Sodium uptake and loss rates are given for three gammarids acclimatized to media ranging from fresh water to undiluted sea water.
In Gammarus zaddachi and G. tigrinus the sodium transporting system at the body surface is half-saturated at an external concentration of about 1 mM/l. and fully saturated at about 10 mM/l. sodium. In Marinogammarus finmarchicus the respective concentrations are six to ten times higher.
M. finmarchicus is more permeable to water and salts than G. zaddachi and G. tigrinus. Estimated urine flow rates were equivalent to 6·5 % body weight/hr./ osmole gradient at 10° C. in M. finmarchicus and 2·8 % body weight/hr./osmole gradient in G. zaddachi. The permeability of the body surface to outward diffusion of sodium was four times higher in M. finmarchicus, but sodium losses across the body surface represent at least 50 % of the total losses in both M. finmarchicus and G. zaddachi.
Calculations suggest that G. zaddachi produces urine slightly hypotonic to the blood when acclimatized to the range 20 % down to 2 % sea water. In fresh water the urine sodium concentration is reduced to a very low level.
The process of adaptation to fresh water in gammarid crustaceans is illustrated with reference to a series of species from marine, brackish and freshwater habitats.
Previous papers have described sodium uptake and loss across the body surface and in the urine of the freshwater amphipods Gammarus pulex and G. lacustris, and in G. duebeni from both fresh-water and brackish-water localities (Shaw & Sutcliffe, 1967; Sutcliffe, 1967a, b;Sutcliffe & Shaw, 1967a, b). The present paper deals with sodium uptake and loss in a further three species of gammarids, and a comparison is then made of the main features of sodium regulation in the gammarids examined in this series of papers, representing differing degrees of adaptation to fresh water.
Gammarus tigrinus Sexton is a North American species where it is apparently confined to brackish water (Bousfield, 1958). It was probably introduced into Britain where it is also confined to relatively saline waters in the north and west Midlands, but it occurs in fresh water in Lough Neagh and the River Bann in Northern Ireland (Sexton & Cooper, 1939; Spooner, 1951; Hynes, 1955; Hynes, Macan & Williams, 1960). Gammarus zaddachi Sexton is primarily a brackish-water species occupying the less saline zone at the head of estuaries and, in Britain at least, it is often found in fresh water at the mouths of streams in the zone influenced by salt water during high spring tides (Sexton, 1942; Spooner, 1947; Kinne, 1954; Hartog, 1964). Marino-gammarus finmarchicus (Dahl) is a marine littoral species occurring intertidally along the sea shore and in more saline water at the mouth of estuaries (Sexton & Spooner, 1940).
MATERIAL AND METHODS
Gammarus zaddachi Sexton (Kinne, 1954) (G. zaddachi zaddachi in Spooner, 1947) was obtained from two similar localities, underneath stones on the shore where a small stream discharges over the beach at Budle Bay, Northumberland, and near Barrow-in-Fumess, Lancashire. Care is needed to distinguish this species from Gammarus salinus Sexton (G. zaddachi salinus in Spooner, 1947), which was occasionally found amongst G. zaddachi at Budle Bay. All of the specimens used in the investigation were individually examined under a binocular microscope. For this the specimens were anaesthetized by brief immersion in sea water saturated with CO2. They recovered within a few minutes after placing them in a beaker of sea water with compressed air bubbled through a diffuser, but they were allowed several days to recover before they were used for experimental purposes.
G. tigrinus Sexton was collected from a small tributary of the River Penk at Coven, Staffordshire, a few miles north of Wolverhampton. The water in this and other streams in the area contains a relatively high concentration of salts. Dr H. B. N. Hynes very kindly informed me of this locality, where G. tigrinus was readily available and common in 1962, but less so in 1963. In 1964 only a few specimens were found at this locality and further work was not possible due to the paucity of animals. G. tigrinus was then obtained from fresh water in Lough Neagh, N. Ireland, in March 1965.
Marinogammarus finmarchicus (Dahl) was obtained from underneath stones in the tidal zone of the shore at Cullercoats Bay, Northumberland. Again care is needed to distinguish this species from Marinogammarus obtusatus (Dahl) (see Sexton & Spooner 1940), which was found intermingled with M. finmarchicus. Each specimen was identified under a binocular microscope when anaesthetized with CO2 as described for G. zaddachi.
Experiments on G. zaddachi from Budle Bay, and on G. tigrinus and M. finmarchicus, were carried out in a constant-temperature room at 10± 1° C. Experiments on G. zaddachi from Barrow-in-Furness were carried out in a constant-temperature room at 9 ± 1° C. Animals were acclimatized to these temperatures for about 1 week before commencing experimental work. Between experiments they were fed on dead leaves, and occasionally small pieces of blowfly larvae and worms were also given to G. tigrinus.
Experimental techniques and measurements were carried out in the manner described in the previous papers in this series. Experimental media were NaCl solutions to provide concentrations of less than 8 mM/l. Concentrations greater than this were made by diluting Cullercoats sea water with tap water at Newcastle and with deionized water at Ferry House. For convenience, these sea-water media are usually referred to in terms of NaCl solutions having the same freezing-point depressions.
