Blackwater and low pH tolerance

Peacock Bass Environments

Peacock Bass In Acidic Amazon Waters

peacock bass water
Click image for larger picture

The more anglers know about their quarry, the better they are able to both successfully pursue it while managing its conservation. The giant peacock bass (Cichla temensis) roams a blackwater environment that is so significantly different from that of temperate zone freshwater sportfish, that it is worthwhile for peacock bass anglers visiting the Amazon to gain an understanding of the peacock's home waters. The following article attempts to provide some insights via a research project assessing fishes' acid tolerance in blackwater environments Throughout North America, Europe and Asia, pollution has caused serious damage to aquatic ecosystems. One of the worst culprits is acid rain. Resulting mostly from sulfur emitted by power plant smokestacks, this toxic acidification has been shown to cause massive fish kills and a serious loss of biodiversity in North American lakes, rivers and streams.

On the other hand, in the Amazon basin, highly acidic "blackwater" regions exist that support a huge diversity of fishes in spite of being far more acidic than even the temperate zone’s most damaged waters. In fact, this is the preferred home of the giant peacock bass. The most powerful freshwater gamefish in the world, lives in water with enough acid content to kill most species! The obvious question you might ask is, "How is this possible?" Many researchers have asked the same question. The answer may lie in the tea-colored material that gives blackwater its name.

Blackwater is formed when wet, oxygen-poor soils permit the slow decay of matter from vascular plant material. Cyclical flooding of the surrounding low-lying forests (igapó), coupled with runoff delivers a constant supply of this mixture of dissolved organic matter (mostly made up of tannic and humic acids). Not only does this material deliver blackwater's characteristic coloration, but scientists have found convincing evidence that it actually protects fishes against the poisonous effects of acidic environments.

Acid water causes fishes to lose their body salts. Freshwater species have a biological pumping system in the cells of their gills that keeps the salt in their bodies from leaking out into the salt-free freshwater that surrounds them. Acidic conditions attack these cells and cause them to stop working. The material in blackwater, however, appears to provide a protective effect for these cells, enabling them to continue to work normally. The peacock's own ecosystem may be what protects it from environmental toxicity that kills fish elsewhere.

The Amazon is a giant enigma, with thousands of interlocking puzzles waiting to be solved. We haven't even begun to understand how they fit together. Here is just one more reason why it must be protected at all costs. With more study, we might learn how to use Amazon-based knowledge to protect fishes in each of our various backyards. Perhaps we'll find that reducing the constant deforestation in our countryside might put more of these blackwater materials into our waters and help slow the rate of environmental degradation and fish loss.

Note - The following research paper is the result of a study performed on non-Amazon fishes, with an eye toward understanding more about the nature of Amazon Blackwater systems. The reference materials cited in this paper can provide additional information regarding this subject matter from peer-reviewed sources.

Laboratory Analysis of the Effects of Blackwater on Low pH Tolerance in Fishes

PAUL REISS; Rutgers University, Graduate Program in Ecology and Evolution, New Brunswick, NJ, 08901, USA
Rutgers University Marine Field Station, Tuckerton, NJ


The unusually high level of fish biodiversity found in acidic "blackwater" systems in the Amazon basin suggests that the humic and fulvic acids in blackwater may provide some form of protection against the toxic effects of low pH, or that fishes endemic to this environment may be more tolerant of those effects. These ideas were tested by two experiments in a laboratory study. In the first experiment, seven fish species from three water types were subjected to a treatment regime of reduced pH to compare the species' tolerance to pH toxicity. Species examined included: Enneacanthus obesus, Micropterus salmoides and Aphredoderus sayanus from blackwater; Fundulus heteroclitus, Menidia menidia and Cyprinodon variegatus from brackish water and Lepomis macrochirus from clear freshwater. The results demonstrated markedly different resistance to mortality in low pH among the species, as measured by the cumulative concentration of excess H+ ion over time. For example, Enneacanthus was able to tolerate almost three times as much exposure as Lepomis, a member of the same family, and over eight times the exposure of Cyprinodon, a brackish water fish. The results also demonstrated that fishes from blackwater are more resistant to low pH toxicity, as a group, than fishes from other source waters.

In a second experiment, the effect of water type on tolerance to low pH was measured among a subset of species selected from the first experiment, i.e., Fundulus heteroclitus, Cyprinodon variegatus and Lepomis macrochirus. Resistance to mortality ranged from 20% to 100% greater in both blackwater and brackish water than in clear freshwater for each species. These results indicate that there are effects inherent in both blackwater and brackish water that protect fishes against low pH and which are lacking in clear freshwater. The study examines the physiological aspects of pH toxicity in various water types, considers differences in innate or acquired tolerance to low pH among species and analyzes the relevance of ecosystem management strategies in relation to the toxic effects of acidification.


Acidification of water by acid rain is detrimental to freshwater ecosystems and the survival of fishes (Schindler 1988). Low pH toxicity has been studied in a variety of North American species, including trout (Playle and Wood 1989, MacDonald 1983), and smallmouth bass (Hill et al 1987) in the northeastern U.S. Although the physiological nature of these effects on fish gill epithelium is fairly well understood (Evans 1999; Freda 1988), the larger scale effects of acid rain and lowered pH on freshwater and estuarine ecosystems is more problematic (Moiseenko, 2004) and continues to be a topic of ongoing investigation. Conversely, in the Amazon basin, highly acidic "blackwater" regions exist that support a large diversity of fishes in spite of very low pH (Wilson 1999). This raises general questions regarding the effects of blackwater on pH tolerance (Gonzalez et al. 2006), some of which will be examined here.

Wilson and Wood (1999) studied the impact of lowered pH on three species of Amazon fish with varying degrees of environmental contact with blackwater. They concluded that acid tolerance is related to a species' life history in blackwater and is not simply a characteristic of all fish in the Amazon. Further, it has been theorized that the dissolved organic compounds associated with black water (primarily humic, fulvic and other organic acids) may exert a protective effect against low pH toxicity on fishes in blackwater environments (Gonzalez, et al, 2002). These protective effects were investigated by measuring ionic flux rates under changing pH conditions in Amazon stingrays (Potamotrygon sp.) by Wood (2003) and in a variety of Amazon aquarium species, including silurids, characids and cichlids by Gonzalez (2002). These studies indicate that blackwater appears to be of beneficial protective value to pH tolerance in both elasmobranches and teleosts. This paper reports on two experiments that were undertaken to specifically test these theories on various species found in the Mullica River - Great Bay watershed of New Jersey. Portions of this system have naturally occurring blackwater with low pH and associated native fauna (Able 1999). Tolerance to low pH was compared among seven species collected from freshwater, brackish water and blackwater. The differences in effects of freshwater, blackwater and brackish water on low pH tolerance in three selected species was then examined in the second experiment.


