The Inorganic Composition of the Body Fluids of Three Marine Invertebrates

That the boundary membranes of marine invertebrates are permeable to the ions of sea water appears to have been first clearly demonstrated by Quinton (1900), but the fact has become established only in recent years. The composition of marine invertebrate body fluids, in general, differs but slightly from sea water, and direct chemical analyses have shown that alterations in the ionic composition of the external medium—sea water—cause corresponding alterations in the composition of the internal medium. Experiments of this type have been done on various crabs, Carcinus, Cancer, Eriocheir, and Heloecius, by Quinton (1900), Bethe (1928, 1934), Berger (1931), Berger & Bethe (1931), Dakin & Edmonds (1931), Nagel (1934) and Robertson (1937), on the molluscs Onchidium and Aplysia by Dakin & Edmonds, and Bethe, and on the echinoderms Caudina and Echinus by Koizumi (1932) and Berger & Bethe (1931). In addition, the permeability of several invertebrates to salts has been inferred from data on the osmotic pressure and changes in weight of the animals when immersed in diluted sea water, the inferences being supplemented in a few instances by chemical analyses. In this category we have the polychaete Nereis diversicolor (Ellis, 1937), several crabs (Margaria, 1931 ; Hukuda, 1932; Bateman, 1933; Drilhon-Courtois, 1934; Maloeuf, 1938), the isopod Mesidotea (Bogucki, 1932), and the triclad turbellarian Procerodes ulvae (Pantin, 1931a).

Accepting then the permeability of marine invertebrates to the ions of sea water, the next step is to find out how far the body fluids are in physico-chemical equilibrium with sea water. Does the chemical relation between the blood and sea water across the permeable portions of the integument conform to a simple equilibrium such as would be observed in a dialysis experiment across an inert collodion membrane? If not, what are the factors concerned in rendering the equilibrium dynamic and not static?

An attempt to answer these questions has been made in the case of three invertebrates, the sea urchin (Echinus esculentus L.), the lobster (Homarus vulgaris Milne-Edwards), and the edible crab (Cancer pagurus L.). After analysis of the blood, as initially obtained from these animals, and the sea water in which they were living, dialysis experiments were set up by separating samples of the blood by a collodion membrane from the same sea water. The blood was again analysed after equilibrium had been attained.

Blood from Cancer and Homarus was collected in a pipette through the arthrodial membranes at the bases of the legs. By withdrawing it into a vessel containing a little powdered heparin, clotting was prevented. After aggregation of the cells, the blood was centrifuged and the clear plasma used for analysis and experiments. Antennary gland secretion of Cancer was obtained by raising the operculum of the gland with a needle and withdrawing the secretion by means of a glass cannula.

The coelomic fluid of Echinus was obtained by boring two holes in the shell opposite each other and draining off the fluid. After the few cells present had aggregated, the somewhat opalescent sea-water-like fluid was ready for analysis.

The chemical methods described by Robertson & Webb (1939) were used for the determination of sodium, potassium, calcium, magnesium, copper, chloride, and sulphate in 1 ml. samples of the body fluids and the sea water. Their special applicability to the analysis of sea water and fluids resembling it in composition had been thoroughly demonstrated. Only in the case of the magnesium determination in Homarus was a modification necessary, since the magnesium content was only about a seventh of that of sea water. This small amount of magnesium failed to come down in the second precipitation as magnesium hydroxyquinolate, due undoubtedly to the relatively high concentration of ammonium chloride in the solution. The whole of the second solution was therefore evaporated with a drop of 25 % sulphuric acid and ashed in a silica crucible, thus removing the ammonium salt. The magnesium was then precipitated with a considerable excess of hydroxyquinoline as in the first precipitation, and the analysis completed as usual.

Protein was determined in 1 or 2 ml. samples (5 ml. in Echinus) by the standard gravimetric method (Peters & Van Slyke, 1932, p. 688). Water contents were obtained by evaporating 1 ml. samples and drying the residue in an oven at 100 ° C. for several hours. In any experiments on equilibrium, it is specially desirable to give analyses as parts per unit weight of water, and not parts per unit volume of fluid which do not take into account the space occupied by the colloids.

