1. A study has been made of the relation between blood, urine and medium concentrations in the two amphipod Crustacea G. duebeni and G. pulex.

  2. G. duebeni produces urine hypotonic to the blood but hypertonic to the medium when it is in media more dilute than 50% sea water.

  3. G. pulex forms urine which is hypotonic both to blood and medium when in 2−20% sea water.

  4. G. duebeni begins to form hypotonic urine within 2 hr. of transference from no to 160% sea water to fresh water. Hypotonic urine formation begins in these circumstances when the blood concentration is up to twice that at which hypotonic urine is formed by animals fully adapted to their medium.

  5. It is concluded (a) that the concentration of urine produced by G. duebeni is not dictated solely by the absolute level of the blood concentration; (b) that the formation of urine hypotonic to the blood in a brackish-water animal functions primarily as a means of conserving ions in the body ; (c) that the ability to regulate the concentration of the urine with rapidity will be important in an animal living in environments of fluctuating salinities.

The amphipod, Gammarus duebeni, is essentially a brackish-water species though it extends its range into fresh water in a number of places. (Kinne, 1959; Hynes, 1954.) It occurs in fresh water apparently only when G. pulex is absent (Hynes, 1954) and in brackish water also competition with G. salinus and G. zaddachi appears to limit its range (Kinne, 1959). In consequence the distribution of G. duebeni is discontinuous and it is most commonly found in physiologically ‘difficult’ habitats where its competitors are absent. Such habitats include brackish-water lagoons, splash-zone rock pools, ditches near the sea and salt marsh pans (references in Kinne, 1959). Since such habitats are all subject to extensive and rapid changes in salinity as a result of inundation by sea water or fresh water, G. duebeni might be expected to have a high degree of osmoregulatory capacity.

Beadle & Cragg (1940) have shown that G. duebeni tolerates salinites in the range 100 to 2% sea water and can withstand direct transference between these two extremes of concentration. The osmotic pressure of the blood is maintained at a higher level in 2 % sea water than is that of the fresh-water species G. pulex. In more saline media the blood concentration of G. duebeni rises slowly until it becomes effectively isotonic with the medium when the latter is more concentrated than about 50—60% sea water. The mechanisms responsible for the maintenance of the relatively constant blood composition when the medium is less concentrated than 50% sea water have not been fully investigated. Beadle & Cragg (1940) have shown that G. duebeni loses salt at a rapid rate when the animal is placed in distilled water. Rapid changes in the blood concentration may therefore be expected to follow changes in the concentration of the medium. In such circumstances any mechanism tending to retain ions within the body would clearly be of benefit by serving to slow down the change in blood concentration. One such process might be the production of urine hypotonic to the blood.

Schwabe (1933) demonstrated that the fresh-water G. pulex has a longer excretory tubule in the antennary gland than has the brackish and marine species G. locusta. By analogy with the situation in the decapod Crustacea it has been assumed that this indicates that the fresh-water species can produce hypotonic urine. Recent measurements by Hynes (1954) have extended Schwabe’s work by showing that in G. duebeni the eosinophil section of the excretory tubule is almost twice as long as that in G. locusta of similar size and some three-quarters the length of the corresponding section in G. pulex. Hynes concludes that G. duebeni might also be capable of producing hypotonic urine. This observation is of particular interest as no brackish-water’ crustaceans have previously been thought to produce hypotonic urine. Beadle (1943) regards the capacity to produce hypotonic urine as being one of the later physiological modifications to be developed by fresh-water animals. Further, Potts (1954) has concluded that the production of hypotonic urine by a semi-permeable animal in brackish water would contribute little towards decreasing the expenditure of energy on osmoregulation, though having a marked effect if the medium were fresh water.

In the present paper it will be shown that G. duebeni is not only capable of producing hypotonic urine when in brackish water or fresh water, but that it also has the capacity to vary the urine concentration as the blood concentration changes.

Materials

G. duebeni was collected from the salt-marsh at Aberlady (Firth of Forth) and from the Stour estuary at Flatford Mill. G. pulex was obtained from the Braid Bum, Edinburgh. In the laboratory stocks of G. duebeni have been maintained in their natural medium and fed on ‘Bemax’ and Enteromorpha. They reproduced and appeared normal in every way. Species were identified by reference to Segerstråle (1959) and Reid (1944).

Methods

Analysis

Osmotic pressure determinations on both blood and urine were made using the cryoscopic method of Ramsay & Brown (1955). Osmotic pressure is expressed in terms of the concentration of NaCl (inmM./l.) having the same freezing-point depression. Tritium was counted using a standard liquid phosphor (Popop, Naphthalene and 2,5-diphenyl oxazole) in a Panax liquid scintillation counter. 22Na was counted by a standard Geiger-Muller tube and Labgear Dekatron scaler. At least 2500 counts were taken on each sample reducing the expected error of the count to ±2%.