Sufficient water volumes were collected to fill 1000 liter holding tanks for each of the water types. To reduce relative variability, water samples were selected from sources that provided reasonably similar water parameters, with the exception of the desired distinguishing parameter (i.e. salinity or blackwater content) (Table 1). A multi-parameter YSI Model 556 MPS meter system with its sensing probe arrayed for pH, conductivity/salinity, dissolved oxygen and temperature recording was used to measure water parameters. The unit was calibrated weekly during the experimental period. Selected sample results following calibration were compared with a similarly equipped YSI 600 series meter for confirmation of accuracy.

Blackwater was collected from Lake Pohatcong in Tuckerton, NJ. Samples from this source had greater light absorbance than other known blackwater sources within the watershed. This indicated the likelihood of a greater concentration of the dissolved organic matter known to cause blackwater's typical coloration (Ertel 1986), and provided the largest available differential in this parameter compared to the other water types used. Light absorbance of each of the waters collected was assayed on a Beckman DU-64 Spectrophotometer at 440nm. The absorbance of the collected blackwater was shown to be more than 20 times greater than that of the other two water types examined, specifically indicating a far higher concentration of humic substances, a primary blackwater component (Cuthbert, 1992).

Since reference freshwater was used as both a control and a diluent for parts of this study, a source in Lebanon Township, New Jersey was selected because of its similarity in both conductivity and pH to the blackwater source, and its low absorbance in the range of blackwater components.

Collection of naturally occurring brackish water that was not mixed with the blackwater water ubiquitous throughout the estuary proved problematic. Therefore the brackish water used in this study was obtained by diluting water taken from the estuary at a relatively high salinity (approx. 23.5) and low absorbance at 440 nm and mixing it with aliquots of the reference freshwater (essentially devoid of any blackwater components). The resultant solution yielded salinity similar to that at which the brackish water specimens were collected (approx. 1.2) while remaining essentially free of the blackwater components.

The collected water was stored and used for holding specimens. The stored water provided aliquots for the preparation of each treatment tank at the time of analysis. The water parameters were periodically monitored with the YSI 555 system. During the period following collection and prior to analysis, pH and conductivity/salinity of each water type was found to increase. This was probably due primarily to evaporation and to a lesser extent, the presence of an active fish population over several weeks without replenishment of acidity from outside sources (Schindler 1988). Dissolved oxygen content increased due to artificial aeration (Table 1).

Table 1: Source Water Characteristics at Collection and Prior to Treatment
Water Conductivity Salinity Dissolved O2 pH Abs. @
Initial Final Initial Final Initial Final Initial Final
Ref. FW 0.078 0.126 0.04 0.06 7.23 8.83 6.43 7.47 0.002
Blackwater 0.061 0.112 0.03 0.05 6.65 8.83 4.62 7.23 0.105
Brackish 2.245 2.600 1.16 1.35 7.76 7.76 6.86 7.38 0.005
Table 1: Source Water Characteristics at Collection and Prior to Treatment - Water parameters (as measured by the YSI 555 system) changed between collection and use. The initial values at time of collection are shown next to the final values measured at the time of testing for each parameter. The light absorbance of each of the water types was assayed on a Beckman DU-64 Spectrophotometer at 440nm. The absorbance of the collected blackwater is more than 20 times greater than that of the other two water types examined, indicating a higher concentration of humic substances, and was used as the defining component of blackwater.

Treatment tank pH levels were adjusted in each 38 liter glass tank prior to the addition of specimens. Sulfuric acid (H2S04) was added while utilizing the YSI meter to measure pH, effectively creating a real-time titration system. Tanks were artificially aerated during the preparation period to ensure that adequate oxygen was available to the specimens during the one hour treatment period, when no artificial aeration was provided. Pre-treatment dissolved oxygen (DO) concentration was typically 9.0 mg/L while post-treatment DO concentration dropped to approximately 8.0 mg/L.

Table 2: Amounts of Sulfuric Acid needed for pH changes
pH Ref. FW Blackwater Brackish
From Start to 5.0 10.0 ml 6.0 ml 45.0 ml
6.0 to 4.4 9.0 ml 7.0 ml 15.0 ml
5.0 to 4.0 2.5 ml 3.0 ml 3.0 ml
4.4 to 3.7 3.5 ml 3.5 ml 4.0 ml
4.0 to 3.4 6.0 ml 6.0 ml 6.0 ml
3.7 to 3.2 9.0 ml 9.0 ml 8.5 ml
3.4 to 3.0 13.0 ml 13.0 ml 11.0 ml
3.2 to 2.85 20.0 ml 18.0 ml 15.0 ml
3.0 to 2.7 20.0 ml 27.0 ml 25.0 ml
2.85 to 2.6 40.0 ml 35.0 ml 22.0 ml
Table 2: Amounts of Sulfuric Acid needed for pH changes - A 3% solution of sulfuric acid was used for pH adjustments. The amounts of H2S04 required varied with water type, pH level and the magnitude of pH change. The pH levels in the first column represent the adjustment made to the alternating treatment tanks used. Adequate acid additions were required in each tank to adjust for the change over two pH levels. The volume measurements recorded reflect that span. Initial pH changes required greater acid additions to overcome buffering effects in each water type. After pH 5.0, buffering was exhausted. Subsequent acid volumes added reflect the amount of excess hydrogen ion concentration needed to reach the next pH level.

The adjustment of pH in the treatment tanks required amounts of H2S04 that varied with water type, pH level and the magnitude of pH change (Table 2). Tanks for each treatment were arrayed in pairs, with one tank available for pH adjustment while the other was in use. The alternation of treatment tanks required adjustment of pH across two pH levels at every hourly interval. The volume of H2S04 used for each adjustment was recorded. All three water types required larger volumes of H2S04 for the initial lowering of pH due to the water's natural buffering capacity. Brackish water required the most H2S04. As pH reached 5.0, buffering capacity was diminished and smaller volumes of H2S04 were necessary for adjustment at subsequent levels. Further pH decreases required H2S04 additions proportional to the desired H+ concentration.

peacock bass environmental research
Figure 1 - Specimen Transport System

A fish transport system was used in order to avoid physical injury to specimens between treatments, to transport all specimens simultaneously and to enable separation of fish groups within each treatment. The system consisted of three flexible plastic mesh cubes, each divided into 3 equal sized sections by mesh panels. One cube was dedicated to the treatment regime in each water type. The system allowed water to pass through unencumbered and allowed access to the full volume of water in each treatment. The small size of the fishes and the relatively large water volume (28.5 L in each 38 L tank) readily allowed accommodation of 45 specimens per treatment tank (Figure 1).


