in this case the chloride content ranged from 0.35 to 40.9 mg per litre). The effect of chloride on nitrite toxicity is so marked that the results of tests made without recording the chloride concentrations in the water cannot be compared with those of other tests.
It is now known that the chloride cells in the fish gills cannot distinguish between nitrite and chloride ions; both are transported across the gill epithelium. The rate of nitrite uptake depends therefore on the nitrite-chloride ratio in the water.
Nitrite toxicity can be also influenced by bicarbonate, potassium, sodium, calcium and other ions, but their effect is not so great as that of chloride. Among these, potassium is the more significant, and that of sodium and calcium is less. These monovalent ions are also involved in the ionic fluxes across the gill epithelium and so directly or indirectly influence the uptake of nitrite. The pH value has also been considered as important for nitrite toxicity; pH and temperature control dissociation between NO2 and nondissociated HNO2 and it was believed that the uptake of nitrites into fish blood plasma depended on the diffusion of nondissociated HNO2 across the gill epithelium. However, the results of later experiments refuted these theories and showed that within the acidity-alkalinity range encountered in natural waters the pH is of little importance in nitrite toxicity.
Another factor that influences nitrite toxicity is the dissolved oxygen concentration and water temperature. This is associated with the fact that fish need a fully oxygenated water when the oxygen-carrying capacity of the blood is reduced by the formation of methaemoglobin, and the oxygen requirement of fish increases with temperature.
Long exposure to sublethal concentrations of nitrites does not cause much damage to the fish. Concentrations corresponding to 20–40% of the minimum levels having a lethal action on the fish may slightly depress their growth but no serious damage has ever been recorded.
For estimating the safe nitrite concentration for particular locations, it is necessary to measure the ratio of chloride to nitrite. These ratios (expressed as mg l-1 Cl: mg l-1 N-NO2-) are recommended to be no less than 17 for rainbow trout and 8 for fish of low economic importance.
The toxicity of nitrates to fish is very low, and mortalities have only been recorded when concentrations have exceeded 1000 mg per litre; 80 mg per litre is considered to be the maximum admissible nitrate concentration for carp and 20 mg per litre for rainbow trout. In surface waters and in fish farms where the water contains ample oxygen with no danger of denitrification (i.e. conversion of NO3- to NO2- and then to elementary nitrogen or N2O and NO), it is not so necessary to monitor the concentration of nitrates. However, as with ammonia, water quality standards need to be set for nitrate to prevent eutrophication, and the excessive growth of algae and plants, which can have a secondary effect on fish.
Hydrogen sulphide occurs in organically polluted waters from the decomposition of proteins. It is also present in industrial effluents including those from metallurgical and chemical works, paper pulp plants, and tanneries. It has a high to very high toxicity to fish; the lethal concentrations for different fish species range from 0.4 mg H2S per litre (salmonids) to 4 mg per litre (crucian carp, tench and eel). The toxicity of H2S decreases with increasing water pH, because of a reduction in the ratio of the nondissociated toxic H2S to the less toxic HS ions). The concentration of nondissociated H2S can be calculated from the measured total hydrogen sulphide (HS + H2S + S2-) concentration and the pH value of the water, using the formula:
where p = activity coefficient depending on the ionic strength of water. For natural water it is about 0.92.
Hydrogen sulphide can be formed in decomposing rich organic mud, and escapes into the overlying water together with other gases (e.g. methane and carbon dioxide) formed by anaerobic degradation. In aerobic waters the H2S is rapidly oxidized to sulphate; however, it is possible for fish living close to the surface of such muds to be exposed to hydrogen sulphide.
Carbon dioxide is dissolved in water in its molecular gaseous state; only 10 % is in the form of carbonic acid H2CO3. These two forms of carbon dioxide together constitute what is termed free CO2. The ionic forms, i.e. fixed carbon dioxide, are represented by the bicarbonate and carbonate ions (HCO3- and CO32- respectively). Their presence is important for the buffering capacity of the water. The amounts of CO2 present in flowing surface waters are typically in the order of a few mg per litre, and seldom rise above 20 to 30 mg per litre. In stagnant surface waters the CO2 levels are stratified because of photosynthetic assimilation by phytoplankton, the upper strata usually having less free CO2 than the lower strata. If all the free CO2 in the surface strata is used for photosynthesis, the pH of the water there may rise above 8.3, and in waters of moderate bicarbonate alkalinity to 10.0 and above during the daylight hours. Ground waters from limestone or chalk strata usually contain several tens of mg of free CO2 per litre, and this may be important where well water is used for fish culture.
The toxic action of carbon dioxide is either direct or indirect. The indirect action of both free and bound CO2 is exerted on fish through its influence on water pH, especially where, as described earlier, the values rise to toxic levels. Also, changes in pH affect the toxicity of those chemicals which exist in the dissociated and nondissociated forms of which only one is toxic, such as H2S and ammonia.
A direct adverse effect occurs when there is an excess or absence of free CO2. In waters of low oxygen content, such as where intensive biodegradation is taking place, or where fish are kept or transported in a high density, or when poorly aerated ground waters are used, free CO2 may reach harmful levels. In such cases the diffusion of CO2 from the fish blood into the respiratory water is reduced, the blood CO2 rises and acidosis develops. If the rise in CO2 concentration is relatively slow (e.g. over 1 day), fish can adapt to the acidosis by increasing the bicarbonate concentration of the blood. Adapted fish can then suffer from alkalinosis if returned to water of low CO2 content.
In water of low O2 and high CO2, where gaseous exchange at the respiratory surface is limited, the fish increase their ventilation rate, become restless, lose equilibrium, and may die. Twenty mg free CO2 per litre is considered the maximum permissible concentration for trout (higher concentrations can cause kidney problems) and 25 mg free CO2 per litre is the maximum for carp (if the acid capacity is 0.5 mmol per litre at a pH of up to 4.5). The sensitivity of fish to free carbon dioxide declines with increasing acid capacity of water.
