Responses to Hypersaline Exposure in the Euryhaline Crayfish Pacifastacus Leniusculus: I. The Interaction Between Ionic and Acid-Base Regulation

Variation in ambient salinity exerts profound effects on iono- and osmoregulation, acid-base balance, respiratory gas exchange and intermediary metabolism in decapod crustaceans. Previous investigations have focused largely on the effects of external dilution in marine species. A. C. Taylor (1977) and E. W. Taylor, Butler & Al-Wassia (1977) both examined the ventilatory and cardiovascular responses of Carcinus maenas to dilute environments, and the effects of hyposaline exposure on O₂ transport for this species have been assessed by Truchot (1973) and for Callinectes sapidus by Weiland & Mangum (1975). Changes in O₂ uptake were related to amino acid metabolism by Gilles ( 1973) and the overall changes in intermediary metabolism during salinity stress have been extensively reviewed by Huggins & Munday (1968) and Schoffeniels & Gilles (1970).

The present study is predominantly concerned with the interaction between ionic and acid-base regulation. Hydrogen ion concentration is determined by three independent quantities, the difference between strong acid anions and base cations (S.I.D.), the CO₂ partial pressure and the total weak acid present (Stewart, 1978). The S.I.D. may be adjusted by both branchial and urinary ion exchange mechanisms. may be directly affected by alterations in ventilatory/circulatory flows and additionally by changes in the chemical equilibria arising indirectly from ionic disturbance. Finally, total haemolymph protein levels may fluctuate with external salinity affecting haemolymph buffering capacity. An increase in haemolymph pH during exposure to hyposaline environments was reported in Carcinus (Truchot, 1973, 1981) and in Callinectes (Weiland & Mangum, 1975). This alkalosis was due to excess base related to modifications in ion exchange rather than respiration, since was maintained.

In the present study, the contrasting problem of hypersaline exposure in a euryhaline freshwater species was investigated. Successful invasion of fresh water demands active uptake of ions, production where possible of a hyposmotic urine, and reduction in boundary permeability in order to maintain a hyperosmotic condition of the blood (Kirschner, 1979). As the external concentration is increased, equilibration with haemolymph is attained, whereafter the nature of the response may alter to ensure hyporegulation of body fluids. By comparison, the problem of hyposaline exposure is essentially unidirectional.

The species selected for study was the western North American crayfish Pacifastacus leniusculus, which extends into brackish water up to 13‰ along regions of the Pacific coast in Oregon and Washington (Miller, 1960). The ability of this species to regulate over a wide range of salinities was recently demonstrated by Kerley & Pritchard (1967) and contrasts with the lower salinity tolerance exhibited by most freshwater species (cf. Hermann, 1931; Lienemann, 1938; Bryan, 1960).

Changes in haemolymph acid-base status resulting from exposure to a range of increasing ambient salinities were examined and correlated with ion- and osmoregulatory capacity. In the course of the investigations, the constants and were derived experimentally under controlled in vitro conditions and used for accurate calculation of via the Henderson-Hasselbalch equatio.n. Values have been published for the brachyuran crab C. maenas (Truchot, 1976) but problems inherent in their application to other species have recently been presented (Wilkes, deFur & McMahon, 1980; deFur, Wilkes & McMahon, 1980).

Experiments were performed on adult Pacifastacus leniusculus leniusculus (Dana, 1852), identified from taxonomic criteria of Miller (1960), of either sex and wet weight 57·3–106·8 g, which were obtained from commercial sources (Monterey Bay Hydroculture, California) and judged to be in intermoult stage C (Drach & Tchernigovtzeff, 1967). Prior to experimentation they were housed for at least two weeks in holding tanks 120×62×10cm deep continuously provided with flowing well-aerated, lechlorinated Calgary tap water (ca. 750 ml min⁻¹) maintained at the acclimation temperature of 15 ± 1 ·C, and fed twice weekly on chopped smelt.