Collodion membranes of pyroxylin were made according to the instructions of Adair (1925). All the membranes used were completely impermeable to protein, as determined by complete recovery of protein in the dialysed plasma. The dialysis apparatus was similar in design to Adair’s osmometers. Air passed through sea water covered with a layer of toluene was bubbled slowly through the external sea water which had an initial pH of 7 · 9 –8·0. Dialysis was allowed to proceed at room temperature (13–15· 25°C.) for 30–65 hr. No bacterial infection of the blood was observed or any tendency to precipitation, but the pH of the sea water at the end of dialysis had fallen to 7· 5–7· 6. Since the normal pH of crustacean blood appears to be between 7· 4 and 7· 8 (Quagliariello, 1916; Duval, 1924), the slight fall in p of the dialysate is less disadvantageous than at first appears.

The dialysis experiments were done in duplicate. Millport sea water has been analysed by Robertson & Webb (1939), and the sea water figures in the following tables were calculated from the ionic ratios established by them and from the chlorinity of the sample. The slight changes in the sea water after dialysis were calculated from the known volumes of plasma and sea water and the change in composition of the plasma.

The inorganic composition of the perivisceral fluid of an Echinus was found to be identical with that of the sea water in which the specimen was living, any slight differences being within the experimental error of the methods (Table I). On dialysing the fluid against the same sea water, there was no significant change in composition. Consequently, the system appears to represent a case of complete physico-chemical equilibrium between the body fluid and the external sea water.

Analysis of the pooled blood of three specimens, two females and one male, showed that the concentrations of Na, K, and Ca ions in the plasma were higher than those of the sea water, while Mg and SO₄ were much lower. After dialysis, the composition of the blood approached much more closely to that of the sea water; but, as seen in Table II, the concentrations of Na, Ca, and Mg in the plasma are greater than the corresponding values in the sea water, while the concentration of Cl is lower. This result is to be expected, on the grounds of the Gibbs-Donnan equilibrium effect between a protein-containing solution and its dialysate. The slight opposite tendency shown by the other cation K and the anion SO. is probably not significant, but due to the relatively greater experimental error of the methods of analysis for these two ions.

The significantly higher amounts of Ca and Mg in the dialysed blood as compared with Na is interpreted as showing that a proportion of these radicles is in the form of undissociated compounds with the colloids, and as such does not come under the equilibrium laws.

From a consideration of the data as a whole, it is clear that the ionic composition of the blood of Homarus must be actively maintained.

Blood and antennary gland secretion were obtained from two crabs, a male and a female. The plasmas and also the secretions were mixed proportionally, so as to retain the relative ionic composition of the two fluids.

Table III shows that Na, K, and Ca were present in greater concentration in the original blood than in the sea water in which the crabs were living, while Mg, Cl, and SO₄ were lower in the blood. After dialysis the differences in relative concentrations were very much reduced, and one must conclude that the original differences were actively maintained.

The antennary gland fluid showed important differences in composition from the blood (Table IV). All the ions were lower in concentration in the secretion, with the exception of Mg and SO₄ which were markedly higher. These data would seem to indicate that the antennary gland may be of great importance in controlling the ionic composition of the blood. The differences between the original equilibrium of the blood with sea water and the experimentally found equilibrium after dialysis could, at least in part, be caused by the activity of this organ.

The extent by which the original plasmas of Homarus and Cancer and the perivisceral fluid of Echinus differed from static equilibrium with the surrounding water is shown in Table V.

The apparently complete static equilibrium of Echinus body fluid is correlated with the peculiar morphological features of echinoids. The water vascular system brings sea water into intimate relation with the perivisceral fluid, and diffusion equilibrium must exist between the two through the thin walls of the tube-feet ampullae. In addition the perivisceral fluid is separated only by a thin coelomic membrane from the lantern coelome, the fluid of which fills the external gills. Finally, there are no definite excretory organs to play a part in altering the composition of the body fluid relative to sea water. Absorption of salts present in the food may cause slight temporary increases in the concentration of any of the ions in the coelomic fluid, but outward diffusion of the excess ions would soon restore equilibrium.