However, the more frequent occurrence is a lack of free carbon dioxide in water. Carbon dioxide deficiency occurs when too much free CO2 is utilized for photosynthetic activity by the phytoplankton, or when the water used in thermal power plants is artificially softened or when water is aerated more vigorously than necessary with CO2 free air. Free carbon dioxide concentrations below 1 mg per litre affect the acid-base balance in the fish blood and tissues, and cause alkalosis. A lack of free carbon dioxide is particularly harmful to cyprinid fry when they pass from endogenous to exogenous nutrition. Cyprinid fry respire through their body surface and are unable to regulate their acid-base balance by gill respiration. A low partial pressure of free CO2 in water is conductive to a high CO2 diffusion rate from the body, leading to alkalosis and finally to death. If the fry of cyprinids suffer from free CO2 deficiency, they gather close to the water surface and show symptoms of suffocation even though the concentration of oxygen in the water is adequate (Taege, 1982).
The factors considered in this Chapter have been those which can occur in the natural environment, and which can be enhanced by man's activities. Fish have a limited ability to adapt to changes in these factors, if they occur sufficiently slowly; rapid changes can be harmful. If fish are affected to some extent by such changes, a full recovery is possible on return (in some cases, e.g. free CO2, this needs to be gradual) to normal conditions. Unless irreparable damage has been caused to fish tissues, there are unlikely to be any long-term consequences to their health.
This section briefly describes the toxicity to fish of chemicals that are likely to occur in surface waters. Where possible, the acute toxic concentrations are given to provide information useful for cases of sporadic discharges where high concentrations may exist for a short time, and maximum admissible concentrations which are relevant for low-level continuous discharges. Clinical and patho-anatomic effects are also described. For more detailed information standard reference works should be consulted.
Active chlorine2 can be discharged into water courses, lakes and ponds in effluents from textile and paper plants. Chlorine and compounds that release active chlorine into water are used as disinfectants in both public health and veterinary medicine. Thus, chlorine can be discharged in water from public swimming pools and from sterilizing procedures for equipment in dairy farms.
Chlorinated lime is used for a total disinfection of pond bottoms (application rate of 600 kg per ha), fish storage ponds and other facilities for fish culture and transport. If fish suffer from a gill disease, a recommended remedial proceedure is to spread chlorinated lime on the surface of the pond at a rate of 10–15 kg per ha (if the average depth of the pond is 1 m). However, overdosage or improper handling of chlorine or chlorine-releasing compounds can damage or kill fish. Marketed fish may also be harmed by chlorine if retailers keep them in tanks supplied with chlorinated tapwater which contains 0.05 to 0.3 mg active chlorine per litre. Higher, rapidly lethal, concentrations can occur if the water supply works abstracts water containing a high content of organic matter; excessive chlorine then has to be used to disinfect the water. If chlorinated water from a public supply has to be used, it should be passed through an activated charcoal filter to remove the chlorine; for small-scale use, small amounts (c. 10 mg per litre) of sodium thiosulphate can be added to the water to react with the chlorine. Low concentrations of chlorine can be naturally absorbed by organic matter in the water and in sediments.
Active chlorine is very toxic to fish. Its toxicity largely depends on water temperature: for example, an active chlorine concentration of 3.5 mg per litre has a sublethal effect on carp at a water temperature of 3–7°C but when exposed to the same concentration at a temperature of 15–20°C they die in 1 to 2 hours. In general, a prolonged exposure to active chlorine concentrations of 0.04 to 0.2 mg per litre is considered to be toxic to the majority of fish species.
Active chlorine may affect specific parts of the fish (e.g. the skin and gills) or the whole body (i.e. when chlorine is absorbed into the blood). The systemic effect manifests itself mainly as nervous disorders. The clinical symptoms of chlorine intoxication include a considerable restlessness, leaping out of the water, muscle tetanus, lying on one side, and spasmic movement of the mouth, fins and tail. The buccal spasms hinder respiration, so that the fish suffocate, and ultimately die. The skin and gills of the poisoned fish are covered with a thick layer of mucus and if the concentration of active chlorine is very high the gills become congested and can haemorrhage. The body surface of such fish becomes pale and the margins of the gill filaments and fins are covered with a grey-white coating. Histopathologically, there is a marked dystrophy and necrobiosis leading to necrosis, with desquammation of the gill respiratory epithelium and of the epidermis of the skin.
Cyanides do not occur naturally in waters; they can be discharged in various industrial effluents, particularly from metal plating works and from the thermal processing of coal (e.g. for town gas production). Cyanides may be present in water either as simple compounds (nondissociated HCN, simple CN ions) or as complex compounds (e.g. complexes with iron, cobalt, nickel and other metals). Simple cyanides are very toxic or extremely toxic to fish species; lethal concentrations for the majority of species are in the range of 0.03 to 0.5 mg per litre. Cyanide toxicity is affected by the pH of the water; if the pH is low the proportion of nondissociated HCN increases and so does the toxicity (Table 2). Cyanide toxicity is also markedly enhanced by an increase in water temperature and a decrease in the concentration of dissolved oxygen in the water.
With complex cyanides, the toxicity varies according to their ability to dissociate into metal and HCN. For example, the complex iron cyanides which do not dissociate are of low to very low toxicity to fish but the complex cyanides of zinc, cadmium, copper and mercury which do are highly toxic. The concentrations of different cyanide compounds proposed as maximum admissible levels for fish culture are in the range of 0.002 to 0.02 mg per litre.
The mechanism of the toxic action of cyanides is based on their inhibition of respiratory enzymes (i.e. cytochromoxidases). This blocks the transfer of oxygen from the blood to the tissues, reduces tissue respiration and leads to tissue asphyxia. The clinical symptoms of the cyanide poisoning of fish include increased depth of respiration, nervous disorders, and loss of equilibrium. If the fish are transferred to clean water while they are in the early stages of overturning, they will recover in 1 to 2 hours. The characteristic features of the patho-anatomic examination in cases of cyanide poisoning include a cherry-red colour of the gills and sometimes also leakage of tissue fluid mixed with blood into the body cavity.
Table 2: The dependence of HCN content (% of simple cyanide content) on water pH
| pH | % HCN | % CN |
| 6 | 100 | 0 |
| natural 7 | 100 | 0 |
| range 8 | 93 | 7 |
| 9 | 60 | 40 |
| 10 | 10 | 90 |
| 11 | 2 | 98 |
Trace quantities of metals present in waters may be of natural origin. If waters are polluted with metals at greater concentrations, the source may be traced back to ore mining and processing, to smelting plants, rolling mills plants for the surface treatment of metals, film, textile and leather industries and other sources. Atmospheric precipitation can wash out metals in dust and aerosols generated by the burning of fossil fuels, by the exhaust gases of motor vehicles, and from other sources.