Two experiments were performed on specimens selected from the same population of collected fishes. The first experiment was designed to compare the differences in low pH tolerance among each of the species collected. The second experiment was designed to determine the effect of water type on low pH tolerance in three selected species. Although the hypothesis tested in each experiment (and consequently the design and analysis) differed, each used the same treatment regime, source specimen population and water sources. Both experiments treated several groups of small fishes derived from each of the water types (blackwater, reference freshwater and brackish) with a sequentially lowering pH regime.

The pH treatment levels utilized were selected to maximize the resolution of specimen responses to pH exposure in relation to pH as a logarithmic representation of relative excess H+ concentration. For example, pH 7.0 was considered to have a relative excess H+ concentration of 100, essentially a nil concentration of excess H+. pH 6.0 therefore was considered be 101, or a relative excess H+ factor of 10. pH 5.0 therefore was considered be 102, or a relative excess H+ factor of 100 (Table 3). The pH levels used were selected with proportionately reduced intervals between levels as pH became lower, in order to maximize linearity and maintain proportionality of the differential in H+ concentration between treatments.

Table 3: Treatment pH levels (for both experiments)
Reference Freshwater (6.97) Brackish water (7.35) Blackwater (6.78)
pH level Power H+ Con pH level Power H+ Con pH level Power H+ Con
pH 6.0 101 10 pH 6.0 101 10 pH 6.0 101 10
pH 5.0 102 100 pH 5.0 102 100 pH 5.0 102 100
pH 4.4 101.6 398 pH 4.4 101.6 398 pH 4.4 101.6 398
pH 4.0 103 1000 pH 4.0 103 1000 pH 4.0 103 1000
pH 3.7 103.3 1995 pH 3.7 103.3 1995 pH 3.7 103.3 1995
pH 3.4 103.6 3981 pH 3.4 103.6 3981 pH 3.4 103.6 3981
pH 3.2 103.8 6310 pH 3.2 103.8 6310 pH 3.2 103.8 6310
pH 3.0 104 10000 pH 3.0 104 10000 pH 3.0 104 10000
pH 2.85 104.15 14125 pH 2.85 104.15 14125 pH 2.85 104.15 14125
pH 2.70 104.3 19953 pH 2.70 104.3 19953 pH 2.70 104.3 19953
pH 2.60 104.4 25119 pH 2.60 104.4 25119 pH 2.60 104.4 25119
      pH 2.50 104.5 31623      
Table 3: Treatment pH levels - The treatment regime shown was used for both the species sensitivity experiment and the pH tolerance experiment. Both experiments treated several groups of small fishes derived from each of the water types (blackwater, reference freshwater and brackish) with a sequentially lowering pH regime as shown above in the first column for each water type. The second columns in each set display the reciprocal power of each treatment level. The pH treatment levels utilized were selected in order to maximize the resolution of fishes' responses to pH exposure. The selected intervals between levels were reduced as pH became lower, in order to maximize the linearity and maintain the proportionality of the differential in H+ concentration between treatments. The third column displays the relative hydrogen ion concentration. The pH level selected was identical for each of the three water types. The pH 2.60 and 2.50 levels were used only with Enneacanthus.

Blackwater fishes were collected from Lake Pohatcong in Tuckerton, NJ (in the same area as the source water) and included: Enneacanthus obesus (banded sunfish); Micropterus salmoides (largemouth bass) and Aphredoderus sayanus (pirate perch). Brackish water fishes were collected in a marsh area contiguous with Tuckerton Seaport and included: Fundulus heteroclitus (mummichog); Menidia menidia (Atlantic silverside) and Cyprinodon variegatus (sheepshead minnow). Lepomis macrochirus (bluegill sunfish) was collected (along with the freshwater used) from a clear freshwater pond in Lebanon Township, NJ.

Fishes collected from each environment were held in large (1000 liter) holding tanks with pH ? normal for the water type, at the Rutgers University Marine Field Station prior to treatment. Fishes were acclimated for a minimum period of one week to reduce variability due to possible capture injury or failure to tolerate aquaria. Specimens selected at the time of collection were small in size (3 - 8 cm) to allow an adequate number to be supported in each of the holding and treatment tanks. Larval fishes are considered to be more susceptible to low pH toxicity because of their additional ion exchange locations (Hill 1988). It is possible that smaller or very immature specimens would therefore be disproportionately susceptible to low pH than larger specimens. To avoid confounding by size or stage of maturity, consistency in size and maturity within and across species was maintained. None of the fishes retained at collection or used in the experiments were larval. All captured fishes retained in the holding tanks were judged to be in a stage of maturity ranging from large juvenile to small adult.

Experimental subjects were randomly selected from the pool of retained captured specimens. All treatments were processed sequentially on all selected specimens without returning to the normal pH environment of the holding tanks between levels of treatment. This regimen ensured the continuity of the same specimens for each of the treatments, reducing variability and the possibility of confounding due to potential after-effects of a treatment.

Table 4: Species Sensitivity Experiment; Species tested and Treatments applied.
Source Water Group Species Included Treatment 1
Dropping pH in Ref. Fresh water
Treatment 2
Dropping pH in Brackish water
Treatment 3
Dropping pH in Blackwater
Specimen Total - All Water
Blackwater fishes Enneacanthus obesus 6 replicates 7 replicates 5 replicates 18
Micropterus salmoides 1 replicate 1 replicate 1 replicate 3
Aphredoderus sayanus 1 replicate 1 replicate 1 replicate 3
Brackish water fishes Fundulus heteroclitus 4 replicates 4 replicates 5 replicates 13
Menidia menidia 5 replicates 5 replicates 5 replicates 9
Cyprinodon variegatus 5 replicates 4 replicates 5 replicates 14
Freshwaterfishes Lepomis macrochirus 10 replicates 10 replicates 8 replicates 28
Table 4: Species Sensitivity Experiment; Species tested and Treatments applied. - Species collected from the three source waters used in the species sensitivity experiment are shown along with the number of specimens tested for each species in each water type. Replicates for several species were limited by specimen availability.