Homarus and Cancer differ from Echinus in two important features, the presence of significant quantities of colloids, particularly protein, in their blood, and the presence of excretory organs, the antennary glands.

Let us consider a system consisting of sea water bounded by a membrane permeable to salts and water, and immersed in a large volume of sea water. Complete osmotic and ionic equilibrium will exist between the two solutions. Suppose the inner solution contains protein to which the membrane is impermeable. The colloid osmotic pressure of the protein will cause an absorption of water accompanied by salts since the membrane is permeable to both. Equilibrium will be reached when the tension of the boundary membrane equals the colloid osmotic pressure. If a colloid-free filtrate were to be passed into the sea water at one area of the membrane, the osmotic pressure of the colloids would immediately cause an intake of water through the rest of the membrane.

The conditions in the marine decapod Crustacea resemble those of the model. One can visualize the passage of a colloid-free filtrate of the blood through the epithelium of each antennary gland if the pressure in the vessels supplying the glands exceeds the colloid osmotic pressure. This concept of filtration is supported experimentally by the work of Picken (1936) and Bethe et al. (1935). The increased hydrostatic or blood pressure in parts of the system is probably determined by the tone of the body muscles as well as by the pulsations of the heart (Picken). Intake of water to balance the output through the glands would occur where the internal fluid near the more permeable portions of the integument had a hydrostatic pressure lower than the colloid osmotic pressure.

If we accept the process of filtration as the first stage in the formation of the excretory fluid of Cancer, the final composition of the fluid (Table IV) must have been attained by the processes of resorption and secretion. Resorption of Na, K and Cl, and secretion of Mg and SO₄ have taken place, while the concentration of Ca has apparently remained unchanged, since its value in the antennary gland fluid is approximately that of the diffusible Ca of the blood plasma. An alternative explanation of the high concentrations of Mg and SO₄ in the excretion is obtained by postulating resorption of water as well as ions. This would appear less probable, since the marine decapod Crustacea produce an excretion approximately isotonic with the blood (Schlieper, 1935; Picken, 1936) and are not faced with the necessity of conserving water.

The output of fluid through the antennary gland in crabs is balanced by an intake of water through permeable portions of the integument. By plugging the openings of the glands of a large Cancer pagurus (300 g.) and measuring the increase in weight of the animal after 4–5 hr., the intake of water was found to equal 10% of the body weight per 24 hr. in one experiment, and 3 % on repeating the experiment the following day. Since the blood is maintained in approximate isotonicity with sea water (Hukuda, 1932), the fluid absorbed must be approximately isotonic with both fluids, but its precise ionic composition is unknown at present. It is clear from Table III that the process of diffusion cannot account for the intake of Na, K, and Ca ions, since there is an adverse concentration gradient in each case. Active transport of these ions against the concentration gradient by the activity of the cells of the surface membranes must take place. There are insufficient data at the present time to decide whether the intake of water is caused by the colloid osmotic pressure of the blood or by an active process. A necessary condition for the former alternative is a colloid osmotic pressure higher than the hydrostatic pressure in the parts of the blood system into which the fluid is drawn. Even in the sternal sinus, where perhaps the lowest hydrostatic pressures might be expected, Picken found that twenty-seven of forty-one specimens of Carcinus had values for hydrostatic pressure greater than the colloid osmotic pressure of the blood. The hydrostatic pressures were minimal values taken when the animals were quiescent. Picken’s data, therefore, rather point towards some active process being necessary for the uptake of water.