Crayfish were exposed to experimental salinities equivalent to 25, 50 and 75% sea water, which were obtained by diluting artificial sea water (Instant Ocean - s.g. 1·020–1·023 at 24 °C) with tap water (see Table 1). As a change in salinity will alter the acidbase status of the medium and thereby the extracellular balance (Dejours, Armand & Gendner, 1978; Truchot, Toulmond & Dejours, 1980), ambient pH and titratable alkalinity were also continually monitored, the latter by HCl titration following the procedure of Strickland & Parsons, 1965 (see Truchot et al. 1980).

Crayfish were housed individually in chambers of dimensions 21 × 14 × 7 cm deep (volume 2 l) supplied at a continuous rate of 40 ml min⁻¹ with recirculated well-aerated water (O₂ tension ≃ torr-elevation 1048 m), maintained at the experimental temperature. The salinity was raised rapidly and without animal disturbance by switching to an alternative circulation.

Haemolymph was anaerobically sampled from submerged crayfish in fresh water or acclimated for 48 h to each of the experimental salinities, analysed immediately for osmolality, pH and total CO₂, and subsamples frozen rapidly for ion analysis. Individual animals were thus sampled once only in view of the reported detrimental effect on acid-base balance of repetitive sampling (Truchot, 1975; McMahon, Butler & Taylor, 1978).

Pre-(venous, v) and post-(arterial, a) branchial samples were removed respectively from the ventral sinus via the arthrodial membrane at the base of the right 4th walking leg and from the pericardial sinus through a small hole ( < 1 mm) drilled through the carapace above the pericardium and lateral to the heart, which had been prepared and sealed with a latex diaphragm one week prior to sampling (see McDonald, 1977).

Osmolality was determined on a 7 μl subsample using a vapour pressure osmometer (Wescor-5100 B). Inorganic ions were routinely analysed on frozen subsamples, preliminary comparison with test measurements on fresh haemolymph having indicated that changes had not resulted from freezing. Cations were analysed by atomic absorption spectrophotometry (Jarrell-Ash 850) using singleelement hollow cathode lamps (Corning). Mg²⁺ and Ca²⁺ were diluted 1 in 500 with 0·1 % LaCl₃.6H₂O, and K⁺ and Na⁺ were diluted 1 in 500 and 4000 respectively in 0·1 % CsCl, to suppress chemical interference occurring in an air/acetylene flame and calibrated against similarly diluted certified stock solution (Fisher). Cl⁻ concentration was determined on a 20 ^μl subsample using a digital chloridometer by coulometric titration with silver ions (Searle Buehler 4–2500).

pH was determined on a 50 μl subsample using a liquid junction capillary electrode (Radiometer, G299 A) connected to an acid-base analyser (Radiometer, PHM71) and calibrated against precision buffers. Total CO₂ (⁠⁠) was measured on a 40 μl subsample either using the Cameron technique (1971) or a Corning 960 CO₂ analyser. Results using both methods on the same sample agreed to ± 2%. CO₂ tension was calculated using values for solubility coefficient of CO₂ (⁠⁠) and pH-dependent first apparent-dissociation constant of carbonic acid experimentally determined in vitro.

Measured changes in haemolymph ionic strength will affect CO₂ solubility and equilibrium constants as outlined in Albers (1970). The use of correct values in the calculation of is imperative, since an error of 0·01 unit in for example can result in an error of 2·3 % (Severinghaus, Stupfel & Bradley, 1956). The values derived for the shore crab by Truchot (1976) cannot be employed for freshwater decapods whose haemolymph ionic concentrations and ion ratios are markedly different, since they lack correction for pH dependence (see Albers & Pleschka, 1967). Furthermore, Truchot’s salinity-dependent nomograms, which were calculated for varying external [Na⁺] by assuming a constant ratio between values in sea water and serum, did not take into consideration variations in haemolymph buffering capacity which the present investigation demonstrates to be an important response to salinity acclimation. Whilst a new operational approach has proved useful for calculation of in freshwater crustaceans (Wilkes et al. 1980; Morgan & McMahon, 1982) the method is limited to a prescribed set of experimental conditions, and in the present study where four experimental media are involved it was considered necessary to derive these constants experimentally.

