Water Quality

Ambient Water Quality Criteria for Fluoride

---

5.0 Aquatic Life

---

General Effects

Fluoride ions are directly toxic to aquatic life, and accumulate in the tissues, at concentrations where absorption rates exceed excretion rates. Some accumulation occurs in all tissues, but in most tissues subsequent losses may occur when ambient fluoride levels decrease. However, in bone, tooth and scales, accumulation is permanent and cumulative. Temperature affects fluoride toxicity (Table 5.14), in part because metabolic rates and thus uptake rates double for every 10°C rise in temperature. The duration of exposure also affects toxicity. The fluoride level necessary to cause an LC₅₀ decreases as the time of exposure increases. For brown trout living in the Firehole River of Yellowstone Park, where natural fluoride levels are 12 to 14 mg/L, there is a linear relationship, fluoride (mg/kg) = 5.501 length (mm)-471, between the fluoride levels in the skeletal bone and the length (presumably a function of the age) of the fish (Neuhold and Sigler, 1960).

Pretreating fish in high chloride solutions confers protection against subsequent high fluoride levels (Neuhold and Sigler, 1961). The sizes (ages) of fish also affect the fluoride toxicity level and accumulation rates and levels. Larger fish are more tolerant of higher fluoride levels and accumulate less fluoride on a per weight basis (Hemens et al., 1975).

Water hardness, mostly calcium, also affects fluoride toxicity; however, there exists considerable confusion in this area. The solubility of CaF₂ is 16 mg/L (7.8 mg/L fluoride and 8.2 mg/L calcium). When high fluoride levels are used in experiments with hard water, both the hardness and the fluoride level of the solution drop rapidly due to precipitation of CaF₂. This precipitation is noted by several experimenters in their published papers. If additional fluoride is added to try and maintain a fluoride level, precipitation will continue as long as free calcium is available.

One can eventually achieve a high fluoride solution but only after most calcium and magnesium have been reacted. This means that the water is no longer hard, but soft. Thus, one can not carry out an experiment on the effects of high fluoride in hard water, and if hardness is maintained, fluoride levels will be driven down to less than 10 mg/L (Ericsson, 1970; and Vallin, 1968). CaF₂ precipitation has been reported at 7.4 mg/L fluoride, pH 7.0 to 8.0, temperature 19.5° to 21.5°C and hardness 250 mg/L (Dave, 1984). One paper reported that any protective effect of water hardness is slight, if present, and probably due entirely to precipitation of CaF₂, thus subjecting the fish to lower fluoride levels than were added (Smith et al., 1985).

In one experiment, fluoride was analyzed and it was found that total fluoride in the system, dissolved plus precipitate, was within 5% of the amount added, but that the amount in solution was lower than that added and also varied greatly on a daily basis (Herbert and Shurben, 1964). Therefore, one can not relate organism survival to some known fluoride level. All reported fluoride levels over 7.8 mg/L, when calcium ions were available, are suspect; all hardness levels reported are also suspect (see Table 5.13). Thus, the only reliable data are for low fluoride levels, less than about 5 mg/L, in very soft water, when the experimenters actually measured F^- and hardness in the solution, rather than weighing out fixed amounts of chemical and calculating fluoride and hardness values.

Several papers were reviewed in which fluoride levels were actually measured and maintained by replenishment (Neuhold and Sigler, 1960; Fieser et al., 1986; Hemens and Warwick, 1972; Wright, 1977; Hekman et al., 1984; Pimental and Bulkley, 1983a; Kaplan et al.., 1964; Barbaro et al., 1981; Milhaud et al., 1981; and Neuhold and Sigler, 1961). The control of hardness in these experiments was not always clear. Some were marine water experiments, and in some water was deliberately de-ionized on a water softener column before use so hardness was known to be negligible (less than 10). In several other papers it was not clear whether or not fluoride was actually measured in the solution after it was added (Wright and Davidson, 1975; Angelovic et al., 1961b; Wallen et al., 1957 ; Moore, 1971; Hassall, 1969; and Anon, 1960).

Theoretical equilibrium concentration calculations based on reactions used to reduce fluoride levels in drinking water or effluent, by precipitation with calcium, produce lower limits to the fluoride levels of 0.06, 0.2 and 1.3 mg/L fluoride when residual hardness levels of 1000, 100 and 10 mg/L, respectively, are achieved. These hardness values correspond to residual calcium levels of 400, 40 and 4 mg/L respectively. In practice, such low fluoride levels are not achieved due to other competing reactions.