The mechanism of the toxic action of metals on fish is varied. Most of the metals have a great affinity for amino acids and the SH groups of proteins: as such, they act as enzyme poisons. The toxicity of metals to fish is significantly affected by the form in which they occur in water. The ionic forms of metals or simple inorganic compounds are more toxic than complex inorganic or organic compounds. The toxic action of metals is particularly pronounced in the early stages of development of the fish.
Another potentially harmful property of many metals is their ability to accumulate in the sediments and in the aquatic flora and fauna (bioaccumulation). This property is quantitatively described by the accumulation coefficient (concentration in substrate/ concentration in water) and such values can range from several hundred to many thousands; mercury, selenium and cadmium have a particularly high bioaccumulation capacity. Hence, the concentration of these metals in water does not provide a true indication of the total pollution of the aquatic medium; it is better to use the content of metals in the sediments, and especially also in the bodies of predatory fish which are the final link in the food chain, as an indicator.
The metals found to be of highest importance to fisheries in practice include aluminium, chromium, iron, nickel, copper, zinc, arsenic, cadmium, mercury and lead.
The toxicity of aluminium to fish depends to a large extent on the physico-chemical properties of the water and particularly on its pH. Aluminium is soluble at pH values below 6.0; a number of chemical species can be formed, the most toxic occurring in the pH range of 5.2 to 5.8. At higher pH values, aluminium is precipitated as the hydroxide, which can flocculate in the water. It is possible that freshly precipated aluminium (as a colloid) may be toxic; the fully flocculated hydroxide has a low toxicity similar to that of suspended solids in general.
In toxicity tests, rainbow trout fry were exposed to different aluminium concentrations at a pH of 7.0. A concentration as low as 0.52 mg aluminium per litre was found to markedly reduce the growth of these fish. When an even lower concentration, 0.05 mg per litre, was tested no such adverse effect was found, so this concentration can be regarded as safe at this pH. A mass kill of maraena and peled fry, reared in a public supply water clarified by flocculated aluminium sulphate, is a practical example. The aluminium concentration of the water was up to 0.3 mg per litre and the pH value was between 7.0 and 7.5. All the fry of maraena and peled died within 10–14 days of hatching. It is not known whether this was due to ionic aluminium or to micro flocs affecting fish respiration.
In surface waters, the most stable forms of chromium are the oxidation states III and VI. Of these two forms, chromium III is poorly soluble and is readily adsorbed onto surfaces, so that the much more soluble chromium VI is the most common form in fresh water. For this reason, maximum admissible concentrations for chromium are generally based on toxicity data for the hexavalent form.
Chromium compounds in the trivalent state (III) are more toxic to fish and other aquatic organisms than are those in the hexavalent state VI. From the LC50 data obtained for different fish species, chromium III compounds are among those substances with a high toxicity to fish (LC50s of 2.0 to 7.5 mg per litre) whereas chromium VI compounds are among those substances of medium toxicity (LC50s of 35 to 75 mg per litre). The toxicity of chromium compounds to fish is also considerably affected by the physico-chemical properties of water, especially the pH value and the concentrations of calcium and magnesium. At a high pH and a high concentration of calcium, the toxicity of chromium to aquatic organisms is reduced, compared to that in soft acid water. With acute poisoning by chromium compounds, the body surface of the fish is covered with mucus, the respiratory epithelium of the gills is damaged and the fish die with symptoms of suffocation. Fish suffering from chronic chromium intoxication accumulate an orange yellow liquid in their body cavity.
In surface waters, iron occurs in ferrous state II (soluble compounds) or ferric state III (mostly insoluble compounds). The ratio of these two forms of iron depends on the oxygen concentration in the water, the pH and on other chemical properties of the water. Fish may be harmed by iron compounds in poorly oxygenated waters with a low pH where the iron is present mainly in the form of soluble compounds. Because the gill surface of the fish tends to be alkaline, soluble ferrous iron can be oxidized to insoluble ferric compounds which then cover the gill lamellae and inhibit respiration. At a low water temperature and in the presence of iron, iron-depositing bacteria will multiply rapidly on the gills and further contribute to the oxidation of ferrous iron compounds. Their filamentous colonies cover the gills; at first they are colourless but later the precipitated iron gives them a brown colour. The precipitated iron compounds and tufts of the iron bacteria reduce the gill area available for respiration, damage the respiratory epithelium and may thus suffocate the fish. In a similar toxic action, iron compounds can precipitate on the surface of fish eggs which then die due to a lack of oxygen.
The lethal concentration of iron for fish is not easy to measure because it depends to a large extent on the physico-chemical properties of the water. In cyprinid culture, it is generally accepted that the concentration of the soluble ionized forms of iron should not exceed 0.2 mg per litre; for salmonids this limit is 0.1 mg per litre.
Nickel can be discharged into surface waters in effluents from metal plating plants. Nickel compounds are of medium toxicity to fish. With short periods of exposure, the lethal concentration is between 30 and 75 mg per litre. As with the toxicity of other metals, the toxicity of nickel compounds to aquatic organisms is markedly influenced by the physico-chemical properties of water. For example, in soft waters with low calcium concentrations, the lethal concentrations of nickel compounds for the stickleback were less than 10 mg per litre. In such cases nickel can be regarded as highly toxic to fish. After toxic exposure to nickel compounds, the gill chambers of the fish are filled with mucus and the lamellae are dark red in colour.
Although copper is highly toxic to fish, its compounds are used in fish culture and fisheries as algicides and in the prevention and therapy of some fish diseases. The physical and chemical properties of the water exert a strong influence on the toxicity of copper to fish. In water containing high concentrations of organic substances copper can become bound into soluble and insoluble complexes. In very alkaline water it forms hydroxides of low solubility, and in waters with a high bicarbonate/carbonate concentration copper precipitates as poorly soluble or insoluble cupric carbonate. Compounds that are slow to dissolve or are insoluble are unlikely to be taken up to any extent into the fish body, so their toxicity to fish is low. A good example of this effect of solubility is a comparison between the different LC50s for carp recorded during 48 hours exposure to CuSO4.5H2O in a pond water [pH 7.6; acid capacity to a pH of 4.5 (a measure of bicarbonate alkalinity), 2.2 mmol per litre; CODMn 32 mg O2 per litre] and in a well water [pH 6.2; acid capacity to a pH of 4.5, 0.6 mmol per litre; CODMn 2.2 mg O2 per litre]: in the hard pond water the LC50 was 8.1 mg per litre and in the softer well water it was 1.5 mg per litre.