was determined in haemolymph from crayfish in fresh water or acclimated for 48 h in each experimental salinity by the method outlined in Truchot (1976). Briefly, solubility was determined in a salt-free solution in which precipitated proteins are redissoluble below the isoelectric point during the necessary acidification of the blood (Redfield, Mason & Ingalls, 1932). This indirect method was based on the assumption that solubility in whole haemolymph can be considered as that of a solvent modified by the presence of protein.

Serum was dialysed for 36 h against four changes of distilled water and two changes of 1·2 × 10⁻³N-HCl at 4 °C. Wheatly & McMahon (1982b) demonstrated that the deterioration of haemocyanin over this time period was minimal. Protein concentration was assayed according to the dye-binding method of Bradford (1976) using Coomassie Brilliant Blue G-250, and was determined at 15 °C after 40 min equilibration with 40% CO₂ (⁠ ca-300 torr) in gently shaken round-bottomed tonometers. was similarly determined in acidified distilled water and saline solutions corresponding to the ionic levels found in the haemolymph after equilibration to the experimental salinities, with the intention of applying to these a protein-concentration-dependent correction.

Since in crustaceans [CO₃²⁻] can, depending on pH, form an appreciable part of measured and yet cannot be determined independently of [HCO₃^-], was estimated from Lyman (1956), for a sea water of corresponding ionic strength, corrected for pH dependence ( – 0·102 per pH unit according to Albers & Pleschka, 1967) and used to assess the contribution of CO₃²⁻ to using the second dissociation equation. was then calculated at a range of pH values, by substitution of corrected [HCO₃⁻], measured pH, equilibration and experimentally determined into the first dissociation equation.

values calculated using these constants and others existing in the literature were compared.

Values throughout are expressed as mean ± 1 S.E.M. and significance at the 5% level as calculated by Student’s t-test (unpaired) is denoted.

No mortality resulted from exposure of P. leniusculus to any of the salinities for the duration of the present experiment.

Osmolality and concentrations of inorganic ions in postbranchial haemolymph are given for each experimental salinity in Table 1. Afferent and efferent values were not significantly different.

Except for K⁺, which remained constant throughout, all major inorganic ions were significantly elevated in 50% SW and osmolality likewise in 75 % SW. These increases however were modest in comparison with ambient changes. Fig. 1 indicates the difference between blood and medium ion levels and osmolality. In fresh water, Pacifastacus maintains the ECF hyperosmotic and hyperionic with respect to ambient. With the exception of Ca²⁺, a similar trend was observed in all ions and osmolality during hypersaline exposure. Essentially, the magnitude of the blood-medium difference became progressively reduced and increasingly negative, implying a change to hypoionic regulation. Regression analysis determined that the transition point occurred at 427 mOsm kg⁻¹ external osmolality (approximately 44% SW). Isoionicity with respect to Cl⁻ was attained at this salinity (corresponding to an external [Cl⁻] of 204 ·5 m-equiv l⁻¹). The medium did not become isoionic with respect to either Na⁺ or K⁺ until 51% SW (external concentrations of 254 ·6 and 4 ·6 m-equiv l ⁻¹ respectively). [Mg²⁺] in the haemolymph in fresh water is maintained only marginally hyperionic, and equilibration with the external milieu occurred at 5 ·08 m-equiv l ₋¹ external concentration and thereafter was maintained increasingly hypoionic.

Ca²⁺ regulation departed from this overall pattern. Levels in local tap water and 25 % SW were essentially similar and in both cases haemolymph concentrations were 6 –8 m-equiv higher. This differential was maintained during hypersaline exposure.