SODIUM REGULATION IN GAMMARUS ZADDACHI
Survival in dilute media
Animals were transferred from undiluted sea water to 10 mM/l. NaCl (2 % sea water) without harm, and it was possible to keep them alive and healthy for several weeks at this concentration at a temperature of 9–10° C. Survival at lower external concentrations was very variable. For example, most of the animals in one batch from Barrow-in-Furness survived in 0–25 mM/l. NaCl for nearly a week, but practically all of the animals in two other batches died within 3 days at the same concentration. Experiments to determine the minimum external sodium concentration at which sodium balance was maintained also gave rather variable results. In one experiment, after repeated sodium loss into about 30 ml. de-ionized water, six groups of ten animals originally acclimatized to 0·3 mM/l. NaCl maintained balance at a concentration of 0·21 ± 0·05 mM/l. sodium for 24 hr. At the same time the external chloride concentration was 0·18 (6) ± 0·03 mM/l. But 48 hr. later the external sodium concentration had increased to 0·37 ± 0·07 mM/l. In another experiment six groups maintained balance for 24 hr. at a concentration of o·16 ±0·03 mM/l. sodium, and during the next 24 hr. this increased to 0·22 ± 0·05 mM/l.
In general these results indicate that the ability to maintain sodium balance at very low external concentrations closely resembles that of G. duebeni from brackish-water localities (Sutcliffe, 1967a).
Blood sodium concentration
Estimates of the blood sodium concentration are shown in Table 1. Only small specimens were available at the time, so the blood drawn from about six individuals, pooled under liquid paraffin, was used for each measurement. Over the range from about 50 % to 2 % sea water the blood sodium level fell by only 16%, but it was reduced by a further 25 % in animals acclimatized to 0-25 mM/l. NaCl for only 24 hr. These blood concentrations are very similar to the concentrations found in G. duebeni when acclimatized to the same media (Shaw & Sutcliffe, 1961; Sutcliffe, 1967a).
Sodium influx and loss at low external concentrations
Sodium influx was determined over a range of external concentrations from 0·3 to 6 mM/l. NaCl in groups of animals from Budle Bay acclimatized to 10 mM/l. NaCl at 10° C. These animals were then acclimatized to 0·3 mM/l. NaCl and the sodium influx measurements were repeated. The average weight of these animals was 40 mg. The results are shown in Fig. 1. The relation between the influx and the external sodium concentration is very similar to that found in G. duebeni from brackish water. The influx reached a maximum at an external concentration approaching 10 mM/l. sodium and the transporting system was half-saturated at an external concentration of about 1 mM/l. sodium. The maximum influx in animals acclimatized to 0·3 mM/l. NaCl was nearly double the maximum influx when acclimatized to 10 mM/l. NaCl, but a curious feature of this increase is that the influx from 0·3 mM/l. sodium in animals acclimatized to this concentration was not significantly greater than the influx from 0·3 mM/l. when the animals were previously acclimatized to 10 mM/l. NaCl (Table 2). This is different from the situation found in G. duebeni, G. pulex and G. lacustris where the influx from very low external concentrations was significantly increased (Shaw & Sutcliffe, 1961; Sutcliffe, 1967a, b; Sutcliffe & Shaw, 1967a).
The total sodium loss into de-ionized water was determined with the same batch of animals used to determine sodium influxes. The results are also given in Table 2. In animals acclimatized to 10 mM/l. NaCl the sodium loss rate (0·25 µM/hr.) was considerably less than the sodium influx (0·6–0·7 μM/hr., Fig. 1), indicating that approximately 60% of the influx was due to an exchange component. On the other hand, when acclimatized to 0·3 mM/l. NaCl the loss rate just balanced the influx (Table 2). Net sodium uptake in these animals was then determined by placing groups of ten animals in 20 ml. of 2·0 mM/l. NaCl. The decrease in the external sodium concentration was measured at intervals of 30 min. The results are shown in Table 3, where the total sodium uptake rate was deduced by including the loss rate found in animals acclimatized to 0·3 mM/l. NaCl, ie. 0·13 μM/hr. (Table 2). It appears that nearly all of the increased influx in animals acclimatized to 0·3 mM/l. NaCl represents a true uptake of sodium, and only some 10–15 % of the influx was now due to an exchange component. These features of the sodium transporting system in G. zaddachi are also present in G. duebeni and G. pulex.
Another feature also found in G. duebeni and G. pulex is that the sodium loss rate was immediately affected by an increase in temperature. Three groups of ten animals acclimatized to 10 mM/l. NaCl at 10° C. were transferred to de-ionized water at 20° C. and the sodium loss was measured over a period of 80 min. The increase in the external sodium concentration was linear with time, and the mean sodium loss rate from three groups was 0·47 μM/animal/hr. This is approximately double the loss rate into de-ionized water at 10° C. (Table 2).