The composition of the blood would seem to be regulated by two interacting mechanisms, (1) the excretion of a fluid differing from the blood in ionic composition, and (2) an active control by the surface membranes of the passage of Na, K, and Ca ions into and from the crab, and perhaps also of the passage of water. The other ions, Mg, SO₄ and Cl may pass into the crab by diffusion through the surface membranes. Work is performed by the activity of the heart in producing the hydrostatic pressure in the antennary arteries necessary for filtration, by the epithelium of the antennary gland in altering the composition of the filtrate by resorption and secretion, and again by the boundary membranes in the transport of ions against concentration gradients.

It has been assumed up till now that the output of ions in the antennary gland secretion is balanced by the absorption of equivalent ions from sea water through the body surface. This is not always and perhaps never exactly the case. The net result is sometimes an actual loss or gain of particular ions. Occasionally, other factors operate. Thus immediately after moult much more Ca is absorbed by Carcinus and Cancer (Robertson, 1937, and unpublished) than is excreted, this being related to the need for Ca for hardening the new exoskeleton. Immediately prior to ecdysis, on the other hand, large amounts of Ca pass into the surrounding sea water. This Ca has probably accumulated in the body fluid during the process of mineral resorption from the old exoskeleton, and has diffused through the surface membranes, although a certain proportion has undoubtedly been excreted by the antennary glands.

Absorption of salts with food must also affect the composition of the blood to some degree, but this factor is difficult to assess at present.

Within the last ten years several authors have published analyses of the body fluids of marine invertebrates. Unfortunately, the micro-methods used by most of them were quite unsuitable for such analyses. When they were applied by these authors to sea water, a medium of practically constant ionic composition (Thompson & Robinson, 1932), variations from the accepted ionic ratios of up to 40% were found (Robertson & Webb, 1939). The data of Bethe & Berger (1931) included in Table VI are seriously vitiated by the inaccuracy of the methods used in their determination. Besides finding wide variations in the ionic composition of Heligoland sea water within an interval of 2 months, they claim that in Echinus each of-the four principal cations is in higher concentration in the coelomic fluid than in sea water. A further claim is made that the fluid in the water vascular system differs from the coelomic fluid in composition, particularly in having a much higher K content. These results are certainly erroneous. The present writer’s data showing that the perivisceral fluid and the surrounding sea water are identical in composition are supported by the spectrographic analyses of Webb (1937).

From the very nature of the dynamic equilibrium between crustacean blood and sea water a certain amount of variation in blood composition within a single species is to be expected. Apart from such individual variations others occur, especially in relation to moulting. Post-moult Carcinus have lower Ca concentrations than normal intermoult crabs (Robertson, 1937). Since a proportion of the Ca in the blood is bound to colloids, the concentration of the principal colloid, protein, also influences the Ca level.

The data on Homarus and Cancer in Table VI were obtained from intermoult specimens. The present analysis of Homarus blood agrees fairly well with that of Macallum (1910) in which macro-methods were used, but some of Bethe and Berger’s figures, especially the high K content, differ from the other two analyses. Compared with Cancer pagurus, Homarus has much less Mg and SO₄ in its blood. Until more is known about the conditions which determine these low values, it is perhaps premature to discuss them. It is possible, as Macallum believed, that the ancestors of the marine members of the Astacura (lobsters and crayfishes) were originally fresh-water forms as Astacus is at present. Whether the tissues of these ancestors became adapted to prevailing low concentrations of Mg and SO₄ and are still adapted to low concentrations of these ions can only be a speculation. If, as is more probable, the fresh-water forms were derived from marine ancestors, the low values of these ions in the blood of a lobster compared with a crab have to be accepted as a fact which has no apparent explanation at present.