No significant change in pH followed exposure to either 25 or 50% SW (Table 2), although a slight (non-significant) alkalosis was observed in mean pH in 25 % SW. A significant acidosis, greater in pre- (0 ·147) than postbranchial haemolymph (0 ·122 pH units) resulted from 48 h exposure to 75 % SW. A progressive reduction in accompanied these changes and became significant around 50 –75 % SW.

⁠. Protein concentration, which was 63 ·59 ± 6 ·17 (15) mg ml ⁻¹ in fresh water decreased significantly to 37 ·30 ±5 ·44 (5), 44 ·62 ± 2 ·17 (16) and 46 ·75 ±2 ·71 (15) mg ml ⁻¹ respectively in 25, 50 and 75% SW. Circulating levels however did not influence CO₂ solubility since the values measured in salt-free acidified haemolymph protein solutions in no case differed from the value of 0 ·0576 mmol l ⁻¹ torr ⁻¹ measured in acidified distilled water. Solubility in acidified salines equivalent in ionic strength to haemolymph after acclimation to the experimental salinities (from Table 1) was therefore assumed to be representative of that in whole serum. Mean values of 0 ·0563 ± 0 ·0002 (5), 0 ·0561 ±0 ·0002 (5), 0 ·0554 ±0 ·0003 (5) and 0 ·0544 ± 0 ·0005 (5) mmol l ⁻¹ torr ₋¹ were measured in FW, 25, 50 and 75 % SW respectively, indicating a progressive decrease in CO₂ solubility with acclimation salinity.

CO₂ buffer curves combined from a number of individual animals in each acclimation salinity (Fig. 2) demonstrated that, accompanying the progressive reduction in circulating [HCO₃⁻ +CO₃^{2 −}] during hypersaline exposure, there was also a reduction in the CO₂ buffer value (expressed as Δ [HCO₃⁻+ CO₃^{2 −}]/ ΔpH) which declined from – 11 ·6 in FW to –8 ·56, –4 ·73 and –7 ·02 in 25, 50 and 75% SW respectively. Equations describing each of the curved CO₂ isopleths were obtained by linearizing the plot of log vs. log pH for all the animals in each acclimation salinity using the operational procedure devised by Wilkes et al. (1980). The points of intersection of these with the buffer line were employed in the calculation of over the given pH range. Table 3 indicates the values for which were used in the estimation of the ratio [CO₃^{2 −}]/[HCO₃⁻]. Having deducted molecular CO₂ (i.e. × -using experimentally determined and equilibration ⁠) from measured [HCO₃⁻], and thereby could be estimated from the determined ratio.

The regression analyses for pH dependence of in fresh water, 25 and 50% SW (see legend to Fig. 3) show reasonable closeness of fit, and there is good correlation between haemolymph salinity ([Cl ⁻] × 1 ·807) and both Y intercept and gradient as follows.

Operational decreased as a function of pH (see Fig. 3), and the pH dependence increased with ionic strength of the haemolymph. This contention is supported by comparison of the FW and 25 % SW lines, where gradients were similar ( – 0·138 and – 0·152 per pH unit respectively) yet ion concentrations were unchanged (Table 1). The difference in elevation of these two lines indicates that other changes in blood composition may affect the absolute values of At physiological values of pH, operational became progressively reduced with increasing acclimation salinity.

Experimentally determined values for corrected for pH and ionic strength, were then used in conjunction with estimated and measured values of pH, and to solve the Henderson-Hasselbalch equations for [HCO3⁻], [CO₃²⁻] and ultimately (Table 4). Calculated values are also given using values for sea waters of comparable ionic strength taken from Lyman (1956) and also from Truchot (1976) at a salinity corresponding to ion levels found in Pacifastacus, assuming the relationship between [Na⁺] and external concentration for Carcinus in Truchot (1973). This analysis attempted to highlight sources of potential inconsistencies in the calculation of CO₂ tension. From in vivo measurements of pH and it is evident that during hypersaline exposure progressively decreased ; [HCO₃⁻] and [CO₃²⁻] were also reduced.