Species Sensitivity Experiment - Specimens from each of the water types were randomly selected for the treatments shown in Table 4. This study measured the low pH tolerance of different species from different source water groups. Ambient pH was lowered by the addition of dilute (3%) sulfuric acid (Freda 1988; Hill 1987) into 38 liter treatment tanks containing 28.5 liters of the source water of each water type. Two alternating treatment tanks were used for each water type at each of the prescribed pH levels, allowing adjustment to the selected pH level prior to the addition of specimens and ensuring that specimens would not be exposed to fluctuating pH levels during treatment tank pH level adjustments. A transfer system (as described in Materials) was used to ensure that specimens would not be damaged by the physical effects of netting and handling when changing tanks. Fishes remained in this transfer system for the duration of the experiment. Fishes were transferred into each treatment tank for one hour periods at the specified pH. At the end of each period, all specimens from each of the treatment groups were transferred simultaneously into the next level treatment tank. Specimen behavior was observed and noted during each period until exposure became fatal (the measured response). Observation was performed in two forms. Behavior was observed from a distance, without impinging on the open space above treatment tanks and without eliciting any observable fright response. Physical status was observed directly from the open space above each treatment tank. This elicited an overt and observable fright response. Observation was essentially continuous and alternating in form. Mortality was defined as the cessation of respiration, swimming and response to touch. Time and pH level at death were recorded.

pH Tolerance Experiment - The same methodology and pH treatment schedule as previously described was used to test the differences in fishes' pH tolerance among different water types. Based upon the results of the species sensitivity experiment and availability of adequate numbers of specimens, three species, shown to be reasonably acid-tolerant, were selected for this experiment. Fifteen specimens each of Fundulus heteroclitus, Cyprinodon variegatus and Lepomis macrochirus were subjected to each of the pH treatments in each of the three water types. The protocol used tested 45 specimens of each species across the three water types, totaling 135 experimental subjects.

Hydrogen Minutes
Graph 1 - Total Hydrogen Minutes of Exposure Generated by Treatment Structure - As specimens progressed through the stepwise treatment regime, they were exposed to increasingly greater concentrations of excess hydrogen ion, as exhibited by the stepwise progression of calculated hydrogen ion exposure plotted over time. The toxic effects are cumulative, therefore the total area under the curve until the time of mortality is recorded, represents a specimen's total hydrogen ion exposure (THIE) and is expressed in terms of Hydrogen Minutes (HM). Sample pH levels are superimposed on the y-axis to illustrate the logarithmic relationship between pH and relative hydrogen ion concentration. Hours are superimposed over the x-axis to coincide with the stepwise reduction of pH.

Total Hydrogen Ion Exposure - Hill et al (1988) dTotal Hydrogen Ion Exposure - Hill et al (1988) demonstrated that the total time of exposure to low pH, combined with the degree of exposure, affects the survival of fishes. This indicates that a true measure of low pH tolerance in fishes must consider both the degree of exposure and the time of exposure. This form of measurement of tolerance to toxicity (or inversely drug efficacy) is used in both pharmacology and clinical laboratory analysis (Tietz 1994). It is displayed in graphical form as the plot of the concentration of an active agent over time (in this study measured in minutes). The total exposure to the active agent (in this study H+) is represented by the area under the curve produced by this plot. The general equation representing this calculation is "The Definite Integral of the Area under the Curve" (Abbot and Neil 2003) and is expressed as;

equation 1

In both experiments performed in this study, the measured response to the treatments was mortality of the subject. The response was recorded as the final level of pH tolerated and the total time of exposure until mortality. The unitless expression pH is the logarithmic expression of the relative concentration of hydrogen ion in a solution. The degree of exposure to low pH is therefore expressed in linear form as the relative concentration of H+. Consequently, the total hydrogen ion exposure (THIE) is measured as the H+ concentration integrated over time in units of Hydrogen Minutes (HM) and expressed as;

equation 2

Since pH, and therefore relative H+ concentration, was changed in a stepwise fashion during the treatments in this study, the resultant plot is a series of steps (Graph 1) where area can be calculated arithmetically by summing the rectangular areas under the curve. The equation used to calculate total hydrogen ion exposure within the treatment structure of this experiment can be expressed as;

The calculation of total hydrogen ion exposure (THIE) provides a linear measure of pH toxicity tolerance by integrating relative hydrogen ion concentration with time. The final calculated values of THIE, expressed in units of Hydrogen Minutes (HM) are proportional to the low pH tolerance of the fishes observed in this experiment. Higher values coincide with greater tolerance.


Species Sensitivity to Low pH
Graph 2: Species Sensitivity Experiment Results - The average THIE to mortality in HM for each species is displayed at the top of each column. Results for each water type were pooled to provide a wider range of information, therefore standard error (represented by error bars) is greater than that otherwise occurring within individual water types for each species. ANOVA revealed a significant difference between species F(6, 81) = 41.60, p<.01. Tukey's Post hoc analysis revealed homogeneous subsets; Menidia and Cyprinodon; Cyprinodon, Fundulus ,Lepomis; Lepomis, Aphredoderus; Aphredoderus, Micropterus and Micropterus, Enneacanthus. Differences between homogeneous subsets were collectively significant at p<.05.

The behavior of fishes was observed and noted during both experiments and throughout the course of all treatments. Comparisons between the fishes being treated were made against the behavior of fishes remaining in the 1000 liter holding tanks as well as in a separate series of 38 liter tanks used in the random selection process. These tanks remained at circumneutral pH and served as control tanks for each water type. At pH levels ranging from circumneutral to 5.0 in all water types, the behavior of all specimens appeared to be indistinguishable from that of conspecifics that remained in the circumneutral pH tanks. Fishes were distributed throughout the water column and moved readily within the treatment tanks, as they did in circumneutral pH. Specimens appeared to be unstressed. Interactions among specimens were ongoing and included aggression as well as predation. Fright responses, as defined by rapid startling, cessation of other activity, and subsequent attempts to hide in the corners and bottom of the transport system, were noted when specimens were observed from directly above each treatment tank, where there was impingement on the open space above the tanks.

At treatment levels below pH 5.0, differences in behavior began to appear between species and water types. In freshwater at pH 5.0, Lepomis macrochirus, Menidia menidia and Cyprinodon variegatus began to congregate at the surface while they continued to remain distributed in both blackwater and brackish water. Subsequently, these same species all moved to the surface in brackish water at pH 4.4, while only Cyprinodon and Menidia did so in blackwater. At pH 4.0 Lepomis lost their fright response in clear freshwater, but remained sensitive in blackwater and brackish water, while Fundulus specimens formed into a tight school on the bottom in brackish water. By pH 3.7, the fright response was no longer observable in all species in all waters and overall activity levels diminished. All species tended to congregate at the bottom of the water column where they remained in an upright orientation, appearing to respire at a similar rate as in circumneutral pH. All fishes began to appear sluggish in their movements. As specimens reached fatal levels of H+ exposure, a short period of erratic swimming (1 to 3 minutes) typically preceded death. At the point of mortality, all apparent respiration, swimming movement and response to touch ceased.

Species Sensitivity Experiment - Low pH tolerance varied significantly between the individual fish species studied (Graph 2). One way ANOVA revealed a significant difference between species F(6, 81) = 41.60, p<.01. Tukeys Post hoc analysis revealed homogeneous subsets; Menidia and Cyprinodon; Cyprinodon, Fundulus ,Lepomis; Lepomis, Aphredoderus; Aphredoderus, Micropterus and Micropterus, Enneacanthus. Differences between homogeneous subsets were collectively significant at p<.05.