In the case of G. duebeni from brackish-water localities acclimatized to 10 mM/l. NaCl it was shown that the sodium loss into de-ionized water was shared equally between sodium loss in urine slightly hypotonic to the blood and sodium loss across the body surface. When transferred to 0·25 mM/l. NaCl the sodium loss rate into deionized water was reduced by 60 % and the influx was greatly increased. The reduction in loss rate was partly due to a fall in blood sodium but mainly due to elaboration of a dilute urine containing about 40 mM/l. sodium (Sutcliffe, 1967a). From Table 2 it appears that in G. zaddachi the sodium loss rate into de-ionized water was also halved when animals were moved from 10 to 0·3 mM/l. NaCl, and again this reduction is only partly accounted for by the fall in blood sodium concentration (Table 1). The following section examines the possibility that there is also a change in the urine sodium concentration in G. zaddachi.
Sodium loss in the urine
This was investigated by comparing the sodium loss rates into de-ionized water and into sucrose solutions made slightly hyper-osmotic to the blood of animals acclimatized to a range of dilute sea-water media at 9° C. The technique is discussed by Sutcliffe (1967a,b). Since the blood sodium concentration is almost identical with that of G. duebeni the total concentration of the blood was assumed to be also the same as in G. duebeni. When groups of animals were placed in the experimental sucrose solutions, sodium loss into the sucrose declined quite rapidly during the first 10–20 min., particularly in animals acclimatized to concentrations greater than 10 mM/l. NaCl. The loss rate was then constant and remained linear with time during the next 60–80 min. The initial high loss rate was interpreted as due to the initial production of urine containing sodium, and this urine flow then ceased in the absence of osmotic water uptake. The lower rate of sodium loss in the sucrose solutions is attributed to outward diffusion of sodium across the body surface.
The results of a series of experiments at 9° C. using a batch of animals with an average weight of 48 mg. are given in Fig. 2. The animals were acclimatized to the sea-water media in the following order; 10, 170, 270, 540 and 115 mM/l. NaCl. Sodium loss into de-ionized water was always greater than the loss into sucrose, and whereas the loss rate into de-ionized water increased sharply at external concentrations above 115 mM/l. NaCl the loss rate into sucrose increased more gradually. When acclimatized to 540 mM/l. NaCl (undiluted sea water) the sodium loss rate into sucrose was double the rate when acclimatized to 115 and 10 mM/l. NaCl. This increase in sodium loss across the body surface was presumably due to an increase in the blood sodium concentration, which would also be doubled in animals acclimatized to un-diluted sea water.
From Fig. 2 it is clear that when acclimatized to 170 and 270 mM/l. NaCl the increased rate of sodium loss into de-ionized water must have been mainly due to a greater sodium loss in the urine. This resembles the situation in G. duebem, where the urine was isotonic with the blood of animals acclimatized to 270 mM/l. NaCl and higher external concentrations but was slightly hypotonic to the blood in more dilutemedia (Lockwood, 1961; Sutcliffe, 1967a). Hence it is reasonable to assume that in these experiments the urine of G. zaddachi was also isotonic with the blood when acclimatized to 270 mM/l. NaCl, but that it was hypotonic to the blood when acclimatized to 10 and 115 mM/l. NaCl. Lockwood (1961) notes that G. zaddachi can produce hypotonic urine when in dilute media. Now when acclimatized to 270 mM/l. NaCl the difference between the sodium loss rates into sucrose and de-ionized water was 0 · 23 μM/hr. (Fig. 2). If this loss was due entirely to an isotonic urine containing about 300 mM/l. sodium, the urine flow rate in de-ionized water must have been equivalent to about 38 % body weight/day at 9° C. Similarly, the difference of approximately 0·20 µM/hr. when acclimatized to 170 mM/l. NaCl (from Fig. 2) could be accounted for by a roughly isotonic urine containing 260 mM/l. sodium produced at a rate equivalent to 38 % body weight/day. This rate of urine flow is similar to the estimated flow rate of 28 % body weight/day at 9° C. in G. duebeni (Lockwood, 1961; Sutcliffe, 1967a).
With a urine flow rate equivalent to 38 % body weight/day, sodium loss attributed to the urine when acclimatized to 115 and 10 mM/l. NaCl was 0·12 and 0·11 μM/hr. respectively (Fig. 2), which would require a urine sodium concentration of about 145–160 mM/l. This would be hypotonic to the blood.