The maximum admissible copper concentration in water for the protection of fish is in the range of 0.001 to 0.01 mg per litre, depending on the physical and chemical properties of water and on the species of the fish. The characteristic clinical symptoms of fish poisoned by copper ions and copper compounds include laboured breathing and, in cyprinids, gasping for air at the water surface. The typical patho-anatomic appearance includes a large amount of mucus on body surface, under the gill covers and in the gills. Acute copper intoxication can be diagnosed on the basis of a chemical analysis of the gills in which the concentration of copper is much greater than in other parts of the body of the fish. The gills of fish caught in waters free of copper contamination contain up to 10 mg of copper per 1 kg of dry matter.
Zinc poisoning of fish is most frequently encountered in trout culture. Rainbow trout and brown trout, and especially their fry, are extremely sensitive to zinc and its compounds. The lethal concentrations are around 0.1 mg per litre for salmonids (some authors even suggest a level of 0.01 mg per litre) and 0.5 to 1.0 mg per litre for cyprinids. Resistance to zinc compounds increases with age. The toxicity of zinc to fish is influenced by the chemical characteristics of water; in particular, increasing calcium concentrations reduce the toxicity of zinc. The clinical symptoms and patho-anatomic picture of zinc poisoning in fish are similar to those found for copper. The best remedy to avoid frequent occurrences of zinc toxicity in trout culture is to avoid using galvanized pipes for the supply of water and to avoid using galvanized containers and equipment, especially in areas where the water is soft and acid.
As a rule arsenic occurs in water in the oxidation state V but some of it may also be present in non-stable forms, i.e. in the oxidation state III. As with mercury (see later) biological (particularly bacterial) activity may lead to the formation of organic methyl derivatives of arsenic. The main sources of arsenic pollution in surface waters include industrial effluents e.g. from tanneries, ore processing plants and dyestuff production plants. Arsenic is able to accumulate in large quantities in the sediments on the bed of water courses and reservoirs, and in aquatic organisms. Arsenic compounds in the third oxidation state (arsenites) are absorbed fairly rapidly into fish and are more toxic than arsenic compounds in the oxidation state V (arsenates). From concentrations found to be lethal to different species of fish during 48 hours of exposure, diarsenic trioxide, for example, can be included among those substances which have a medium to high toxicity to fish; lethal concentrations are between 3 and 30 mg per litre.
Cadmium in surface waters is usually found together with zinc but at much lower concentrations. The cadmium present in surface waters may be either dissolved or insoluble. Of the dissolved forms, those which may be poisonous to fish include the simple ion and various inorganic and organic complex ions. Apart from an acute toxic action which is similar to that of other toxic metals (damage to the central nervous system and parenchymatous organs), very small concentrations of cadmium may produce specific effects after a long exposure period. Chief among these specific effects are those exerted on the reproductive organs. An adverse influence of long exposure to cadmium upon the maturation, hatchability and development of larvae in rainbow trout was recorded at concentrations as low as 0.002 mg l1. The acute lethal concentration of cadmium for different species of fish is from 2 to 20 mg l1. Cadmium toxicity is reduced with increasing levels of calcium and magnesium in the water. For salmonids, the maximum admissible cadmium concentration in water is 0.0002 mg per litre, and for cyprinids 0.001 mg per litre (Schreckenbach, 1982).
Mercury is transported to the aquatic environment mainly in discharges of industrial effluents and by atmospheric precipitation. Unpolluted waters will contain trace amounts of mercury which do not exceed 0.1 μg per litre.
Mercury concentrations found in surface waters are not a true measure of the actual total amount of mercury present and therefore do not represent the extent of the mercury pollution there; this is because mercury is transferred from water to the sediments on the bed of water courses, lakes and reservoirs where it accumulates mainly as the sulphide. Elementary mercury and its organic and inorganic compounds can undergo methylation (a process induced by the activity of microorganisms) in the sediments. The toxic end-product of this methylation, methyl mercury, enters the food chains and bioconcentrates in increasing amounts in aquatic organisms up the food chain.
Mercury can be taken up into fish from food via the alimentary tract; the other routes are through the gills and skin. Absorption from the alimentary tract has proved to be of the greatest importance in methyl mercury accumulation; evidence for this has been provided by the results of investigation at sites in the drainage area of the Berounka River in Central Bohemia. The total mercury content in the flesh of fish from these localities is about 10 times that recorded in their food. This coefficient of bioaccumulation can be compared with the food efficiency coeficient of fish living in open waters and feeding on the aquatic invertebrates. Of the other aquatic organisms in the drainage area of the Berounka River, the greatest mercury levels were recorded in leeches and this can be ascribed to their exclusively predatory mode of feeding. With their wide distribution in different types of waters, leeches (e.g. Helobdella stagnalis) may be considered as good indicators of mercury contamination of the aquatic medium.
Carnivorous fish contain the highest amounts of mercury because they form the final link in the aquatic food chain. The problem of mercury in the aquatic medium is important not only for environmental and hygienic reasons but also from the viewpoint of fish culture. It has been shown that mercury compounds can cause damage to some vital tissues and organs in fish and may also have a harmful effect on reproduction. At very low concentrations they reduce the viability of spermatozoa, reduce egg production and affect the survival rate of developing eggs and fry. Acute lethal concentrations of inorganic mercury compounds are in the range of 0.3 to 1.0 mg per litre for salmonids and 0.2 to 4.0 mg per litre for cyprinids, depending on the physical and chemical properties of the water. The acute lethal concentrations of commonly found organic mercury compounds are from 0.025 to 0.125 mg per litre for salmonids and from 0.20 to 0.70 mg per litre for cyprinids. For salmonids the maximum admissible concentration of inorganic forms is about 0.001 mg per litre and for cyprinids about 0.002 mg per litre. For fish in general, the maximum admissible concentration of mercury in organic compounds has been suggested to be as low as 0.0003 mg per litre.