In a stimulating review Pantin (1931b) pointed out that the body fluids of marine invertebrates had relatively more K and less Mg and SO₄ than sea water, although osmotic pressure differences between the fluids were very small. Taking into consideration the facts as known at that time, he put forward a hypothesis to explain this distribution. Pantin believes that the concentrations can be explained on a basis of the relative mobilities of the ions and the activity of the excretory organ. “Suppose a vessel is allowed to float in an aqueous salt solution to which it is permeable, and that the permeability is not the same for all the substances in the solution. After a long time all these substances will have penetrated and there will be an internal fluid identical with the outside solution. But the first substances to penetrate will be those that do so most rapidly. Consequently the internal fluid will at first differ from the external by accumulation of the most rapidly penetrating substances. This condition will gradually give way to equilibrium as time goes on. But if the accumulating fluid is pumped out of the vessel at a steady rate, substances from outside will continuously pass in through the walls of the vessel. A steady state is arrived at in which the internal fluid continuously maintains a composition different from that outside the vessel by having a greater proportion of the most rapidly penetrating substances.” The differences in ionic mobility are greatly increased by the presence of negative charges on biological membranes. On this hypothesis, the fast moving K ion has accumulated in the body fluids of marine invertebrates at the expense of the slow moving Mg and SO₄ ions.

Baldwin (1937) has accepted this hypothesis, while admitting that it is perhaps too simple to explain the observed facts completely.

To the present writer the theoretical basis of the hypothesis seems quite fallacious. In an animal like Echinus the internal medium is in diffusion equilibrium with sea water, the concentrations of ions inside and outside being practically identical. Suppose there is now developed an excretory organ which actively filters off fluid. No matter how quickly the K and Na ions move, the incoming fluid would never contain more of these ions per unit volume than the body fluid, unless forces other than diffusion were at work. If equilibrium between the sea water and body fluid concentrations of Mg and SO₄ was not attained owing to the slowness of penetration of these ions, water would be withdrawn from the body fluid to compensate for the decreased osmotic pressure of the incoming fluid. The result of such differential mobilities would be the diminution in volume of the body fluid. Ca is a slower moving ion than Na (Koizumi, 1932, 1935). Yet, as seen in the Cancer and Homarus analyses (Table V), Ca is accumulated to a much greater extent than Na.

In order to explain the ionic composition of the body fluids of marine invertebrates with excretory organs, it has to be assumed

(1) that K, Na and Ca ions are actively transported through the surface membranes against concentration gradients, and

(2) that the inward diffusion of the slowly moving ions Mg and SO₄ and the active transport of Na, K and Ca are co-ordinated at the body surface so that the incoming fluid has approximately the same osmotic pressure as the blood and the excretory fluid which it is replacing.

Such assumptions, of course, destroy the simple concept of attributing “the observed differences between the internal and external media to different ionic mobilities, exaggerated by the presence of negative charges on the boundary membranes, the whole effect being called into play by the activity of the excretory organ” (Baldwin).

In Cancer it has been seen that the composition of the blood is controlled by the activity of the antennary glands and the body surface. The same mechanism probably controls the blood composition of Homarus, and it would seem that the antennary gland is partly responsible for the low Mg and SO₄ concentrations in the blood as compared with sea water. Peters (1935) has shown that Homarus lacks the canal portion of the gland which is present in Astacus. This portion appears to be responsible in Astacus for the resorption of salts which results in the excretion of a fluid hypotonic to the blood. Both Peters and Schlieper (1935) therefore conclude that no resorption of salts takes place in Homarus, but it is probable that resorption of K, Na and Ca ions does occur in addition to an excretion of Mg and SO₄, although the excretory fluid is isotonic to the blood (Schlieper).

The importance of the antennary gland in modifying the ionic composition of the blood was shown first by Bialaszewicz (1932) who found that in Maia squinado the gland fluid compared with an ultrafiltrate of the blood contained more SO₄, Ca and Mg and less K per unit volume. Scholles’ (1933) analyses of the body fluids of Eriocheir sinensis showed similarly that the antennary gland played a part in regulating the high Ca and low Mg concentrations in the blood of this estuarine crab when in sea water. The more complete and probably more accurate analysis of the excretory fluid of Cancer pagurus in the present paper points again to the importance of the antennary glands in maintaining the ionic composition of the blood, a function at least as important as the longer known one of nitrogenous excretion.

I am indebted to the Carnegie Trustees for a Fellowship during the tenure of which this work was carried out, and to the Managers of the Balfour Fund for a grant towards expenses incurred at Millport.