To test the possibility that, like G. duebeni, the sodium concentration in the mine of G. zaddachi can be reduced to a very low level, a batch of large animals was acclimatized to 10 mM/l. followed by 0·25 mM/l. NaCl at 9° C. The average weight of these animals was 82 mg. Sodium loss rates into de-ionized water and slightly hyperosmotic sucrose are given in Table 4. The loss rates in these animals are approximately double the comparable loss rates in 48 mg. animals shown in Fig. 2. When acclimatized to 10 mM/l. NaCl, sodium loss attributed to the urine was 0·29 μM/hr. If the urine flow rate in these animals was also equivalent to 38 % body weight/day at 9° C. then the concentration of the urine would have been about 223 mM/l. sodium, i.e. slightly hypotonic to the blood. But after 3–4 days acclimatization in 0·25 mM/l. NaCl the loss rate into de-ionized water was reduced by 50 % (Table 4), as it was in a previous batch of animals acclimatized to 0·3 mM/l. NaCl (Table 2), whereas the loss rate into sucrose was only slightly reduced. Hence most of the large reduction in loss rate into de-ionized water must have been brought about by a fall in the urine sodium concentration. Indeed the results indicate that the urine sodium concentration was close to zero since the difference between the loss rates when acclimatized to 0·25 mM/l. NaCl, shown in Table 4, is not significant (P > 0·05, N = 10, t = 2·14). However, if this difference of 0·03 μM/hr. was actually due to sodium loss in urine produced at the above flow rate then the urine sodium concentration would have been approximately 23 mM/l. In either case it is clear that the urine concentration was reduced to a very low level, and it appears that in general the regulation of urinary sodium losses in G. zaddachi closely resembles the regulation found in G. duebeni. Since the reduction in sodium loss via the urine is linked with an increase in the sodium influx rate (Fig. 1) it seems likely that the overall regulation of sodium uptake in G. zaddachi is controlled by an internal mechanism operating in the manner suggested for G. duebeni (Sutcliffe, 1967a).
Sodium loss across the body surface
The permeability constant K′ for sodium loss across the body surface in the direction in-to-out is calculated in Table 5 for animals with an average weight of 48 mg. The mean value for K′ is 0·00067 and this is very similar to the value of K′ in both G. duebeni and G. pulex (see later in this paper). Using this value for K′ the amount of sodium lost across the body surface can be calculated for animals living in dilute sea-water media, and this can be compared with sodium loss in the urine assuming that the urine flow rate is equivalent to 38 % body weight/day at 9° C. for an osmotic gradient of 300 mM/l. NaCl between the blood and external medium. The results are shown in Fig. 3 for 48 mg. animals fully acclimatized to the range of media where active sodium regulation is necessary to maintain the normal high blood sodium level. Over the range of external concentrations 270–10 mM/l. NaCl (50 % to 2 % sea water) sodium loss across the body surface accounts for at least 50% of the total sodium losses, even if the urine was maintained isotonic with the blood over the entire range of sea-water media. This is yet another aspect of sodium regulation in which G. zaddachi resembles G. duebeni.
SODIUM REGULATION IN GAMMARUS TIGRINUS
Survival in dilute media
The salinity tolerance of G. tigrinus corresponds with the salinity tolerance of G. duebeni and G. zaddachi. Werntz (1963) reported that about 50% survived direct transference into both 150% sea water and fresh water. Some animals from the River Penk and from Lough Neagh survived for several days in 0·15 mM/l. NaCl and, after repeated sodium loss into de-ionized water, sodium balance was maintained at even lower external sodium concentrations for at least 24 hr. (Table 6). But it is unlikely that G. tigrinus could survive for long periods at these very low external concentrations as mortality was rather high after about 2 weeks in 0·25 mM/l. NaCl. Death may have been partly due to prolonged starvation, since it was noted that these animals did not feed, whereas they did feed on pieces of leaves, worms and other materials at higher external concentrations. A few lived for more than 6 months in 2 mM/l. NaCl. The sodium concentration in the tributary of the River Penk was 4 mM/l., but only 0-46 mM/l. in Lough Neagh (Table 7).
Measurements of the total blood concentration in animals from the River Penk acclimatized to sea-water media are shown in Fig. 4. The results are similar to values for the blood concentration obtained by Werntz (1963) with G. tigrinus in North America, indicated by the broken line in Fig. 4. The blood concentration in G. tigrinus is intermediate between that of some brackish-water and freshwater species of Gammarus.
Sodium influx and loss
Figure 5 shows the relation between the external concentration and sodium influx in G. tigrinus from Lough Neagh acclimatized to 0·5 mM/l. NaCl at 10° C. The average weight of the animals was 30 mg. Although the maximum influx was not determined it is clear that the sodium transporting system closely resembles that of G. zaddachi, and also that of G. duebeni from brackish water and fresh water in Britain (Sutcliffe, 1967a), in that the transporting system is less than half-saturated at external concentrations below about 1 mM/l. sodium. The regulation of total sodium loss into de-ionized water also resembles that found in G. duebeni from brackish and fresh water in Britain, as the sodium loss rate in animals from the River Penk (average weight 42 mg.) was progressively reduced in animals acclimatized to the range 10–0·15 mM/l. NaCl (Table 8). Similar results were obtained with G. tigrinus from Lough Neagh (Table 9), and sodium regulation over this range of external concentrations appears to have been achieved almost entirely by reducing the rate of sodium loss, as the sodium influx was not significantly altered when acclimatized to progressively lower concentrations (Table 9). This is very surprising in view of the fact that a marked reduction in the sodium loss rate is closely linked with an increased rate of sodium uptake in G. duebeni, G. zaddachi, G. pulex and G. lacustris, and these particular results with G. tigrinus must be viewed with caution until the matter has been studied in greater detail.