Data from different sources are often conflicting since the variables mentioned above are frequently not even recognized as pertinent variables, and are rarely controlled or reported. Several papers have commented on these conflicting results and have postulated reasons, but have not apparently appreciated all the uncontrolled confounding factors. "The available data suggest that a uniform consensus about the maximum safe level of fluoride ion for fish in natural waters of varying hardness has not yet been achieved" (Smith et al., 1985). We concur completely with this statement and will set a criterion for fluoride in very soft waters only since this is the only condition for which there are reasonably reliable data. Such a criterion should be the worst case, or most sensitive condition, criterion and will be an interim value until more carefully controlled experiments can be carried out to determine what, if any, adjustments should be made for hardness and other environmental variables.

Effects on Marine Organisms

Of the natural fluoride level in the ocean, 1.2 to 1.4 mg/L, only about half is in the biologically available fluoride ion form; the rest is present as a relatively insoluble magnesium fluoride complex (Riley and Skirrow, 1965). Thus the effects on marine organisms noted for ambient fluoride levels are essentially due to actual levels of about 0.6 to 0.7 mg/L of available fluoride.

Fish
Experimental work on the mullet, Mugil cephalus, has not demonstrated mortality in 96 hours at 20°C, at 10 to 100 mg/L fluoride as NaF. In a 72 day experiment at 25°C, and only 20 ppt salinity, there was only a 70% survival when fish were subjected to 52 mg/L fluoride. These surviving fish were in poor condition since they were not fed for the duration of the experiment (Hemens and Warwick, 1972). Mullet exposed to 5.88 mg/L fluoride, as NaF, for 68 days had a 100% survival rate and did not demonstrate any growth effects when they were 52 mm long, but a 90% survival rate and decreased growth were observed when they were only 17 mm long. In 113 day experiments at 25°C in 5.70 or 5.65 mg/L fluoride, as NaF, survival was just over 80% (Hemens et al., 1975).

Invertebrates
The crabs Portunus depurator, Cancer pagurus and Carcinus maenas were subjected to NaF in 90 day experiments. The fluoride concentrations used were 1.0, 2.4, 10, and 30 mg/L. No deaths occurred under these conditions. The mussel, Mytilus edulis showed no lethal effects in 42 days at 2.4 mg/L F , 75% mortality in 30 days and 100% mortality in 36 days at 10 mg/L fluoride, and 75% mortality in 14 days and 100% mortality in 21 days at 30 mg/L fluoride (Wright and Davidson, 1975). The prawn, Penaeus indicus, and the crab, Tylodisplax blephariskios, were subjected to 5.5 mg/L fluoride, as NaF, for 113 days. The prawns showed increased body weight and survival over the controls while the crabs showed no effect. In 68 day experiments at 5.9 mg/L fluoride, as NaF, there was no effect on growth or survival in either the crabs or the prawns (Hemens et al., 1975). The prawns Penaeus indicus and P. monodon were tested for 96 hours at 20°C in 10 and 100 mg/L fluoride, as NaF. No mortality was noted at either fluoride level (Hemens and Warwick, 1972). The crab, Tylodisplax blephariskios, shrimp, Palaemon pacificus, and prawn, Penaeus indicus were subjected to 52 mg/L fluoride, as NaF, for 72 days at 25°C. The survival percentage, compared to controls at 1 mg/L fluoride, was 32% for the crabs, 71% for the shrimp and 100% for the prawns (Hemens and Warwick, 1972).

The crab, Callinectes sapidus, showed a 4.5% reduction in growth per moult when subjected to 20 mg/L fluoride, as NaF (Moore, 1971). The shrimp, Palaemon pacificus, suffered 23 to 45% mortality in a 72 day experiment when subjected to 52 mg/L fluoride and oysters showed 100% mortality in 60 days when immersed in 32 to 128 mg/L fluoride (Connell and Miller, 1984).

Effects on Freshwater Organisms

Tables 5.2, 5.3, 5.5, 5.6, 5.7, 5.8, 5.11, 5.13 and 5.14 give data on the effects of fluoride on freshwater and marine organisms. Table 5.2 lists the effects on fish, presented in sections by fish species. Table 5.3 deals with other non-fish organisms and is also sorted into sections by groups of related organisms. Tables 5.5 and 5.6 deal with the relationship between temperature and fluoride for LC₅₀ and non LC₅₀ data respectively. Tables 5.7 and 5.8 do the same for hardness and fluoride interactions. Table 5.11 gives the results of one set of experiments on Catla catla, an Indian fish, relating mortality at various fluoride levels to time till death. Tables 5.13 and 5.14 give data on the effects of fluoride on rainbow trout as related to hardness and temperature respectively.