Among groups based on source water, fishes collected from blackwater showed the greatest tolerance to low pH in all of the water types, while fishes collected from brackish water showed the lowest tolerance (Graph 3). ANOVA revealed a significant difference between water type F(2, 85) = 82.45, p<.01. Tukeys post hoc comparisons showed that all three water types were significantly different p<.01. The only species collected from reference freshwater, Lepomis, demonstrated tolerance midway between the other groups. The average Total Hydrogen Ion Exposure (THIE) tolerance across all water types ranged from a high of 3,864,632 Hydrogen Minutes (HM) for Enneacanthus obesus, a native blackwater species, to a low of 138,101 for Menidia menidia, collected from brackish water, an approximate 2800% difference. The tolerance levels of the other species examined fell within this range.

Graph 3 - pH Tolerance by Source Water Group
Graph 3: pH Tolerance by Source Water Group - The average THIE to mortality in HM for all species collected from a specific source water type (group) is displayed at the top of each column (± SEM). Results for each water type were pooled to provide a wider range of information, therefore variance is greater than that otherwise occurring within individual water types for each species. ANOVA revealed a significant difference between water type F(2,85) = 82.45, p<.01. Tukey's post hoc comparisons showed that all three water types were significantly different p<.01.

Among the blackwater fishes, Enneacanthus obesus demonstrated the highest tolerance to low pH of any species tested. The THIE tolerated ranged from 3,137,540 HM in reference freshwater to 4,429,000 HM in blackwater. This species proved so resistant to low pH that its responses occurred far beyond the limits of other species, even outside the initially planned scope of treatments. The other two blackwater species tested, Micropterus salmoides and Aphredoderus sayanus demonstrated average THIE's of 2,860,434 HM and 2,667,559 HM, respectively. These results were higher than any of the non-blackwater species. It must be noted that, unlike all of the other species examined, Micropterus and Aphredoderus were collected in limited numbers. As a result, testing was performed on only three samples of each. The blackwater group, taken as a whole, averaged a THIE tolerance of 3,130,875 HM.

The only species collected from freshwater, Lepomis macrochirus demonstrated an average THIE of 953,277 HM in freshwater, 1,588,311 in blackwater and 1,994,840 in brackish water during the species sensitivity experiment. The overall average tolerance in all water types for Lepomis was 1,506,702 HM, placing its tolerance, as the only member of the freshwater group, between that of the other two groups.

The brackish water group was the least tolerant of low pH, with an average THIE of 770,937 HM. Individual species within the group, however, differed significantly. Menidia Menidia, Cyprinodon variegatus and Fundulus heteroclitus had average THIE tolerance values of 138,101 HM, 668,009 HM, and 1,529,218 HM, respectively. Menidia was the most pH intolerant of any of the species tested. Its tolerance ranged from 80,480 HM in blackwater to 238,976 HM in brackish water. Menidia was the only species with tolerance not greater in blackwater than in reference freshwater. The other two brackish water species both demonstrated greater tolerance in blackwater than in freshwater, and interestingly, even greater than in their native brackish water.

Although low pH tolerance varied significantly between the different fish species studied, the variability within each species was not great. Standard error for all species ranged between 2% to 12% of the species' average THIE. The only exception was Enneacanthus, at 26% and Lepomis at 19%. This is likely due to the relatively extreme pH levels (at both ends of the scale) at which mortality was reached with these species. The grand average of the pH tolerances of all individual specimens from each of the species in the sensitivity experiment taken together, were greater in blackwater (2,198,145 HM) and brackish water (1,886,871 HM) than in clear freshwater (1,558,301 HM).

pH Tolerance Experiment - This experiment focused on whether there are differences in the low pH tolerance of fishes when placed in different water types. Tolerance was found to be significantly greater in blackwater and brackish water than in freshwater for each of the three species (Fundulus, Lepomis and Cyprinodon) tested. Although the overall average levels of tolerance remained different among each of the species, as was seen in the earlier species sensitivity study, the effect of the different water types on pH tolerance followed the same, qualitatively consistent trend for each species (Graph 4) with blackwater enabling significantly longer survival than freshwater.

Graph 4 - pH Tolerance Experiment Results
Graph 4 - Effect of Water Type on pH tolerance - Mean Hydrogen minutes (± SEM) for Cyprinodon, Fundulus and Lepomis in freshwater, blackwater and brackish water. The average THIE to mortality is displayed in HM at the top of each column for the 15 replicates of each species treated in each water type. n = 15 specimens of each species for each water type. A total N=135 total specimens were used. ANOVA revealed a significant interaction between water type and species p<.01. There was a significant main effect of water type and species p<.01. Post hoc tests revealed significant differences between all three water types p<.01 and a significant difference between Cyprinodon and Fundulus and Cyprinodon and Lepomis p<.01.

Water type significantly affected species tolerance to pH. ANOVA of hydrogen minutes for Cyprinodon, Fundulus and Lepomis in freshwater, blackwater and brackish water showed a significant interaction between species and water type (F (4,127) = 19.076, p<.01). There was also a significant main effect of water type (F (2,133) = 14.925, p<.01) and species (F(2,133) = 101.425, p<.01). Tukey's post hoc tests revealed significant differences between all three water types (p<.01). Means (± SEM) for freshwater, blackwater and brackish water were 581825.4 ± 2451.97, 967627.5 ± 4108.56, 1265905 ± 4682.93 respectively. There were also significant differences between Cyprinodon and Fundulus and Cyprinodon and Lepomis (p<.01).

The least tolerant species in this experiment was Cyprinodon. Low pH was fatal to this brackish water species at a THIE of 140,492 HM in freshwater, 198,993 HM in blackwater and 329,232 HM in brackish water. Both Fundulus and Lepomis were more tolerant, with the freshwater native Lepomis showing slightly higher overall levels of tolerance. Low pH was fatal to Fundulus at a THIE of 896,796 HM in freshwater, 1,090,682 HM in blackwater and 1,657,107 HM in brackish water. In freshwater, Lepomis tolerated a THIE of 708,189 HM, slightly lower than that of Fundulus. However, its tolerance was greater than Fundulus in blackwater and brackish water, at 1,572,858 HM and 1,822,395 HM, respectively. As in the species sensitivity experiment, the grand average of the pH tolerances of all individual specimens from each of the species in the pH tolerance experiment taken together, were greater in blackwater (967,627 HM) and brackish water (126,5905 HM) than in clear freshwater (581,825 HM).