SODIUM REGULATION IN MARINOGAMMARUS FINMARCHICUS
Blood concentration and survival in dilute media
The blood sodium concentration was determined in animals acclimatized for 8 days to undiluted sea water and sea-water media equivalent to 270 and 115 mM/l. NaCl. Some determinations were also made with animals acclimatized to 60 mM/l. NaCl for 3 days. The results are shown in Fig. 6, together with data from Werntz (1963) on the total blood concentration in North American specimens of this species, obtained by measuring the freezing-point depression. The blood sodium determinations agree very well with the freezing-point measurements, and it is clear that although this species is usually confined to the marine littoral zone and the mouth of estuaries (Sexton & Spooner, 1940) the blood concentration is strongly regulated at external concentrations from 450 mM/l. down to about 115 mM/l. NaCl. In fact, over this range of sea-water media the values for blood sodium and total concentration are higher than the corresponding values in G. duebem and G. zaddachi. M. finmarchicus also differs from these brackish-water species in that the blood concentration falls rapidly at external concentrations below 60 mM/l. NaCl, and this is reflected in the poor survival at low external concentrations. Werntz (1963) found that some individuals of M. finmarchicus survived for 48 hr. in 0·03 molal sea water (approximately 16 mM/l. NaCl) at 15° C. but none survived 24 hr. in fresh water. On one occasion, out of seventy M. finmarchicus from Cullercoats kept at 10° C., twenty-five were still alive after 11 days in 10 mM/l. NaCl. But this was exceptional, and in general it was found that a concentration of 115 mM/l. NaCl (about 20% sea water) was required to keep the majority of individuals alive and healthy for a period of 1-2 weeks at a temperature of 10° C.
A few measurements of sodium influxes from low external concentrations were made with animals acclimatized to 10 and 115 mM/l. NaCl. The average weight of these animals was 55 mg. The results are shown in Fig. 7. There appears to be no difference between the influxes of animals acclimatized to the two sea-water media, and the general relationship between the external concentration and sodium influx resembles that found in G. zaddachi, G. tigrinus and G. duebeni from brackish water.
In these three species the influx increases gradually to a maximum at an external concentration of about 10 mM/l. NaCl, but in the case of M. finmarchicus it is probable that the sodium influx only reaches its maximum saturation rate at higher external concentrations. With the method employed here it was not possible to obtain an accurate measurement of sodium influx from external concentrations greater than 8 mM/l. But a rough estimate of the external concentration at which the influx is half-saturated can be made if it is assumed that, as in G. duebeni (Shaw & Sutcliffe, 1961; Sutcliffe, 1967 a), sodium influx is described by the Michaelis equation, influx = K[C/(Km+C)], where K is the maximum rate of transport, C the external concentration, and Km is the external concentration at which half the maximum influx is reached. By extrapolation from the influx curve in Fig. 7 it appears that the maximum influx is not less than 0·6 μM/hr. When K = 0·6, Km is equal to 2·5 and the sodium influx would be 90 % saturated at an external concentration of about 20 mM/l. However, when the animals were acclimatized to 115 mM/l. NaCl the total sodium loss rate from six groups in de-ionized water was 1·29 ± 0·15 μM/animal/hr. This and other observations presented below indicates that in order to maintain sodium balance the influx must be at least 1·0μM/animal/hr. When K = 1·0, is equal to 60 and the influx would be 90 % saturated at an external concentration of 60 mM/l. In fact the influx curve drawn through the measurements presented in Fig. 7 is very close to a theoretical curve derived from the Michaelis equation when K = 1·0 and Km = 6·0.
Sodium loss across the body surface
Sodium loss into de-ionized water and sucrose solutions made slightly hyperosmotic to the blood was measured in groups of animals acclimatized to increasing dilutions of sea water at 10° C. The results are given in Table 10. The sodium loss rate into sucrose when acclimatized to 115 mM/l. NaCl was reduced to 55% of the loss when acclimatized to 540 mM/l. NaCl (undiluted sea water) and this reduction in sodium losses across the body surface is satisfactorily accounted for by the 60 % fall in blood sodium concentration over the same range of external concentrations (Fig. 6). The data can therefore be used to estimate the permeability constant K′ for sodium loss across the body surface (Table 11). The mean value of 0·0028 for animals with an average weight of 55 mg. is four times larger than the value of K′ in G. zaddachi of similar size and weight, which indicates that at least some part of the body surface in M. finmarchicus is more permeable to the outward diffusion of sodium.