Fish
In fish, fluoride toxicity is known to depend upon may factors. Species is one factor. For example, rainbow trout, Oncorhynchus mykiss, are susceptible to about 5% of the fluoride level which affects carp, Cyprinus carpio, as inferred by the LC₅₀ values (Neuhold and Sigler, 1960). Larger fish survive longer than smaller fish, as determined by weight or length; there is little effect on the LC₅₀ value only on the length of time needed to achieve the same LC₅₀ (Neuhold and Sigler, 1960; and Sigler and Neuhold, 1972). The prior history of the fish also affects the LC₅₀ value. Marine rainbow or those raised in low fluoride water have an LC₅₀ around 3 mg/L, while fish native to high fluoride waters, 14 mg/L fluoride, are able to live and breed successfully (Sigler and Neuhold, 1972). Fish subjected first to high chloride levels become less susceptible to subsequent fluoride (Neuhold and Sigler, 1961). Fish are more susceptible to fluoride at higher temperatures (Angelovic et al., 1961b), and more susceptible in soft as opposed to hard waters (Vallin, 1968; Pimental and Bulkley, 1983a; and Spohn, 1984). Fish also respond to the duration of exposure to fluoride; LC₅₀ values decrease for longer exposure periods.

Experimental conditions reported in the literature do not account for all these variables or even report some of them. Thus, comparisons of data are impossible on a quantitative basis. Much of the published data on the interactions of fluoride and water hardness are not reliable because thermodynamically unstable combinations of hardness and fluoride were being attempted and the CaF₂ simply precipitated out of solution. Thus effects reported for calculated fluoride and hardness levels were actually occurring at lower fluoride and lower hardness levels.

Tables 5.2, 5.5-5.8, 5.11, 5.13 and 5.14 give the effects of fluoride on fish. Most fish are much less sensitive to fluoride than are trout or salmon. The LC₅₀ values vary greatly in the literature since they are dependent upon many variables, most of which were not controlled in any specific experiment. As previously discusssed, it is impossible to reconcile much of these data and much are not even valid due to precipitation of CaF₂ in high fluoride experiments using hard water.

Other Freshwater Organisms
For the frog, Rana pipiens, the 30 day LC₅₀ has been calculated to be between 150 and 200 mg/L fluoride at 25°C, hardness 12 (Kaplan et al., 1964); however, embryonic and developmental effects are noted at fluoride levels as low as 1.0 mg/L (Cameron, 1940; and Kuusisto and Telkka, 1961). In Daphnia magna, experiments in hard water at 20°C showed long-term effects on reproduction at 0.6 mg/L (Dave, 1984). Many other experiments do not show effects until much higher fluoride levels are reached. Bacteria and protozoans appear to be relatively insensitive (Wantland, 1956; Bringman and Kuhn, 1959a; Bringman and Kuhn, 1959c; and Vajdic, 1966), but Chlorella, a freshwater algae, shows some growth effects at fluoride levels as low as 2.0 mg/L (Groth, 1975a; Groth, 1975b; and Smith and Woodson, 1965). Table 5.3 lists the effects of fluoride on aquatic organisms other than fish.

Accumulation

Fluoride accumulates in hard or mineralized tissues such as bones, teeth and invertebrate exoskeletons. In bone, the presumed process is F^- ion exchange in the hydroxyapatite complex. In the amorphous crustacean skeletons, most fluoride is likely to be simply CaF₂ precipitated in the open matrix. This latter inorganic fluoride is much more readily extracted in the stomachs and crops of predators. The stomach contents of cod, Gadus, caught off the British Coast were primarily crustacean remains which had fluoride levels over 100 mg/kg (Wright and Davidson, 1975). Dietary studies have shown that fish accumulate fluoride in hard tissues (Ke et al., 1970; and Zipkin et al., 1970), and in parts of South East Asia this results in some human populations having a high fluoride diet (Minoguchi, 1970). Marine mammals and birds which live on fish can also receive excessive fluoride in their diets from this source. Those parts of the fish in contact with the water, such as scales, fins and gills, have the highest fluoride levels. Skin is very high in fluoride and predators consuming the whole fish are subject to much higher fluoride levels than man who often removes the skin first. Canned salmon and mackerel have high fluoride levels in the bones (Lee and Nilson, 1939), and some prepared feeds containing fish meal also have high fluoride levels (Fisher, 1951).