In the discussion which follows in vitro determinations have been critically appraised independently of in vivo measurements which resulted from hypersaline exposure.

The mean value of measured in acidified distilled water (0·0576 mmol l⁻¹ torr⁻¹) agrees with values reported by Truchot (1976) and Bartels & Wrbitzky (1960) at 15 °C. All, for some reason, yield lower values than 0·0593 mmol l⁻¹ torr⁻¹ measured gravimetrically by Murray & Riley (1971), although Van Slyke et al. (1928) postulate that the depression of caused by acidification was accountable for only 0·01 % of the reduction.

This investigation confirmed that circulating protein levels do not affect CO₂ solubility, which is effectively a function of ionic strength of the haemolymph. The values determined agree well with predictions from Bohr’s original data derived from Buch (1951) and corrected in accordance with Lyman (1956) for solutions of NaCl, and also with the more recent data of Murray & Riley (1971) for solutions of comparable salinity.

Agreement with Truchot’s data (1976) for haemodilution in Carcinus is not so good. When exposed to dilute media, Carcinus exhibits some reduction in [Na⁺], but never to the level recorded in Pacifastacus in fresh water. For this reason, in order to use the salinity nomograms constructed by Truchot, considerable extrapolation must be employed. The values of 0·0543, 0·0542, 0·0539 and 0·0537 mmol l⁻¹ torr⁻¹, which correspond to FW and 25, 50 and 75% SW respectively, are 4% lower than the measured values.

One potential source of error in the determination of was the estimation of which was unavoidable in the absence of a method for separately determining [HCO₃⁻] and [CO₃²⁻]. In particular, in view of the data obtained for (i.e. Fig. 3), the pH dependence which was taken from work on mammalian blood (Albers & Pleschka, 1967) may have been underestimated, especially at increased ionic strengths. None the less, the ratio of [CO₃²⁻]/[HCO₃⁻] is never greater than 3 %, and the resulting errors may be insignificant.

In all experimental salinities decreased with haemolymph pH. The pH dependence in FW and 25 % SW was similar to that observed in mammalian blood, but appeared to increase with the ionic strength of the haemolymph, although there is little in the literature with which to compare this finding. In addition, factors other than ionic strength may well affect the absolute levels of calculated.

Empirical values for are consistently lower than those predicted by Lyman (1956) for sea water solutions of comparable salinity. The determination of from Truchot’s nomograms meets with similar difficulties in interpretation as were experienced for and yields values generally higher than those experimentally determined, though exhibiting less variation with acclimation salinity. The extent of the error inherent in their use in the determination of CO₂ tension is demonstrated in Table 4. Values for from Lyman (1956) vary from 6·242 (FW) down to 6·173 (75% SW) and result in calculated s which are 4·4, 31·4, 46·7 and 70·6% higher than those calculated in the present investigation in fresh 75% SW, the discrepancy becoming exacerbated at low Truchot’s (1976) values for show still less variation as a function of ionic strength. The derived in FW results in an 18·4% underestimation of those in 50 and 75% SW in 17·3 and 46·1 % overestimations respectively, while the 25 % SW value provides an accurate estimation.

Direct measurement of is difficult owing to the low levels of character-, istic of waterbreathers (Rahn, 1966), and is complicated by rapid clotting in this species. A technique for direct measurement of has recently been described (deFur et al. 1980) and is to be used to confirm calculated values on haemolymph from high salinity acclimated animals which exhibits reduced tendency for agglutination (Troup et al. 1961).