Sodium loss in the urine
The higher loss rates into de-ionized water compared with sucrose (Table 10) are believed to be due to additional sodium loss in the urine and, if it is assumed that the urine was isotonic with the blood, the rate of urine flow can be calculated. Thus in animals acclimatized to undiluted sea water and then transferred to de-ionized water, a sodium loss rate of 2·01 μM/hr. (Table 10) in urine with a sodium concentration of 536 mM/l. would require a urine flow rate equivalent to 6–8 % body weight/hr. at 10° C. with a 1 osmolar gradient between the blood and external medium. When acclimatized to 270 mM/l. NaCl and then transferred to de-ionized water, the sodium loss rate of 0·95 μM/hr. in an isotonic urine containing 368 HIM /I. sodium requires a urine flow rate equivalent to 4·7 % body weight/hr., and the loss when acclimatized to 115 mM/l. NaCl indicates a flow rate of 3·7 % body weight/hr. These flow rates are respectively equivalent to 6·6 and 6·o % body weight/hr. for a 1 osmolar gradient between the blood and medium, and so the urine flow rate in M. finmarchicus is very close to the estimated rate in G. oceanicus, which is also a marine littoral species. Here the flow rate is equivalent to 10·5 % body weight/hr. for a 1 osmolar gradient at 15° C. (Werntz, 1963) or approximately 7 % body weight/hr. at 10° C. These flow rates in M. fin-marchicus and G. oceanicus are two to three times faster than the comparable urine flow rate in G. zaddachi.
As a check on these results with M. finmarchicus an estimate was made of the urine flow rate in M. obtusatus, another marine littoral species. With animals acclimatized to undiluted sea water the mean sodium loss rate from six groups in de-ionized water was 2·92 ± o·22 μM/animal/hr., and the mean loss in slightly hyper-osmotic sucrose was 1·14 (6) ± 0·10μM/hr. at 10° C. Measurements of the osmotic pressure of the blood made on this species by Beadle & Cragg (1940) indicate that the blood sodium concentration of animals in sea water will be very similar to that found in M. fin-marchicus. Hence with an average weight of 36 mg. and a sodium loss of 1·78 /m/hr. in urine isotonic with blood containing about 536 mM/l. sodium, the urine flow rate in de-ionized water was apparently equivalent to 9·2 % body weight/hr. for a 1 osmolar gradient. This fast rate agrees well with the estimates made on the other marine species.
The overall regulation of sodium balance in Marinogammarus finmarchicus
With a value of 0·0028 for K’ and an isotonic urine produced at a rate equivalent to 6·5 % body weight/hr. for a 1 osmolar gradient, it is possible to estimate the losses of sodium across the body surface and in the urine of 55 mg. animals living in a range of sea-water media (Fig. 8). For the purposes of calculation the total blood concentration was reckoned to be 20 mM/l. greater than the blood sodium concentration. It is immediately clear from Fig. 8 that the partition of sodium losses is very similar to the situation found in G. zaddachi and G. duebeni. Thus sodium losses by diffusion represent more than 50 % of the total loss rate, even when isotonic urine is produced at a fast rate in dilute media. But compared with the total sodium loss rate in G. zaddachi in dilute sea-water media (Fig. 3) the total loss rate in M. finmarchicus is six to seven times higher. This is due to a combination of three factors: (1) the higher permeability of the body surface to sodium; (2) the higher permeability to water and hence a faster urine flow rate; and (3) the higher blood concentrations in dilute sea-water media increase the diffusion gradients of both sodium and water.
In order to maintain sodium balance at external concentrations of 60-115 mM/l. NaCl the total sodium loss rate of 1·1–1·2 μM/hr. must be met with an equivalent rate of sodium uptake. Now the sodium influx was not increased when animals were moved from 115 to 10 mM/l. NaCl (Fig. 7) and the results indicated that the maximum influx was approximately 1·0 μM/hr. at an external concentration of about 60 mM/l. NaCl. At lower external concentrations the sodium influx declined rapidly and this is illustrated in Fig. 8 by a curve where the maximum influx K′ is made equal to 1·1 and Km, = 10. It is clear that there will be a net sodium loss at external concentrations below about 115–60 mM/l. NaCl and this accounts for the poor survival at concentrations below 115 mM/l. NaCl or approximately 20% sea water.
One other point of interest emerges from Fig. 8. In order to maintain a steady blood sodium concentration at an external concentration of 270 mM/l. NaCl the sodium uptake rate must be increased by a factor of 2 · 5 compared with the uptake rate required to balance losses at 540 mM/l. NaCl, and the rate must then be nearly doubled again at an external concentration of 115 mM/l. NaCl. This five-fold increase in the uptake rate is almost exactly the same as the overall increases in sodium uptake in G. zaddachi (Fig. 3) and in G. duebeni from brackish water (Sutcliffe, 1967a). But there is a marked difference in the ‘timing’ of the increases in uptake rate. In G. zaddachi and G. duebeni these increases apparently come into operation when the blood sodium level is maintained strongly hypertonic to external concentrations of less than about 270 mM/l. NaCl (50 % sea water). In both species the uptake rate reaches its maximum saturation level at an external concentration of only about 10 mM/l. NaCl, and the uptake mechanism continues to be stimulated at concentrations of less than 1 mM/l. In contrast, the uptake rate in M. finmarchicus apparently is immediately increased when normal sea-water is diluted by only a small amount, and this is reflected in the strong hypertonic regulation of the blood at concentrations below about 450 mM/l. NaCl (Werntz, 1963). But the maximum uptake rate is reached at a concentration of about 115 mM/l. NaCl, and further stimulation of the uptake mechanism at lower external concentrations apparently does not occur.