The results of the two experiments of this study are complementary in nature, consistently demonstrating that fishes have greater tolerance to low pH toxicity in blackwater and brackish water than in freshwater. This principle appears to apply equally whether species are highly sensitive or highly tolerant. A large body of previous work has demonstrated that the magnitude of sodium ion loss is a key factor determining the survival time of fishes exposed to low pH. It has been shown that a sodium loss in the range of 50% to 60% is the threshold at which mortality occurs (Freda 1988). The process of sodium loss is decelerated when fishes are in brackish water, due to the greater concentration of readily accessible ionic sodium than is present in freshwater. This study demonstrated that survival time follows a similar pattern in blackwater. It appears, therefore, that fishes are also able to slow the process of sodium loss in this medium, relative to freshwater. Hence, these results support the existence of an action, effect or material ameliorative to the toxicity of low pH, which is present in blackwater and lacking in clear freshwater.

Specimen Transport System
Specimen Transport System in use
(Click image to enlarge)

Species differences - The species sensitivity experiment demonstrated that fishes from blackwater, freshwater and brackish water, as groups, showed differences in their pH tolerance. The species collected from blackwater were most tolerant of low pH. Fishes found in blackwater are routinely exposed to low pH due to the acidifying effect of naturally occurring humic and fulvic acids in blackwater (Ertel 1986). Wilson et al. (1999) postulated that fishes that spend a greater proportion of their life history in blackwater will tend to have a higher tolerance for low pH. Enneacanthus obesus, a species found primarily in blackwaters (FishBase 2006) demonstrated far greater tolerance of low pH than any other species examined in this study, lending support to Wilson's theory and suggesting an evolutionary adaptation to low pH. Interestingly, Micropterus salmoides and Aphredoderus sayanus, two species that can be found in a wide range of water types but were collected in blackwater for this experiment, also showed high tolerance to acid pH. This suggests that there may be an accommodation to low pH conditions by acclimatization (Helfman 1997) in populations of these species found in blackwater. Further investigation into the innate or acquired pH tolerance of blackwater species, as well as the physiological mechanism of its implementation, might be an interesting area for future study.

Non-blackwater species that evolved in clear fresh waters, conversely prove to be extremely sensitive to low pH (Moiseenko 2006). Lepomis macrochirus, a freshwater native typically found in circumneutral waters (FishBase 2006), was more sensitive to low pH than the blackwater natives, suggesting that no accommodation to low pH, either by evolution or acclimatization has occurred in this species. The three euryhaline species collected from brackish water displayed the lowest pH tolerance. In their native environment, these species would rarely be hyperosmotic to their external medium due to the ion-rich nature of brackish water. Adequate ambient sodium is typically readily available to protect against the extreme sodium efflux associated with mortality in low pH. Coupled with the normally circumneutral pH and the greater buffering capacity of brackish water, these fishes are also unlikely to commonly encounter conditions of low pH toxicity. This suggests that there is likely little survival value in developing mechanisms for low pH tolerance in these species, whether by evolution or acclimatization.

The species sensitivity experiment also showed that the different species examined exhibited significantly different levels of pH tolerance regardless of their source waters or treatment water. This is likely a result of variations in osmoregulatory mechanisms among species. Gonzalez (2002) identified two strategies for ion regulation in fishes from the highly acidic, blackwater Rio Negro region of the Amazon basin in Brazil. The first is a high affinity, high capacity system. As an example, the neon tetra (Paracheirodon axelrodi, a characid) maintains such a high rate of influx that it is essentially insensitive to low pH levels and is able to ionoregulate normally at pH 3.5. Conversely, the angelfish (Pterophyllum scalare - a cichlid), has a low affinity, low capacity system that is adversely affected by low pH. Its low rate of efflux, however, is so slow that it enables fish to limit sodium loss for extended periods and effectively wait out low pH exposure.

It is likely that species not normally exposed to low pH would lack adaptive mechanisms and would therefore be less tolerant to its toxic effects. This experiment demonstrates this with the fishes collected from brackish water. Both Menidia menidia, a more marine oriented species and Cyprinodon variegatus, a species that commonly occurs in hypersaline lagoons and connecting channels (Fish Base) were less tolerant than Fundulus, a species routinely associated with freshwater and mildly brackish habitat. Conversely, Enneacanthus, a blackwater native, was far more tolerant of low pH than other species in this experiment, even in non-blackwater treatments.

Effects of Water type on pH tolerance - In the second experiment in this study, all three species tested showed significantly higher tolerance to low pH in blackwater than in clear freshwater. This supports the proposition by Gonzalez (2002) that one or more of the components of blackwater may have an ameliorative effect on pH toxicity. Blackwater in temperate regions, although not likely identical to tropical blackwater, similarly contains an array of organic acids, including tannic, humic and fulvic acids that are responsible for its characteristic color and acidity. In temperate forested regions, the breakdown of leaf litter and forest debris is among the main sources for these materials (Cuthbert 1992), as is jungle vegetation in the tropical Rio Negro system.

Brackish water also ameliorated low pH toxicity in all three species in this study, likely due to the greater availability of Na+ and other ions and thus resulting in decreased osmotic pressure differences. Loss of internal Na+ is postulated to be the primary mechanism by which pH toxicity affects fishes (Evans 1999; Wilson 1999). The ionic concentration of freshwater is lower than that of the normal ionic concentration of the extracellular fluids of fishes. This differential drives the tendency to constant diffusive loss of Na+ from fishes to ambient freshwater, particularly through the gills. Under normal conditions, freshwater fishes overcome this osmotic tendency by limiting efflux through the low ionic permeability of gill epithelium in the presence of Ca+ (McDonald 1983) and by actively maintaining influx of Na+ (and Cl-). These functions are disrupted in low pH conditions in ion-poor freshwater (Matsuyo 2002), where excess H+ is present and adequate ionic sources are unavailable (Gonzalez 2002; Evans 1999).

Various chemical and cellular mechanisms have been investigated to explain how low pH toxicity triggers catastrophic sodium loss in fishes. Three key factors that occur under low pH conditions in natural waters have been identified: (1) excess H+ is present; (2) ionic aluminum (Al3+) (and other metallic ions) becomes bio-available; and (3) a deficiency of Ca++ exists. These ionic differences from circumneutral conditions in waters with differing constituents cause action at the chemical level and result in specific effects at the cellular level of biological organization. The principal toxic effect of these factors is a disturbance of ionoregulation in the gills, resulting in a net loss of sodium (and chloride) (McDonald 1983) that if unmitigated ultimately leads to mortality.

Acidity is the direct expression of the presence of excess H+. Excess hydrogen ion is known to interfere directly with ionic transport, especially of Na+. The exact mechanism for inhibition of influx remains in discussion, however it has been explained variously by competition of H+ with Na+ directly upon the ion exchange mechanism, or by some investigators as an H+ induced slowing of Na+ATPase enzyme activity, a transport channel energizer in ionic transport cells (Wood 2003). Excess H+ has an even more severe effect on efflux, greatly accelerating massive outward Na+ leakage (Freda 1988) and leading to the catastrophic losses associated with mortality at low pH.