In vivo. Although chronic tolerance of hypersaline exposure has not been tested in the present study, our animals survived in 75 % SW for 48 h. Kerley & Pritchard (1967) have reported an LD50 of 8 days in 90% SW. In contrast, although Potamobius astacus survived 6 weeks exposure to 50 % SW, it rapidly succumbed at higher levels (Hermann, 1931), and Kentall & Schwarz (1964) report LD 50 values of 72–96 h for Orconectes virilis and Cambarus bartonii in 50 % SW. The greater salinity tolerance of P. leniusculus is associated with an ability for hypo/hyperosmotic regulation, which is found in some other crustaceans (e.g. Pachygrapsus crassipes -Prosse, Green & Chow, 1955), but is not unexpected since this species extends into coastal river outlets (Smith, 1959), although rarely in salinities above sea water (Miller, 1960).

The measured levels of haemolymph osmolality and major inorganic ions show good agreement with values for other freshwater crayfish (see Taylor & Wheatly, 1982; Morgan & McMahon, 1982; Wilkes & McMahon, 1982). The suspected mechanisms of hyper-regulation should perhaps be clarified before considering how they may be adapted for regulation in iso-or hyperosmotic media. The crayfish shows two characteristics diagnostic of complete adaptation to fresh water. First, a reduction in ECF concentration minimizes the gradient for H₂O and NaCl diffusion (although water permeability is high) and the production of a dilute urine (implicating renal filtrate reabsorption) reduces salt loss. NaCl is actively transported across the branchial epithelium by ion exchange mechanisms, hydration of metabolic CO₂ producing the counterions (H⁺ and HCO₃⁻) at a rate commensurate with normal ion transfer (Kirschner, 1979).

The increases in osmolality and inorganic ion levels during hypersaline exposure were significant but modest when compared to ambient changes, indicating a switch to hyporegulation. Hyperosmotic regulators generally become isosmotic as the external concentration is raised (see Lienemann, 1938; Bryan, 1960). Facing now the reverse problem of desiccation and salt loading, these animals must drink the medium and actively extrude NaCl. Kerley & Pritchard (1967) found that Pacifastacus does not extensively drink the medium even in hypersaline conditions. Cells structurally analogous to chloride cells of fish gills have been observed in crustaceans by Copeland (1967) and may account for Cl⁻ extrusion. Na⁺ extrusion is more controversial. Na⁺/K⁺ ATPase may be important, as may reversal of the Na⁺/NH₄⁺ exchange, which operates in low salinity (Henry & Mangum, 1980). Generally, renal salt extrusion is less important, and the best these animals can do is to produce an isosmotic urine (Potts & Parry, 1964). Kerley and Pritchard’s study, which examined intracellular osmotic regulation during hypersalinity, concluded that osmotic regulation in Pacifastacus does not involve shifts in water between body fluid compartments. The relative contribution of these mechanisms, however, awaits a critical analysis of renal and branchial function in saline-stressed animals, perhaps incorporating radioisotope flux studies.

More pertinent to the present investigation is the integration of ionic/osmotic and acid-base regulation. It has been established relatively recently that the principal mechanism by which pH and bicarbonate are maintained in waterbreathing animals is not the ventilatory control of but adjustment of the strong ion difference (Cameron, 1976, 1978, 1980), which can be adjusted by changes in the relative rates of Cl⁻/HCO₃⁻ and Na⁺/H⁺ exchange at the gills (and kidney). Whilst rates of exchange were not measured in the present investigation, certain conclusions can be drawn from the data.

The changes in acid-base balance resulting from hypersaline exposure were complex, and varied with the degree of salinity imposed. pH was maintained in 25 and 50% SW, an acidosis developed in 75% SW. The response in 75% SW directly contrasts with a haemolymph alkalosis reported by Truchot (1973), Taylor (1977) and Weiland & Mangum (1975) for marine crabs undergoing hyposmotic stress. The accompanying increase in circulating bicarbonate in these studies contrasts with the progressive reduction observed in the present study along with a pioneer investigation on eel migration from fresh to salt water (Fontaine & Boucher-Firly, 1933 a, b), where a similar reduction occurred. The magnitude of the [HCO₃⁻ + CO₃²⁻] change observed in Carcinus on transfer from 36 to 12‰ is comparable with that found in the present study, although the pH increase was more extreme (0·4 pH unit - Truchot, 1973). Weiland & Mangum’s data on Callinectes (1975) are more conservative and more closely parallel the findings in Pacifastacus.