A significant source of airborne lead contamination, and therefore of surface waters, is the exhaust fumes of motor vehicles which contain the break-down product of tetraethyl lead. In surface waters, lead largely accumulates in bottom sediments at concentrations about four orders of magnitude greater than in the water.
Lead toxicity to fish and other aquatic organisms is significantly influenced by the water quality and depends on the solubility of lead compounds and on the concentration of calcium and magnesium in water. The water solubility of lead compounds is reduced with increasing alkalinity and pH value of the water. Also, the toxicity of lead is known to be reduced with increasing calcium and magnesium concentrations in water. The acute toxic concentrations in different types of water are in the range of 1 to 10 mg per litre for salmonids and of 10 to 100 mg per litre for cyprinids.
Acute lead toxicity is characterized initially by damage to the gill epithelium; the affected fish are killed by suffocation. The characteristic symptoms of chronic lead toxicity include changes in the blood parameters with severe damage to the erythrocytes and leucocytes, and degenerative changes in the parenchymatous organs and damage to the nervous system. In trout, a characteristic symptom is a blackening of the caudal peduncle; a biochemical effect is the inhibition of amino levulinic acid dehydrase (ALA-D) in fish blood. The maximum admissible lead concentration in water is 0.004 to 0.008 mg per litre for salmonids and 0.07 mg per litre for cyprinids.
Phenols occur in surface waters from discharges of industrial effluents, especially from the thermal processing of coal, from petroleum refineries, and from the production of synthetic fabrics. Phenols are either monobasic (e.g. phenol, cresol, naphtol, xylenol) or polybasic (e.g. pyrocatechol, resorcine, hydroquinone, pyrogallol, floroglucin).
Phenols can give an unacceptable taint to water and fish, especially chlorophenols which are formed from the chlorination of phenols. The maximum concentrations admissible for fish culture are 0.001 mg per litre for chlorophenol, 0.003 mg per litre for cresol, 0.004 mg per litre for resorcine, and 0.001 mg per litre for hydroquinone. Concentrations of 0.1 mg of phenols per litre of water and 0.02 mg of chlorophenols per litre of water are high enough to cause changes in the flavour of fish flesh. Prolonged exposure to a concentration of 0.2 mg of phenols per litre of water was observed to cause fish to migrate out of the catchment area of a polluted watercourse. Based on the lethal concentrations for fish, the different phenol compounds can be ranked as follows: hydroquinone (0.2 mg per litre), naphthols (2 to 4 mg per litre), phenol, cresol, pyrocatechol and xylenol (2 to 20 mg per litre), resorcine and pyrogallol (10 to 50 mg per litre), floroglucin (400 to 600 mg per litre).
Phenols are anaesthetics which affect the central nervous system. The clinical signs of intoxication are characterized by increased activity and irritability, leaping out of the water, loss of balance and muscular spasms.
The post-mortem appearance include a conspicuous whitening of the skin which is heavily coated with mucus; at high temperature this may be accompanied by haemorrhages on the under side of the body. Long exposure to low concentrations leads to dystrophic to necrobiotic changes in the brain, parenchymatous organs, circulation system and gills.
Polychlorinated biphenyls are recognized as very important environmental pollutants. PCBs are among the most environmentally persistent of organic compounds; although their solubility in water is very low, they are readily soluble in nonpolar solvents and can accumulate in fats. Mixtures of a large number of PCBs isomers are used in the heavy electrical equipment industry (e.g. in power capacitors and high-voltage transformers), mechanical engineering (e.g. as inflammable liquids for heat transfer, in hydraulic fluids and in lubricants for compressors) and in the chemical industry (e.g. the production of synthetic varnishes, dyestuffs and plastics). The world-wide trade names of PCBs include Delor (Czechoslovakia), Aroclor (USA), Clophen (Germany), Kaneclor (Japan), and Savol and Sovtol (USSR). In response to a growing concern about rising levels of PCBs in the environment from diffuse sources, their accumulation in biota, and uncertainty about their toxic effects, the production of polychlorinated biphenyls was restricted in 1971, and successive controls placed on its use and disposal. The main concern is that, once in the natural environment, they cannot be recovered or removed.
In surface waters, PCBs occur at concentrations from 1.108 to 1.104 mg per litre. Polychlorinated biphenyls have a high capacity for accumulation in the bottom sediments and in aquatic organisms for which the bioaccumulation coefficient is from 103 to 105, depending on the fat content.
PCBs present a very difficult ecotoxicological problem; there are 209 individual PCBs, each one with different toxicological properties. Toxicity tests are carried out on commercial formulations which are identified by the extent to which they are chlorinated, and not by the specific PCBs that they contain. This makes it difficult to assess their toxicity in the environment, because differential uptake of the individual compounds leads to a different ratio being found in organisms when compared to that in the tested formulations. Therefore, any assessment of the toxicity of PCBs can be made only in general terms on the basis of tests with commonly used formulations.
PCB formulations are very toxic to extremely toxic to fish, especially in their early developmental stages; their 48 hour LC50s are below 1 mg per litre.
Of the various toxic actions of PCBs reported, they have been found to adversely affect the enzyme systems within the microsomal fraction of the liver. If fish are exposed for a long time to low sublethal PCBs levels, the compounds accumulate in the body and can cause, mainly in the fry, deformities in the skeleton, damage to the skin and fins (the fins disintegrate), to the parenchymatous organs (mainly in the liver where hypertrophy, local dystrophy, and necrobiotic to necrotic changes can occur), and to the gonads. These effects can cause a subsequent mortality during hatching, high mortality of early fry and an increased occurrence of different deformities among the survivors.
The maximum admissible PCBs concentrations in water range from 1.10-6 to 5.10-6 mg per litre for salmonids and from 2.10-6 to 1.10-5 mg per litre for cyprinids (Mattheis et al., 1984). Lower admissible concentrations are recommended during hatching and rearing of the early stages of the fry. However, analytical measurement of these concentrations in solution may be difficult; also, a significant proportion of the uptake of PCBs will be from the food. Analysis of fish tissue will give an indication of the degree of exposure, but the concentrations found must be correlated with the tissue fat content. Where significant amounts occur, the analysis should identify and quantify a number of key individual PCBs for an expert evaluation of the potential hazard.