For marine and estuarine environments, Water Quality Criteria-1968, states: "there is evidence that fluorides are accumulative in organisms. It is tentatively suggested that allowable levels should not exceed those for drinking water" (McNeeley et al., 1979; and Groth, 1975a).

Marine Fish
Fish, mostly mullet, caught in a bay where fluoride-rich effluents were discharged, were found to have tissue fluoride levels 4 to 5 times higher than fish caught in a nearby unpolluted bay. Fluoride levels in the unpolluted bay were about 1. 4 mg/L and in the two polluted areas were about 2 and 3 mg/L. These relatively low, but above background, fluoride levels caused disproportionately large increases in the fluoride of almost all fish tissues (Milhaud et al., 1981). Table 5.10 summarizes fluoride levels found in various tissues of other marine fish as a function of the ambient water fluoride levels. The skin, particularly scales and fins, and the skeleton are found to accumulate fluoride.

Skin without scales or fins generally had a ratio of fluoride in skin to fluoride in water ranging from 6:1 to 73:1 for a mean of 20:1. One unusual specimen had a ratio of 185:1. (The fluoride in the water was expressed as mg/L, and in fish tissue as mg/kg). For whole skin with scales and fins attached, the ratios ranged from 7:1 to 44:1 for a mean of 21:1. In fins, the usual ratios ranged from 118:1 to 468:1 for a mean of 221:1. One specimen had a ratio of 1178:1. In scales, the usual ratios ranged from 62:1 to 404:1 for a mean of 227:1. There was one unusual specimen with a ratio of 570:1. In bone, the ratio increases dramatically once ambient fluoride levels rise above the normal 1.4 to 1.7 mg/L; even as little as 2.0 mg/L causes a significant change in the ratio. For water up to 1.72 mg/L fluoride, the ratio ranged from 18:1 to 214:1 with a mean of 54:1; this includes one high value of 214:1. When water fluoride levels were 2.0 or greater, the ratios typically ranged from 24:1 to 204:1 with a mean of 131:1. There was one unusually high ratio of 596:1.

Marine Invertebrates
Oyster tissues were found to accumulate fluoride when oysters were exposed to fluoride levels exceeding 2 mg/L (Moore, 1969). Barnacles are good fluoride pollution level indicators since they accumulate fluoride in direct proportion to the ambient fluoride level (Barbaro et al., 1981). Fluoride levels in the Lagoon of Venice did not exceed 1.48 mg/L, but the barnacle, Balanus amphitrite, and the mollusc, Mytilus galloprovincialis, accumulated fluoride in their tissues 20 to 70 times the ambient levels on a dry weight basis. The barnacles reached 81 (61 to 101) mg/kg fluoride (Barbaro et al., 1981).

In an unpolluted area of New Zealand, the shells of invertebrates feeding on plankton were found to have fluoride concentrations of 31 to 209 mg/kg, while the skeletons of the blue cod feeding on the crabs, shrimp and shellfish had fluoride levels of 1425 to 1882 mg/kg. This constitutes a bio-magnification factor of at least one order of magnitude for each step in the food chain (Stewart et al., 1974). Blue crab, Callinectes sapidus, muscle tissue accumulates fluoride when the crabs are exposed to fluoride levels exceeding 1.5 mg/L (Moore, 1971). At 1.5 mg/L in seawater the tissue fluoride levels reach 2.5 mg/kg. On a dry weight basis (25% of the wet weight) the blue crab exoskeleton contains 298 mg/kg fluoride, the gills 253 mg/kg, the hepatopancreas 22 mg/kg and muscles 10 mg/kg. Significant amounts of fluoride are accumulated when ambient fluoride levels exceed 2.0 mg/L. Crab muscle accumulates 5 times as much as fluoride, or 50 mg/kg, after 90 days exposure to 50 mg/L fluoride. On a wet weight basis this results in 5.7 mg/pound or 12.5 mg/kg in edible muscle tissue and increases to 150 mg/kg when the water contains 400 mg/L fluoride. The mean daily total fluoride dose from all sources for a 70 kg man is estimated to be 5.3 mg fluoride which is equivalent to 1/2 kg of crab meat for crabs grown in 50 mg/L fluoride seawater or 2 kg of crab meat for crabs raised in normal seawater.