In contrast to the present study, where a progressive reduction in was calculated during hypersaline exposure, Truchot (1981) noted that after long-term exposure to external dilution (4 days) Carcinus haemolymph was unchanged, from which he concluded that the observed pH variation was largely independent of ventilation and resulted from a base excess due to altered ionoregulation. Closer analysis of his data with time reveal that there is some contribution of CO₂ to the overall change in acid-base status. In order to discuss his findings in the context of our own, a ‘Davenport’ analysis of the relative contributions of CO₂ and acids/bases to the overall pH change in postbranchial haemolymph is illustrated in Fig. 5. Because of the effect of ionic strength on and ⁠, and thereby the location of isopleths, the classical analyses performed by Woodbury (1965), Davenport (1974) and more recently Wood, McMahon & McDonald (1977) cannot be employed. On Fig. 5, which indicates the mean acid-base status after 48 h acclimation in 25, 50 and 75 % SW (pts. B, C, D respectively) in comparison with FW(A), CO₂ isopleths for 1–7 torr have been plotted from the constants experimentally determined above. It is encouraging to observe that these agree well with isopleths constructed from the original data using the operational approach of Wilkes et al. (1980). An additional complication was the progressive reduction in buffering capacity indicated by a decrease in the slope of the CO₂ buffer line. This may be partially ascribed to the lowered circulating protein levels and other non-bicarbonate buffers, such as NH₄⁺, whose involvement will be implicated below (see Wheatly & McMahon, 1982a).

Respiratory disturbances are typically characterized by translation along a buffer line in the direction of increasing or decreasing ⁠, whereas metabolic disturbances are represented by translation along a isopleth (i.e. increasing or decreasing |HCO₃⁻] at constant - see Wood et al.1977 ). More frequently, as seen in the present example, the overall pH change represents a combination of both. On transfer rom FW to each acclimation salinity, an acidosis of metabolic origin is seen when comparing points of equal on the two CO₂ buffer lines (i.e. A-X), in addition to a ‘respiratory alkalosis’ which from now on, and for reasons outlined below, will be referred to as a ‘variably compensated hypocapnic alkalosis’ (i.e. X-B, C, D).

These metabolic and respiratory origins of the overall alteration in acid-base balance can be quantitatively assessed. In 25 % SW the magnitude of the hypocapnic alkalosis (0 ·20 pH unit), which accounts for of the overall reduction in bicarbonate, exceeds the metabolic acidosis (0 ·14 pH unit) and causes a slight haemolymph alkalosis. On transfer to 50% SW, (which incidentally is near to the point of equilibration of the haemolymph with the external medium) the two exhibit similar magnitudes of 0 ·23 pH unit, resulting in complete compensation and a maintenance of pH. The bicarbonate lost in excess to CO₂ buffering (1-3 m-equiv I^-1) is 2 ·5 m-equiv l⁻¹. Lastly, the acidosis observed on acclimation to 75 % SW constitutes a metabolic acidosis of 0 ·34 pH unit counteracted by a hypocapnic alkalosis of 0 ·22 pH unit: CO₂ buffering in this case accounts for 1 ·8 m-equiv l⁻¹ of the overall reduction (5 ·6 m-equiv l⁻¹). In summary therefore, whilst the magnitude of the hypocapnic alkalosis remains relatively constant, that of the metabolic acidosis increases with hypersaline exposure, accounting for proportionally more of the observed reduction in circulating bicarbonate.