Surfactants are compounds which, by lowering the surface tension of water, can facilitate the formation of emulsions with otherwise immiscible liquids such as oils and fat. They are widely used domestically and in industry. In recent years, the traditional soaps have been replaced by detergents that contain synthetic surfactants and other ingredients; for domestic washing of garments, these may contain water softeners, optical brighteners, and perfumes.
Surfactants are either ionic (liable to electrolytic dissociation) or nonionic (nodissociating in water). Ionic surfactants are subdivided into anionic (dissociating to a surface active anion and an inactive cation), cationic (dissociating to a surface active cation and an inactive anion), and ampholytic (assuming either anionic or cationic properties, depending on ambient conditions). The anionic surfactants are those most widely used in industry.
Because of the large number of synthetic surfactants in production, it is not surprising that they span a wide range of chemical toxic actions for aquatic organisms. However, they do have a common physico-chemical effect in that they can damage the lipid components of cell membranes. Because the surface tension of the ambient water is decreased, the lipids are less water repellant and this leads to hydration and enlargement of the cell volume. At low surfactant concentrations this enlargement is reversible. Higher concentration can cause a suppression of metabolic processes in the cells. Long-term exposure may damage the cells which then become necrotic in the later stages. These changes result mainly in an impairment of the gill respiratory epithelium. In addition, the exposure of fish to some surfactants can cause changes in the activity of respiratory enzymes, especially cytochromoxidase. Surfactants can also damage the protective layer of mucus on the skin; the layer loosens and the resistance of the fish to infection decreases. Sublethal surfactant concentrations can also damage eggs and sperm.
The toxicity of surfactants to fish is influenced by a number of biotic and, especially, abiotic factors. The age of the fish is a particularly important biotic factor.
During embryonic and larval development, the sensitivity of fish to surfactants is sometimes greater by an order of magnitude in comparison with the juvenile and adult stages. Of the abiotic factors, the molecular structure of the surfactant and the physico-chemical properties of water exert the greatest influence on their toxicity. The results of investigations into the relationship between toxicity and molecular structure indicate, for example with linear alkylbenzene sulphonates, that the toxicity to fish is markedly increased with the length of the molecular chain. A similar correlation between toxicity and chain length was observed with other surfactants. Among the physico-chemical properties of water, increasing calcium and magnesium concentrations have the greatest effect on reducing surfactant toxicity and some influence is also exerted by the pH. This may be important where surfactants are incorporated into a detergent containing water softening chemicals (e.g. polyphosphates). Where both cationic and anionic surfactants are present in waste waters their toxicity is much reduced, due to the formation of insoluble complex.
The acute toxicity of surfactants varies considerably with the species of fish. Nevertheless, these compounds and the detergents that contain them are highly toxic to fish in the majority of cases, the 48-hour LC50 ranging between 1 and 10 mg per litre. A small proportion of surfactants can be classified as having a medium toxicity (48-hour LC50 between 10 and 100 mg per litre) and a few have a very low toxicity (48-hour LC50 to 10 000 mg per litre). For the majority of surfactants, no significant differences in their toxicity to fish were recorded between the anionic, cationic and nonionic groups.
As stated above, surfactants can cause damage to the gill respiration epithelium (e.g. enlargement and vacuolation of the cells with dystrophic to necrobiotic changes). Therefore, the clinical signs of poisoning include respiratory disorders (increased respiration rate, and cyprinids gasp for air at the water surface) and later by inactivity. The characteristics in the patho-anatomic examination are an increased amount of mucus on the skin and in the gills, and congestion to oedematous swelling of the gill apparatus. The mucus is easily removed from the body surface and gills.
In recent years, the number of pesticides available and the quantity used has considerably increased. The term “pesticide” is used to include insecticides, acaricides, herbicides, fungicides and algicides, indeed any chemical which is used to control an unwanted organism (except bacteria), even rotenone which is used to kill unwanted fish. Pesticides are chemicals which have a specific toxic action to which the pest species is particularly sensitive. The chemical is then applied at a concentration which kills the pest but does not affect a wide range of non-target organisms. The ideal pesticide is a chemical which is extremely pest-specific; for the pesticide user it should also be persistent in order to avoid the need for repeated applications. However, on environmental grounds, pesticides should be non-persistent to avoid concentrations building up in environmental compartments and causing unsuspected side-effects. For example, the insecticide DDT is very persistent and thus can build up in food chains to ultimately affect the egg-shell thickness of birds of prey.
Because pesticides are designed and used to kill living organisms, and because of the possibility of unsuspected side effects, it is tempting to implicate them in any incident of fish poisoning where there is no other obvious cause of the damage. There are many cases, therefore, where pesticides have been assumed to be the cause of damage but where the real cause was some other factor.
Some cases of pesticide poisoning of fish are obvious; accidental discharges from road accidents, factory disasters, overspraying of water, or careless disposal of unwanted spray and pesticide containers, can be clearly identified as causes of mortality, especially if the concentrations measured or calculated in the water exceed the 96 hour LC50 by a significant margin. Less easily identified are cases of long-term leaching of persistent pesticides from fields and forests. Besides these acute and chronic direct effects, an indirect action may be important. Inexpert application of aquatic herbicides or algicides to the water, or the accidental contamination of surface waters with these chemicals, may kill excessive quantities of aquatic plants and algae. The rapid decomposition of this organic matter forms a considerable dissolved oxygen demand on the water. This will lead to an oxygen deficit and the fish may die of suffocation.
Another potentially serious indirect consequence of pesticide contamination of the aquatic biota is the reduction or complete destruction of the natural food supply of the fish. Many of the organisms on which the fish feed are much more sensitive, particularly to insecticides, than the fish themselves. For example, the LC50 for the organo-phosphorus insecticide formulation “Soldep” (active ingredient 25% trichlorphon) for common carp is 545 mg per litre of water whereas for Daphnia magna it is 0.0002 to 0.001 mg per litre.
Besides the active ingredient, pesticide formulations contain a number of other chemicals which may sometimes be much more toxic to fish than the active ingredient itself.
When a pesticide enters the aquatic environment, the active ingredient may undergo chemical and biological degradation. In some cases the degradation products may be more toxic to fish than the original active ingredient. For example, parathion is biodegraded to form paraoxon, which is a more toxic compond; similarly, trichlorphon is degraded to form the more toxic compound dichlorvos. It follows that the absence of a specific active ingredient in water cannot guarantee that harmful degradation products are not present.