Table 5.9 summarizes fluoride accumulation in tissues of prawns, shrimp, crabs and mussels grown in seawater with fluoride concentrations from ambient up to 52 mg/L. At ambient fluoride levels, prawn muscle can reach 2.1 mg/kg. Thus, the total adult daily dose of fluoride would be reached after consuming about 2.5 kg of prawn meat.

Freshwater Fish
The Madison River in Yellowstone Park has natural fluoride levels of 12 to 14 mg/L. The resident one to three year old brown trout, Salmo trutta, were found to have bone fluoride levels of 1600 mg/kg (Neuhold and Sigler, 1960). This represents an accumulation of 114 to 133 times the water fluoride level. Such levels are 10 to 100 times higher than bone levels of marine fish raised in ambient fluoride level waters, and are more typical of the ratios found for marine fish raised in water with high fluoride levels. Brown trout 2 cm long raised in pH 6.8 water at 12°C and hardness 73, for 200 hours, accumulated more whole body fluoride as the water fluoride level increased. At a water concentration of 5 mg/L fluoride, tissues measured 10 mg/kg fluoride, at 10 mg/L fluoride in the water the tissue level rose to 18 mg/kg fluoride, and at 20 mg/L fluoride in the water the body level reached 30 mg/kg. These represent concentration factors of 1.5 to 2.0 times the water concentration and were attained in a relatively short time (Wright, 1977).

Catla catla fry, 2.5 to 4.0 cm long and weighing 900 to 1200 mg, were raised in water of pH 7.0 to 7.3 at 37°C for 24 to 96 hours with fluoride added to bring levels of fluoride up to 0.6 to 13.0 mg/L. The results of the experiment indicated that tissue fluoride levels increased with increased duration of the experiment and increased water fluoride levels according to the relationship: Natural log of mg/kg in fish = 4.2133 + 0.1389 mg/L fluoride in the water + 0.0056 times the days of exposure. This equation gives results on an ash weight basis; wet weight equivalents would be about 0.07 of these values (Pillai and Mane, 1985).

For example for 96 hours at 13.0 mg/L fluoride in the water, the wet weight tissue level of fluoride is 51 mg/kg. For only 1.3 mg/L fluoride in the water the tissue level is 6.6 mg/kg; at this concentration 0.8 kg of fish would constitute the full daily dose of fluoride for an adult.

Literature Criteria

The criteria from the literature are summarized in Table 5.1. CCME has no recommendations for the effects of fluoride on aquatic life. In the marine environment, levels exceeding 1.5 mg/L are considered a hazard by EPA (Anon, 1973). The surface fresh water quality objective for fluoride in Manitoba was 1.0 mg/L for all areas and types of use (Anon, 1979), and 1.5 mg/L for class 2B waters suitable for fisheries and recreation (Anon, 1950). Alberta and Saskatchewan have surface fresh water quality objectives of 1.5 mg/L for all uses (Anon, 1975; Anon, 1977; and Anon, undated). No mention is made of hardness or temperature levels with regard to these objectives. They may be suitable for some invertebrates and coarse fish, but are not adequate for trout of the genus Oncorhynchus (Salmo) found in soft, coastal BC waters.

Recommended Criteria

Marine
The total fluoride concentration of marine waters should not exceed 1.5 mg/L (Anon, 1973; and Anon, 1968).

Freshwater
The total fluoride concentration of fresh waters should not exceed 0.2 mg/L when hardness is less than 50 mg/L or 0.3 mg/L when hardness is greater than or equal to 50 mg/L (Angelovic et al., 1961b; Anon, 1973; and Pimental and Bulkley, 1983a). This freshwater criterion is tentative and deliberately conservative. It is designed for 'worst case' soft, coastal waters where Oncorhynchus species (Salmo ) reproduce. It applies everywhere in BC where natural, uncontaminated background levels of fluoride do not exceed the criterion.

It is recommended, as a high priority research topic, that carefully controlled experimental work be carried out in reasonable fluoride, hardness and temperature ranges, to determine fluoride criteria as a function of temperature and hardness. In areas where natural fluoride levels exceed the criterion, either the natural populations of organisms will have evolved to handle this extra fluoride, or naturally higher water hardness levels will have reduced the bioavailability of fluoride to the organisms.