The fact that these observed reductions in haemolymph are not due to changes in ventilation, and therefore not of ‘respiratory’ origin in the accepted sense, is substantiated by recent ventilation studies in our laboratory (see Wheatly & McMahon, 1982a), which demonstrate that significant hyperventilation only occurs in 75% SW. In both 25 and 50% SW, gill ventilation volume and rate are markedly reduced in comparison with FW. We postulate that this part of the acid-base disturbance depends essentially on changes in the acid-base status of the water (i.e. on changes in titration alkalinity, CO₂ capacitance coefficient, and ultimately on ionic movements through the gill epithelium), which explains the preferred use of the terminology ‘hypocapnic alkalosis’.

An inevitable corollary of hypersaline exposure as seen in Table 1 is an increase in the water titration alkalinity (TA), which will determine the shape and position of the CO₂ dissociation curve so that the water capacitance coefficient for CO₂ (i.e. -Piiper, Dejours, Haab & Rahn, 1971) increases. Blood will depend not only on production and excretion rates of CO₂ but also on ⁠. Assuming that the rate of CO₂ output by the gill does not alter as a result of environmental salinity (Taylor (1977) found this to be true for O₂), then inspired and expired CO₂ tensions will be lower (since ), as will haemolymph tensions (see Dejours et al. 1978; Dejours & Armand, 1980; Truchot et al. 1980).

A closer look at Truchot’s data for Carcinus (1981) reveals that prior to complete adjustment of extracellular osmotic concentration an uncompensated hypercapnic acidosis was measured, as one would predict from changes in the acid-base status of the water arising from external dilution. In this case, the change in haemolymph was only transient and returned to normal when base output by the gill increased sufficiently to increase once again the TA of the water passing over the gill.

A careful examination of acid-base and other ionic fluxes across the gill will elucidate whether the acid-base changes occurring during hypersaline exposure in Pacifastacus are linked to extracellular anisosmotic regulation or to metabolic adjustments involved in intracellular isosmotic regulation. In both Callinectes (Weiland & Mangum, 1975) and Carcinus (Truchot, 1981), the pattern of pH change with time closely resembled that of blood osmolality and chloride, in the former case actually extending as far as seasonal differences, in that winter animals have more concentrated blood of a lower pH. Since the haemolymph bicarbonate concentration was found to increase under hypo-osmotic exposure in Carcinus, the involvement of branchial ion exchanges as the source of the acid-base disequilibria would implicate an apparent acid excretion into the water. Measurement of acid-base exchanges between the crab and the external medium indicating net base excretion, however, strongly support a tissular origin of the metabolic disturbance associated with environmental salinity (Truchot, 1981).

It is well known that alteration of the pool of non-essential free amino acids in crustaceans participates in intracellular isosmotic regulation during salinity acclimation (Gilles, 1979). Mangum et al. (1976) suggested that the elevation of blood pH in dilute media resulted from protonation of NH₃ produced in the tissues from enhanced amino acid catabolism by deamination (Jeuniaux & Florkin, 1961) and diffusion into the haemolymph, there accounting perhaps for the measured base efflux.

In a separate investigation, a reduction in circulating [NH₄⁺] during hypersaline exposure in Pacifastacus was coupled with an elevation in free amino acids (see Wheatly & McMahon, 1982a) and may contribute to the observed reduction in S.I.D., since it can be substituted as a counterion for Na⁺ exchange (Towle, 1974). The only problem in the present investigation is the maintained hypocapnic alkalosis. This would preclude the notion of significant acid output by the gill during hypersaline exposure, which can be predicted from the findings of Truchot (1981) and should return the to resting levels. Obviously this aspect of the physiological response to hypersalinity merits further investigation.

The authors gratefully acknowledge financial support provided by Postdoctoral Research Fellowships awarded by the University of Calgary and Alberta Heritage Foundation for Medical Research to M.G.W. and NSERC grant A 5762 to B.R.M.