ADAPTATION TO FRESH WATER IN GAMMARID CRUSTACEANS
The marine species M. finmarchicus is much more permeable to water and salts than G. zaddachi and G. tigrinus, which are essentially brackish-water species that occasionally live in fresh water. This point is clearly shown in Table 12, where it is seen that water permeability as measured by the rate of urine flow is very similar in three marine littoral species, and is roughly three times higher than in several other species from brackish and freshwater habitats. In these the permeability to water is also remarkably similar, and so is the permeability to salts as measured by the outward loss of sodium. Again the marine species are distinguished by a higher permeability to salts, with a permeability constant three to four times greater than in the brackish and freshwater species, so it appears that in this group of closely related animals a reduction in salt permeability has been accompanied by a similar reduction in permeability to water. This has also occurred in the series of decapods studied by Shaw (1961a). On the other hand the permeability to outward loss of sodium in the freshwater prawn Palaemonetes antennarius is about three to four times lower than it is in the brackish-marine P. varions, whereas the permeability to water is the same in both species (Potts & Parry, 1964a; Parry & Potts, 1965).
Blood concentrations in the marine gammarids are higher than in brackish-water species, and marine gammarids begin to osmoregulate strongly more or less as soon as the external concentration is reduced below normal sea water. The brackish-water species do not begin to osmoregulate strongly until the external concentration is lowered to about 50% sea water (Beadle & Cragg, 1940;,Werntz, 1963). Despite a high permeability to both water and salts the marine gammarids maintain osmotic and ionic gradients as large as the gradients maintained by brackish and freshwater gammarids. At first sight this is perhaps rather surprising, particularly since the combination of a large urine flow with a high sodium concentration isotonic with the blood leads to a very fast turnover of sodium. The maintenance of a high blood concentration also results in a very fast sodium loss across the body surface, and this in fact exceeds the loss in the urine. Consequently when M. finmarchicus is kept in less than 20 % sea water the total sodium losses reach a maximum rate of more than 20 mM/kg./hr. at 10 ° C., at least five times greater than the maximum loss rates in G. zaddachi and G. duebeni kept in 2 % sea water.
In order to maintain a steady blood sodium concentration in dilute sea-water media the high sodium loss rate must obviously be balanced by a high rate of sodium uptake, and a fairly generous estimate suggests that the maximum influx in M. finmarchicus is about 20 mM/kg./hr. This is sufficient to balance losses at concentrations down to about 20 % sea water but, unless the maximum influx can be maintained over a long period of time, it seems unlikely that M. finmarchicus would be able to withstand prolonged immersion in less than about 50 % sea water.
G. duebeni, G. zaddachi and G. tigrinus often experience low salinities in their normal habitats, and it is interesting to find that the permeability to both water and salts is reduced to the level found in the freshwater species G. pulex and G. lacustris (Table 12). But as the brackish-water gammarids maintain higher blood concentrations, sodium losses across the body surface are greater; and the elaboration of urine isotonic with the blood except at concentrations approaching 2 % sea water or less (Lockwood, 1961; Sutcliffe, 1967 a) also increases the total sodium loss rate, so that this is some two to four times higher than in the freshwater species. On the other hand the maximum total loss rate at low salinities is considerably reduced in comparison with the marine gammarids, and the rate of sodium uptake required to maintain sodium balance is reduced to only 4–7 mM/kg./hr. It should be noted that the influx of labelled sodium reaches its maximum at a higher rate, due to the presence of a large exchange component in the influx.
In both G. duebeni and G. zaddachi sodium balance can be maintained for short periods in the range of salt concentrations found in fresh water, and this is partly achieved by reducing sodium losses in the urine. At the same time the uptake rate is increased and sodium influx now reaches its maximum saturation level at about 20 mM/kg./hr. (Table 12). This is also the maximum saturation rate in M. finmarchicus, and probably also in the other marine gammarids, which suggests that the sodium transporting system in the brackish-water gammarids has been derived from marine relatives in a comparatively simple way, involving changes in the affinity for sodium ions in the transporting system at the body surface. The acquisition of transporting sites in the antennary glands would also be necessary for the production of hypotonic urine. The successive reductions in the value of Km the external concentration at which the transporting system at the body surface is half-saturated, from marine to brackish to freshwater gammarids (Table 12) is remarkably similar to the reductions in Km found in the series of decapods, Carcinus, Eriocheir and Astacus (Shaw, 1961a).
The greatest changes in the adaptation to very low salinities have occurred in G. pulex and G. lacustris. Here the total sodium loss rate is again reduced (Table 12) by lowering the blood concentration to about one-half the concentration found in G. duebeni and G. zaddachi, and by elaborating a dilute urine. Only a very low rate of uptake is now required to balance sodium losses, and the maximum influx has been reduced to approximately one-seventh of the maximum rate in the brackish and marine gammarids. At the same time the transporting system at the body surface has acquired a very high affinity for sodium ions.