Aluminum in circumneutral freshwater is normally bound in complex molecules and is not readily bio-available. Excess Al3+, however, becomes more bio-available at lower pH due to a decrease of its binding capacity with naturally occurring ligands, prompting its consequent release in ionic form. Although Hill et al. (1988) demonstrated that an increase in environmental acidity alone is sufficient to cause fish losses even in the absence of aluminum and other metals, Moiseenko et al (2006) describe this as a rare condition. They conclude that Al3+ is the main toxic agent at the cellular level in low pH water, rapidly causing thickened epithelia, cell necrosis and lysis, aneurysm and blood vessel obstruction as well as physical damage to chloride (ion transport) cells, effectively disrupting ionoregulation.

Calcium in ionic form modulates ionic efflux through the cell junctions in fish gill epithelial tissue. Ca++ is released from fish gill epithelium in low pH conditions, diminishing its contribution to regulation of gill membrane permeability. Competitive replacement or leaching of calcium from gill cells and intracellular junctions by H+ is thought to be a primary cause for the loss of control over Na+ efflux. Net ion loss and mortality in brown trout was shown to be much lower in hard water, (high calcium content) than in ion-poor soft water (McDonald 1983). Laboratory experimentation (Gonzalez 2002) demonstrated that addition of calcium to low pH water slowed the rate of Na+ efflux in fishes under most conditions.

These factors provide the possible bases for the toxicity of low pH in freshwater. When viewed in concert with the elevated presence of both sodium and calcium in brackish water, they can also help explain why fishes showed greater tolerance to low pH in that medium. In brackish water, the increased concentrations of Ca+ probably help to minimize efflux while the high availability of Na+ in the ambient medium would diminish the tendency toward diffusive loss while helping to maintain influx, thus ameliorating the toxicity of low pH. Blackwater, however, presents different constituent characteristics than brackish water and consequently requires a different analysis.

Blackwater is formed when acidic, wet, anaerobic soils permit the slow decay of organic matter, primarily from vascular plant material. Runoff from these sources generates continual leaching of humic, fulvic and tannic acids along with a complex mixture of other dissolved organic matter (DOM) into freshwater systems (Ertel 1986). These "blackwaters" are typically highly acidic and often ion-poor. Humic acids, insoluble in concentrated acid, can make up as much as 40 to 80 % of the DOM in blackwaters. Acid soluble fulvic acids, phenolics, lignin and other organic acids, comprise the bulk of the balance (Ertel 1986). DOM has several recognized roles in blackwater; it is responsible for the acidity associated with blackwater, it is known to be an efficient chelating agent (causing complexation of aluminum and other metals) and it adsorbs or complexes small molecules (Campbell 1997). Highly acidic and ion-poor waters are generally associated with low abundance, diversity and biomass of animal life. Fish biodiversity, however, is extremely high in the Rio Negro and this has led to speculation by Gonzalez, et al (2002) that DOM might actually be protective against the ion-poor acidic conditions found in these waters. The Rio Negro drainage in the Brazilian Amazon is darkly stained by a high content of DOM (Ertel 1986), giving the trunk river its name (Black River, in Portuguese).

DOM has been shown to accumulate on living cell surfaces at low pH, most likely via hydrogen bonding. Kullberg et al (1993) predicted that DOM would bind to biological surfaces efficiently at low pH because of the reduction of negative charge and Campbell et al (1997) found direct evidence that DOM bonded to isolated fish gill cells at low pH. Various investigators have speculated that this coating of organic material provides a direct protective action against the effects of both H+ and Al3+.

Another possible mechanism whereby DOM may be protective in low pH conditions is its known ability to bind aluminum and other potentially toxic metals (Leenheer 1980, Kullberg 1993). Aluminum in freshwater environments is becomes more bio-available under acidic conditions. DOM is a known chelating agent. Its presence in blackwater causes complexation and subsequent removal of aluminum and other metals from bio-availability. Wood (2003) concluded, based on geochemical modeling, that all of the aluminum in the Rio Negro system would be more than adequately prevented from binding to fish gill epithelium by the measured concentration of DOM present in its waters.

The pH tolerance experiment of this study, using natural blackwater and natural freshwater demonstrates that fishes' ability to tolerate low pH and survive longer, is significantly increased in blackwater and is consistent with the cited studies. Similarly, experimentation with a variety of Rio Negro species by Gonzalez (2002), Wood et al (2003) and Wilson (1999), all of whom examined Na+ influx and efflux in the presence of low pH, has shown that these processes were less affected in natural blackwater than in an artificially prepared medium made with distilled water and added equal ion concentrations, but no DOM (used as an equivalent to natural freshwater).

Ecosystems perspective - The results of the first experiment in this study indicate that individual species vary significantly in the overall level of their tolerance to low pH. The second experiment shows that, in spite of the individual species differences, all of the species tested are able to tolerate low pH significantly better in blackwater than in freshwater. This may be of value in consideration of the growing environmental impact of rapid acidification of freshwaters as a result of acid rain and other forms of environmental degradation. To evaluate the results of this study and to understand the causes and effects of acidification on fishes in natural freshwaters from an ecosystems perspective, it is necessary to address the hydrochemical aspects of the water bodies as well as the physiological processes affecting fishes. Under natural conditions many individual and interrelated factors occur together, making identification or isolation of the effect of a particular factor problematic. These considerations clarify the need to consider the results of field studies and laboratory experimentation and not rely solely on theoretical chemical, physiological and limnologic models and principles when formulating management concepts for the prevention or restoration of damage caused by acidification of freshwaters.

Freshwaters vary greatly in their ability to resist acidification (Schindler 1988). Specific ionic concentrations, hardness and the presence of dissolved and suspended materials directly affect buffering capabilities. Soil characteristics affect runoff that ultimately reaches the individual water bodies and further modulates the impact of acid rain and environmental degradation on actual water pH. This interactivity was reflected in the methodology used to prepare treatments in this study, where variability in the volumes of H2S04 necessary to adjust different water types at different pH levels was recorded during the process of artificial pH reduction.

Initial pH adjustments of all the water types required a larger acid volume than later adjustments, demonstrating a buffering capacity coinciding with that described in naturally occurring lakes, streams and estuaries (Schindler 1988). The amount varied by water type. Brackish water demonstrated a greater buffering capability than either fresh or blackwater. Additionally, elevated ionic concentration, particularly Ca2+, as present in brackish water, was shown by Wood et al (2003) to have a toxicity reducing effect, similar to that of blackwater, directly on fish gill surfaces. Although this suggests that estuarine environments and consequently estuarine and marine fish species might be more resistant to the direct impact of acidification, it must be considered that many estuarine residents, marine species and transients are dependent on spawning cycles or larval stages involving fresh waters (Able and Fahay 1998). Hill (1988) maintains that larval and juvenile fishes are more susceptible to low pH toxicity than adult fishes, further indicating the susceptibility of these species to acidification of freshwaters. Additionally, it is theorized that some fishes, especially anadromous species, are subject to an exacerbated short term toxic effect when acidified freshwater meets seawater, making aluminum suddenly bio-available and precipitating the accumulation of high concentrations of ionic Al3+ on gill tissues (Moiseenko 2006).