Some herbicides are used in fish culture and water management to kill unwanted aquatic plants (e.g. Gramoxone S, Reglone). Trichlorphon based organo-phosphorus insecticides, e.g. Soldep, Masoten, Neguvon, etc. are used to reduce the larger Daphnia in the zooplankton to prevent an oxygen deficit in the pond, to kill predatory cyclopids before stocking the pond with fish at the sac fry stage, to control parasites that infest cyprinids, and for other management purposes. Pesticides based on copper oxychloride may be used to control fish parasites, including the control of gastropod intermediate hosts, and to kill excessive growths of algae.
However, in the majority of cases pesticides have the potential to cause damage to fish. The most toxic pesticides are those based on chlorohydrocarbons (e.g. DDT, dieldrin), organo-phosphorus compounds, carbamates and thiocarbamates, carboxylic acid derivatives, substituted urea, triazines and diazines, synthetic pyrethroids, and metallic compounds.
These pesticides act as nerve poisons. They are highly to extremely toxic to fish (48-hour LC50 < 1.0 mg per litre). Because of their chemical structure and their persistance, their use is now strictly controlled or banned.
The clinical signs of fish poisoning by organochlorine pesticides on the basis of chlorohydrocarbons include increased activity, followed by a long stage of reduced activity. There is no specific patho-anatomic picture in these cases of intoxication; dystrophic alterations have been recorded in the liver and kidneys.
The mechanism of the toxic action of organo-phosphorus pesticides on fish follows the same pattern as their action on homoiothermic animals, in that some hydrolytic enzymes, particularly acetylcholine hydrolase, are inhibited. The degree of inhibition of cerebral acetylcholine hydrolase in fish varies with the specific organo-phosphorus compound causing the effect. Phenitrothion-based pesticides reduce the enzyme activity to 60%, dichlorvos- and imidane-based pesticides cause a greater reduction which leaves only 22% of the physiological activity remaining. The toxicity of these pesticide formulations to fish also varies; from the 48h LC50s obtained they are ranked among those substances of very high to medium toxicity to fish (0.1–100 mg per litre). Also, salmonids are very sensitive to organo-phosphorus pesticides. The typical sign of fish poisoning with these pesticides is a darkening of the body surface at the onset of uncoordinated activity. The patho-anatomic picture of such poisoning is characterized by a considerable mucus production on the body surface and in the gills, a heavy congestion of the gills, and small isolated (“spotted”) haemorrhages in the gills when the pesticide concentration is high.
The water flea (Daphnia magna) is very sensitive to organo-phosphorus pesticides; from the 48h LC50s obtained for these substances, they can be classified as extremely toxic. It is interesting to note that the water flea was found to be sensitive to trichlorphon and dichlorvos at concentrations close to the level of detection by gas-liquid chromatography. Daphnia magna can be regarded, therefore as a sensitive indicator of organophosphorus pesticide pollution.
Carbamate and thiocarbamate compounds also inhibit the activity of acetylcholine hydrolase. However, unlike the toxic action of organo-phosphorus compounds, the inhibition of enzyme activity is readily reversed after carbamate and thiocarbamate poisoning. The toxicity levels of these substances to fish vary from very high to low toxicity (48h LC50s in the range of 1 to 1000 mg per litre). The clinical and patho-anatomic pictures of fish poisoning by these pesticides are not specific.
A number of these pesticides are based on phenoxyacetic acid; the main representative of this group is 2-methyl-4-chlorphenoxyacetic acid (MCPA). Most of the MCPA-based products are of medium to low toxicity to fish (48h LC50s in the range of 10 to 1000 mg per litre). The clinical signs of poisoning are mostly characterized by increasing narcosis. There is no marked patho-anatomic picture in fish poisoned by these herbicides.
Herbicides formulated from substituted ureas are of high to low toxicity to fish (48h LC50s are in the range of 1 to 1000 mg per litre). The clinical symptoms of poisoning are not specific and include increased activity, irregular respiration, uncoordinated movement and a long period of “distress”. The patho-anatomic picture is characterized by an increased amount of mucus on the darkened body surface, hyperaemia of the gills and the presence of a small amount of exuded fluid in the body cavity of the fish.
Triazine-based pesticides are of high to medium toxicity to fish (48h LC50s range from 1 to 100 mg per litre). The clinical signs of fish poisoning by these chemicals are largely characterized by progressive narcosis. The presence of exuded fluid into the body cavity and into the digestive tract is an especially characteristic patho-anatomic sign, particularly in rainbow trout. The presence of exudates causes a marked swelling of the body cavity; in rainbow trout it has even led to a rupture of the body wall in some cases.
Diazine-based herbicidal preparations are less toxic to fish than are triazine-based preparations. Most of the former are of low to very low toxicity to fish (48h LC50 ranging from 100 to 10 000 mg per litre). The clinical course of intoxication is characterized by stages of immobility. The patho-anatomic picture is not specific to these compounds.
The 48h LC50s of these pesticides show that they rank among those substances of high toxicity (up to 10 mg per litre) to extreme toxicity (less than 0.1 mg per litre) to fish. The clinical signs of poisoning are not specific and include respiratory disorders. The most conspicuous patho-anatomic change is the presence of a small amount of exuded fluid in the body cavity.
These include primarily the fungicides formulated from compounds containing copper, mercury and aluminium. In the majority of cases, their toxicity to fish and the clinical and patho-anatomic symptoms correspond to those found in fish poisoned by the respective metals.
Oils and refined products have been responsible for many of the recently recorded pollution incidents in surface and underground waters. Between 1970 to 1990 these substances were responsible for the majority of water pollution accidents recorded on a worldwide basis. These accidents were not associated with problems in sewage treatment plants; most of them were due to careless storage and handling of oil, transport accidents, and defective equipment, all of which can be ascribed either directly or indirectly to human error.
However, oils and refined products can also be discharged into the aquatic environment with industrial effluents. The petrochemical industry is mainly responsible for such effluents; other important sources of pollution include the engineering and metallurgical industry and car and truck repair and service stations. Most of these sources have discharged polluting effluents for many years. To some extent, the large number of reported oil-related pollution incidents is due to the very visible surface film that is formed; it therefore needs no chemical analysis for its detection. Even very small discharges can produce a large area of “sheen” in which the thickness of the oil is about 1 micron. For this reason, few discharges of oil go unnoticed. The harmful effects of such discharges depend on the physical effects of the surface film, and on the transfer of water soluble products into the water.