Rationale

Marine
The natural fluoride level in unpolluted open ocean water is in the 1.3 to 1.4 mg/L range (McNeeley et al., 1979; Benefield et al., 1982; Bewers, 1971; and Warner et al., 1975). In estuarine areas, the levels are usually lower due to dilution by fresh water unless fluoride pollution occurs upstream. Marine organisms accumulate fluoride even at ambient ocean levels and accumulation rises markedly if higher fluoride levels are present. Biomagnification also occurs as fluoride moves up the food chain, at about 1 order of magnitude per level (Stewart et al., 1974). Estimated safe levels of fluoride in the diet of man, based on blood plasma levels, are 0.053 to 0.076 mg/kg (Rose and Marier, 1977). Based on this intake level, approximately 70 mussels would contribute the full daily fluoride intake limit for man if they were grown in normal seawater with fluoride levels of 1.3 mg/L (Barbaro, 1981).

Mullet living in normal seawater had measured fluoride levels of 1.8 mg/kg wet weight in muscle tissue; levels in other tissues were an order of magnitude higher (Milhaud et al., 1981). At this level, a 70 kg man could get his full daily fluoride limit from as little as 2 kg of fish. Blue crab muscle tissue may reach 2.5 mg/kg wet weight of fluoride when the water does not exceed a normal 1.5 mg/L F. At this level, daily intake of about 1.5 kg of crab meat would meet the dietary fluoride limit or 2.0 kg of crab meat would meet the total fluoride limit from all sources.

A 100 kg seal eats about 3 kg of fish per day (a 600 kg steller sea-lion about 15 kg per day). At the upper safe limit of 0.076 mg/kg of fluoride (Rose and Marier, 1977), this works out to (100 x 0.076)/3=2.5 mg/kg fluoride in the fish. The mean fluoride level in the fish tissues consumed by seals and sea-lions is in excess of 2.5 mg/kg when the fish are grown in open, unpolluted seawater. For high energy-using predators like eagles, which also live on fish, the problem is even more acute. There is no extra capacity for seawater to accept additional fluoride. For marine and estuarine environments, "there is evidence that fluorides are accumulative in organisms. It is tentatively suggested that allowable levels should not exceed those for drinking water" (Anon, 1968).

Freshwater
The most sensitive LC₅₀ test for fluoride intoxication appears to be rainbow or brown trout fingerlings, or adult males, at 20°C in soft water for at least 10 to 20 days. The closest approximation found in the literature to this most sensitive situation are rainbow trout 20 day LC₅₀ tests conducted at 18.3°C in water with a hardness of approximately 44 mg/L as CaCO₃ (Angelovic et al., 1961b). The LC₅₀ from this experiment was 4.8 + 2.5 mg/L fluoride (95% confidence limits). Using the relationship calculated by Pimental and Bulkley (1983a), (LC₅₀ fluoride = 92.57 Log₁₀ Hardness-51.73) this value could be expected to drop to 2.0 in water of hardness 10. This hardness level is typical of soft-water coastal BC rainbow trout streams. Such streams would not usually be at 18°C but more likely about 12°C. This lower temperature should raise the LC₅₀ value by about a factor of 2 (Angelovic et al., 1961b) to about 4.0 mg/L fluoride.

A factor of 0.05 is commonly used to convert LC₅₀ values to chronic thresholds for substances which accumulate (Anon, 1972). Application of this factor gives a value of 0.2 mg/L fluoride. A factor of 0.01 has also been proposed (Anon, 1979) for such substances, with 0.05 being used for non-persistent, non-accumulative substances. Application of this factor would result in a value of 0.04 mg/L fluoride. The value 0.2 mg/L fluoride is greater than the mean value for all lakes and rivers in BC. The value 0.04 mg/L fluoride is lower than any lake or river mean value in any area of BC, and is not a practical value. If normal uncontaminated background fluoride levels exceed 0.2 mg/L then organisms living there will have adapted to these high fluoride levels (Neuhold and Sigler, 1960). Water with normally low fluoride levels should not be increased beyond 0.2 mg/L since such levels will stress unadapted organisms.

Apart from naturally high background levels, 0.2 to 0.3 mg/L in the Okanagan Valley, only one coastal and three interior water samples showed fluoride levels over 0.2 mg/L (MOE), excluding contaminated areas around Trail and Kimberley. In interior areas where natural hardness levels are higher and natural fluoride levels are also higher, a higher criterion level for fluoride is not judged to be harmful. Thus a level of 0.3 mg/L in areas where natural hardness levels exceed 50 mg/L is not likely to stress organisms already adapted to fluoride levels in this range.