Following the initial buffering effect, all waters required lower acid volumes for additional pH reduction, approximately equal for each water type. All of the treatment water types lost their buffering capacity at about pH 5.0, the level at which low pH has been shown to begin being directly harmful to fishes (Hill 1988). Corresponding effects occur in natural freshwater environments (Schindler 1988), revealing the dangerous possibility of circumstances in which continuous unabated exposure to acid rain might cause increasingly rapid pH reduction following an initial buffering period.

The increased tolerance of fishes to low pH in blackwater, as demonstrated in this study, points to possible considerations for reducing the effects of acidification. Where headwaters streams and forest drainage provide large proportions of the source water for a system, the resulting blackwater entering the system may simultaneously carry an antidote for the effects of lowered pH. In clear freshwater systems without forest cover, acid rain will cause lowered pH without the presence of the potentially protective effect of blackwater's organic acids. This suggests that maintaining or restoring natural forest cover may be an important ecological mechanism to protect threatened watersheds from the toxicity of acid rain.


Different species of fishes clearly exhibit different, experimentally reproducible, and predictable sensitivity to the toxic effect of low pH. From the results of this and other studies, there is evidence that fishes from blackwater environments are less sensitive than other species. Possible causes for this effect might include; an innate, evolved, species or population-wide characteristic; the acclimatization of individuals or populations in low pH blackwater environments; or a combination of the two. Future research efforts might be designed to identify an explanation by studying different populations of the same species from different native environments.

Low pH toxicity is tolerated better in both blackwater and brackish water than in clear freshwater, by each of the species examined. Higher ionic content is a likely explanation for the lower toxicity in brackish water, while the organic acids in blackwater appear to have a protective effect directly on the gill epithelium and within the aquatic medium. Further investigation into the physiological mechanisms whereby this occurs, as well as analysis of the component(s) of blackwater responsible for this effect, may yield useful information with potential applications toward protection of ecosystems and restoration of degraded aquatic environments.


The author thanks Dr. Ken Able and Dr. Gary Taghon for their guidance in turning this experiment's logistically burdensome design into a manageable reality. Roland Hagan provided invaluable assistance in the implementation of field aspects of the project and Mark Wuenschel with fish ID. The entire staff at Rutgers University Marine Field Station deserves thanks for their patience with the disruptions caused by fire hoses pumping water from trucks to tanks, the commandeering of equipment and similar unusual interferences with their daily routines.

Literature cited

Able K W, Fahay M P (1998) The first year in the life of estuarine fishes in the middle Atlantic bight. Rutgers University Press, New Brunswick, New Jersey

Able K W, (1999) Checklist of fishes of the Jacques Cousteau National Estuarine Research Reserve at Mullica River-Great Bay. Rutgers University, Institute of Marine and Coastal Sciences Technical Report #99-22.

Campbell P G C, Twiss M R, & Wilkinson K J (1997) Accumulation of natural organic matter on the surface of living cells - implication for the interaction of aquatic biota with toxic solutes. Canadian Journal of Fisheries and Aquatic Sciences. 54, 2543-2554

Cuthbert I D, del Giorgio P (1992) Toward a standard method of measuring color in freshwater. Limnology and Oceanography. vol. 37, 1319 -1326.

Evans D H, Piermani P M, & Potts W T W (1999) Ionic transport in the fish gill epithelium. J. Exp. Zoology. 283, 641-652

Ertel J, Hedges J, Devol A, Richey J E & Ribiero M (1986) Dissolved humic substances of the Amazon River system. Limnology and Oceanography. vol. 31, no. 4, 739-754

Freda J, McDonald D G (1988) Physiological correlates of interspecific variation in acid tolerance in Fish. J. Exp. Biol. 136, 243-258

FishBase (2006) Froese, R. and D. Pauly. Editors. World Wide Web electronic publication., Cited version (10/2006).

Gonzalez R J, Wilson R W, Wood C M, et al (2002) Diverse strategies for ion regulation in fish collected from the ion-poor, acidic Rio Negro. Physio. Bio. Zoology. vol. 75 no.1, 37-47

Gonzalez R J, Wilson R W, Wood C M (2006) Ionoregulation in tropical fishes from ion-poor, acidic blackwaters. In:Val A L, Almeida-Val V M, Randall D J (eds.) The physiology of tropical fishes. Elsevier - Academic Press, London

Helfman G, Collette B, & Facey D (1997) The diversity of fishes. Blackwell Publishing Company, Malden, MA

Hill J, Foley R, Blazer V, Werner R, & Gannon J (1988) Effects of acidic water on young-of-the-year smallmouth bass (Micropterus dolomieui). Envir. Biol of Fishes. vol. 21, no. 3, 223-229

Kullberg A, Bishop K H, Hargeby A, et al (1993) The ecological significance of dissolved organic carbon in acidified waters. Ambio. 22, 331-337

Leenheer J A (1980) Origin and nature of humic substances in the waters of the Amazon River basin. Acta Amazonica 10, 513-526

Matsuo A Y, Val A L (2002) Low pH and calcium effects on net Na+ and K+ fluxes in two catfish species from the Amazon River (Corydoras: Callichthyidae)

McDonald D G (1983) The interaction of environmental calcium and low pH on the physiology of the rainbow trout, Salmo gairdneri. J. Exp. Biol. 102, 123-140

Moiseenko T I, Sharova, O N (2006) Physiological mechanisms of degradation of fish populations in acidified water bodies. Russian J. of Ecology. vol. 37, No. 4, 257-263

Schindler D W (1988) Effects of acid rain on freshwater ecosystems. Science. vol. 239, no. 4836, 149-157

Tietz N W (1994) Tietz Textbook of Clinical Chemistry. W B Saunders Co. Philadelphia.

Wood C M, Matsuo A Y, Wilson R W, Gonzalez R J, et al (2003) Protection by natural blackwater against disturbances in ion fluxes caused by low pH exposure in freshwater stingrays endemic to the Rio Negro. Physio. Bio. Zoology. vol. 76, no. 1, 12-27

Wilson R W, Wood C M, Gonzalez R J, et al (1999) Ion and acid-base balance in three species of Amazonian fish during gradual acidification of extremely soft water. Physio. Bio. Zoology. Vol. 72, no. 3, 277-285