However, few of the constituent of oil and refined oil products will readily dissolve in water. There are also large differences between oil and its different products as to their toxicity to fish; most of them have 48h LC50 values within the range of 0.5 to 200 mg per litre. The toxicity varies according to the chemical composition of the different products, with the water solubility of the different petroleum hydrocarbons, and with the degree of emulsification of insoluble components in the water. It is generally agreed that the lighter oil fractions (including kerosene, petrol, benzene, toluene and xylene) are much more toxic to fish than the heavy fractions (heavy paraffins and tars). There are also differences in the sensitivity to oils and refined products between different fish species. The fry of predatory fishes (especially pikeperch and trout) show the greatest sensitivity to refined products.
When oils are discharged to rivers or ponds they spread on the surface, thus reducing (especially in stagnant waters) the transfer of oxygen from the air to water. In cases of pollution of flowing turbulent waters the oil does not from an intact layer on the water surface but becomes dispersed as droplets into the water. In such cases, the gills of fish can become mechanically contaminated and their respiratory capacity reduced. Oil products may contain various highly toxic substances, such as benzene, toluene and xylene which are to some extent soluble in water; these penetrate into the fish and can have a direct toxic effect. These toxic components include the naphthenic acids which are acute nerve poisons and are able to kill fish at concentrations as low as 0.03 to 0.1 mg per litre.
In general, oils and most of the refined products have a narcotic effect on fish; acute symptoms are effects on the nervous system and respiratory activity. The main clinical symptoms include an initial increased activity and respiratory rate followed by a loss of balance (the fish lie on their side), loss of response to stimuli, reduced activity, shallow respiratory movements, and ultimately death.
The scales of the dead fish are dull in colour and are covered with mucus; the skin shows local congestion, the epidermis fractures and peels off, and surface wounds may occur in some cases. Damage to the cornea of the eyes may lead to blindness. The gills show severe dystrophic effects and necrosis and there may also be a proliferation of the respiratory epithelial cells and hypertrophy of the mucus cells. Prolonged exposure to oils at low concentrations can cause severe degenerative necrobiotic effects in the kidneys of the fish and in their eggs. The dead fish have an oily odour and flavour.
Therefore toxicity is not the only harmful consequence associated with oil pollution; the aquatic ecosystem in general, and fish farming in particular, can be badly affected by the oily smell and taint of the water and of the organisms living there. For this reason, a sensory assessment is preferred to toxicological analyses in determining the highest admissible amounts of oil and oil products that can be present in water; on this basis the highest admissible concentrations are in the range of 0.002 to 0.025 mg per litre.
Chemical dyestuffs have also been attracting the attention of toxicologists in recent years. These can be present in the effluents from textile production, food processing and paper mills. Although these coloured effluents are, like oils, very conspicuous even at very great dilutions they seldom cause severe damage to the fish. The toxicity of dyes depends on the physico-chemical composition of the water; in water containing considerable amounts of organic matter the dyes are bound to these substances and their toxicity is decreased.
The mechanism of toxic action of effluents containing dyestuffs on fish is not direct in the majority of cases. If the water is heavily polluted with coloured organic waste, the increase in the organic content alone can lead to an oxygen deficit. Other dyes may increase or decrease the water pH. Some, e.g. aniline, can act as methaemoglobin poisons and as carcinogenic substances.
There is a considerable variation in the acute toxicity of different dyestuffs to fish. Most of the dyes rank among those substances of low to very low toxicity to fish (48h LC50s in the range of 100 to 10 000 mg per litre). This group includes colouring agents used in the food industry and selected organic dyestuffs. Another group, including e.g. acriflavine, rhodamine and also aniline and to a lesser extent methylene blue, are substances of medium toxicity to fish, with 48h LC50s in the range of 10 to 100 mg per litre. The group of dyes of very high toxicity to fish includes, for example, a formulation of malachite green.
The clinical symptoms of fish poisoning by different dyes are not specific. The patho-anatomic changes that indicate such poisoning may include a change in body colour due to the particular dyestuff, and the organs inside the fish body may also take on an intensive colour, e.g. as with malachite green.
As described earlier in this publication increasing eutrophication of surface waters can cause a massive development of phytoplankton and higher aquatic plants. This bloom can cause the water pH to rise to levels above 10, and its collapse and subsequent decomposition together with other decaying organic matter can cause an oxygen deficit. Further, some algal species produce substances (toxins) that may affect not only fish but also domestic animals and man if the water is ingested. These species include, in particular, the blue green algae of the genera Microcystis, Aphanizomenon and Anabaena. An endotoxin, with the properties of cyclic polypeptides, has been isolated from the alga Microcystis aeruginosa.
In exposed fish, the action of these blue-green algal toxins is to increase thiaminase activity and reduce thiamine content in organs and tissues; this leads to a vitamin B1 deficiency. The toxins are released into the water during the period of algal bloom particularly when the cells die and decompose. These toxins can enter the fish through the gills and body surface; some may also be ingested with food. The clinical symptoms of poisoning include damage to the central nervous system. Initially, there is increased activity and respiration, followed by uncoordinated movements; finally the fish lie flat on the bottom and die. The major patho-anatomic signs include haemorrhages on the skin and gills and in the internal organs.
Some phytoplankton species have been found to produce hydroxylamine as a by-product of their metabolism. The occurrence of this hydroxylamine in the heavily eutrophicated waters of some ponds is also accompanied by a high concentration of organic substances; these reduce the oxide reduction potential of the environment, thus allowing the hydroxylamine to accumulate. For this reason, the highest hydroxylamine concentrations are usually recorded in surface waters during the periods when a bloom of blue-green algae is decomposing, and under these conditions, the concentrations may reach toxic levels for a short time. Hydroxylamine is highly toxic to fish, its LC50 for acute exposure being less than 10 mg per litre (in sensitive fishes it may be as low as about 1 mg per litre). The toxic action of this substance includes severe methaemoglobinaemia and damage to the central nervous system. Although there are few cases where damage to fish has been attributable to the action of phytoplankton toxins, this possible source of pollution should not be underrated, especially in warm regions.