Beadle & Cragg (1940) suggested that adaptation of the Crustacea to fresh water has proceeded in two main stages: (1) the maintenance of a high blood concentration by active ion absorption, and (2) the development of renal salt-reabsorption and a lowering of the blood concentration. Shaw (1961a) proposed that stage (1) is in fact the primary process by which the invasion of fresh water is accomplished, and that this primary process involves two interacting factors: (a) a progressive reduction in the permeability of the body surface to salts, and (b) the acquisition of an active uptake mechanism with a high affinity for the ions transported by the mechanism. The way in which these two interacting factors have evolved in the gammarids is illustrated in Fig. 9. Here the maximum rate of sodium influx in M. finmarchicus, Klt is shown as only half-saturated at about 10 mM/l.; it will reach its saturation level and also balance the total sodium loss rate (20 mM/kg./hr.) at an external concentration of approximately 100 mM/l. sodium. In G. duebeni from brackish water the affinity for sodium is increased, and although the saturation rate K2 is only 50 % of K1 the influx compensates the total loss rate L2 at external concentrations down to about 1 mM/l. This lower loss rate is due to the effects of a lower permeability to water and salts combined with the lowered blood concentration. At concentrations below 1 mM/l. the influx is increased to its maximum, and the saturation rate K2 is now equal to Kr in M. finmarchicus. The influx curves K2 and Ks are also good approximations to sodium influx in G. zaddachi and G. tigrinus.
In G. duebeni from fresh water in Ireland the affinity for sodium is increased still, further, and with the saturation rate K4 equal to K2 in brackish-water G. duebeni the loss rate L2 is now balanced by the influx at concentrations down to about 0 ·25 mM/l. sodium. This influx would normally be sufficient to maintain sodium balance in the animals living in fresh water in Ireland as the sodium concentration in these waters is greater than 0 ·25 mM/l. (Sutcliffe, 1967c). At lower external concentrations or in other situations where the blood concentration is lowered—for example, during moulting or following minor injuries—the influx can be increased to its maximum reaching saturation at K5, and it is interesting to note that this saturation rate is still, roughly equal to the maximum saturated rate in M. finmarchicus.
The final step in this series of changes towards complete adaptation to fresh water is seen in G. pulex and G. lacustris. Here a further lowering of the blood concentration has reduced the total loss rate to and the maximum saturated influx is now greatly reduced to Ka. This has been accompanied by the acquisition of an even higher affinity for sodium ions at very low external concentrations, so that the influx is fully saturated at about 1 IHM/I. sodium, i.e. approximately one-hundredth of the concentration required to saturate the influx in M. finmarchicus.
The changes in affinity for sodium ions and the lowered permeability in G. duebeni, G. zaddachi and G. tigrinus are also accompanied by the factors representing stage 2 of Beadle & Cragg. In all three species the blood concentration has been considerably lowered, and all are able to elaborate urine hypotonic to the blood. In fact the evidence obtained so far indicates that at the very low external concentrations found in fresh water these species can elaborate urine as dilute as that produced by the freshwater species G. pulex and G. lacustris. The production of a hypotonic urine must obviously help to reduce the total salt loss, but the exact nature of the role of renal salt-reabsorption in osmoregulation in animals living in dilute media is not clear (Potts, 1954; Shaw, 1961; Lockwood, 1965; Potts & Parry, 1964b). It may be that its role varies in different groups of animals, since the development of renal salt-reabsorption must have arisen independently on many occasions even within a group like the Crustacea. The fact that the freshwater crabs Eriocheir (Scholles, 1933) and Potamon niloticus (Shaw, 1959) produce urine more or less isotonic with the blood shows that the elaboration of dilute urine is not necessarily an essential part of the process of adaptation to fresh water. In the case of the gammarids reabsorption of sodium in the antennary glands of G. pulex has reduced the uptake rate at the body surface to one-half the rate which would be required to balance losses if the urine was isotonic with the blood (Sutcliffe, 19676). The same saving in sodium uptake at the body surface is also gained by G. duebeni, G. zaddachi and G. tigrinus when producing a very dilute urine. Since the rate of renal uptake is closely linked with alterations in the rate of uptake at the body surface this also provides a very delicate means of controlling changes in the blood concentration. Lockwood (1964) has shown that G. duebeni can be induced to elaborate a hypotonic urine, with an associated increase in the influx at the body surface, when the blood concentration is abruptly lowered.
As a final comment the reduction in total sodium losses in the brackish- and fresh-water gammarids appears to be a feature common to all of the freshwater crustaceans in which the salt loss has been determined. This is shown in Table 13, where it is seen that the total loss rates are much lower than the comparable rates in marine relatives. For example, the loss rate in Palaemonetes antennarius is approximately four times less than in P. varions, and in Potamon and Astacus the loss rate is at least one order of magnitude lower than in Carcinus and Pachygrapsus. Although these loss rates include variable amounts lost in the urine, comparable loss rates in the gammarids (Table 12) indicate that in the freshwater species G. pulex and G. lacustris the permeability of the body surface has been lowered to the same extent as in other freshwater crustaceans.
It is a great pleasure to acknowledge the many stimulating discussions which I have had with Prof. J. Shaw, and I thank him for comments on the manuscript of this paper. The work done at Newcastle and travelling expenses were generously supported by a grant from D.S.I.R.