Urine collection

A small hole was burnt with a heated needle through a 1 in. square of thin rubber sheet and the animal was inserted so that it was firmly gripped in the region between the second and third coxal plates. The rear portion of the body, with the gills, thus lay on one side of the membrane and the head and excretory apertures on the other. The membrane was placed over the mouth of a tube containing 50 ml. of the experimental medium, so that the animal’s gills were immersed and allowed room to beat. The entire tube was then immersed about 12 in. below the surface of liquid paraffin. The head and that portion of the thorax exposed to paraffin were carefully dried with filter paper. When the excretory papillae are quite dry urine collects as discrete droplets on the top of each cone.* Such droplets were sucked directly into the pipettes used for freezing-point determinations. Samples were only taken when such a droplet had been observed to form on the excretory papilla so there was no possibility that urine samples were contaminated by extraneous water collecting round the mouth parts.

Blood samples

Blood samples were obtained by snipping off the terminal segments of the flagellum of the first antenna. Successive sampling was possible if only a few segments at a time were taken. The blood was collected into freezing-point pipettes as it emerged and the osmotic pressure was determined at once.

When G. duebeni is acclimatized to media more concentrated than 50% sea water, the urine is isotonic with the blood. In media less concentrated than 50% sea water hypotonic urine is formed. The lower the concentration of the medium the greater is the concentration difference between blood and urine. These results are summarized in Fig. 1. The values shown are all for animals acclimatized for 3 days or longer to the experimental medium. As the concentration of the urine frequently rises when the animals have been in the collecting vessel for some time, only the lowest value recorded for each animal is given. Not infrequently the two urinary papillae produce urine of different concentrations.

A plot of blood concentration against the lowest urine concentration observed for each fully acclimatized individual indicates that there is a general correlation between blood and urine concentrations (circles) Fig. 2. However, it should be noted that animals exposed to dilute media for only a short time before determination (less than 10 hr.), and therefore not yet fully acclimatized, tend to produce dilute urine even though the blood concentration is high (solid circles, Fig. 2).

G. duebeni adapted to fresh water produces urine whose concentration is only about one-third to one-fifth that of the blood. The absolute concentration of this urine is, nevertheless, still markedly greater than that formed by the fresh-water species G. pulex (Table 1)

When G. pulex is acclimatized for 4−7 days to dilutions of sea water in the range of concentration 90−140 mM./I. the urine produced is little more concentrated than it is in fresh water. In consequence the concentration of the urine is markedly less than that of the medium (Fig. 3).

No assessment of the part played by the excretory organ in the osmoregulation of an animal can be made without knowledge of the rate of urine flow. Direct determination of the urine flow by continuous collection of the urine produced was found to be technically rather difficult in the case of G. duebeni. An estimate of urine flow may be made, however, if the blood concentration and the water flux across the body surface are known, and if it is assumed that the net entry of water into the body results solely from the differences between the diffusion of water into and out of the body.

The flux of water across the body wall will be proportional to the activity of the water on either side. A measure of the water activity is obtained by osmotic pressure determinations and activities can be expressed conveniently in terms of the concentration of water in equivalent NaCl solution (1 litre of pure water contains 55·5 mois H2O and 1 litre of N10 NaCl contains approximately 55·3 mois H2O). The net rate of water movement across unit area of the membrane will be K(C0— Ct), where K is a constant and Co and Q are the concentrations of water on the two sides. The proportion of net water movement to influx will therefore be (C0Ci)/C0. If t12 is the half-time of exchange (in minutes) of body water the percentage of body water exchanged in 1 min. is given by the expression . The daily net entry of water into the body will be:

where M is the number of minutes in a day. If the net entry of water into the body involves a bulk-flow component the actual rate of urine flow will be greater than that calculated. No exact measure of urine flow may therefore be obtained, but the formula gives an estimate of the minimum urine flow in terms of percentage of body water per day.

The half-time for water exchange was determined as follows. Animals were acclimatized for 64 hr. in tritiated Cambridge tap water and were then placed in unlabelled tap water. Aliquots of this were taken at intervals and counted. The temperature was 20°C. The half-time for exchange of water was derived from semi-logarithmic plots of Te— Tt against time, where Te is the count at equilibrium and Te the count at time t. The results are given in Table 2 together with calculated values for urine flow per day as a percentage of body water.

Values for urine flow given in the literature are usually presented in terms of percentage of body weight per day. Values for the water content of Gammarus species qnoted by Vinogradov (1953) range from 74 to 83 % of the body weight. The calculated urine flow in percentage of body weight per day will therefore be of the order of 37 and 56 in G. pulex and G. duebeni, respectively.

These values are very high in comparison with the urine flow reported for animals such as: Eriocheir sinensis 4% body wt./day Scholles (1933), Potamobius 4% per day (Scholles 1933), Potamon niloticus 0·05−0·6% per day (Shaw, 19596), Astacus fluviatilis 8·2% per day (Bryan, 1960a), but are not unexpected in view of the small size of the gammarids and consequent high surface/volume ratio. Since urine is formed from the blood it is clear that it is only by the formation of hypotonic urine that a very rapid rate of ion loss from the body can be avoided. Shaw & Sutcliffe (1961), using 40 mg. animals, found a rate of loss of sodium to distilled water of 0·76 μM./hr. in G. duebeni previously acclimatized to 2% sea water and 0·17 μM./hr. in animals acclimatized to 0·25 mM./l. NaCl. In the present experiments animals of this species acclimatized to Cambridge tap water and then placed in distilled water showed a loss of 22Na equivalent to 6·1 % (± 1·6%, n = 5) total body sodium/hr. over a 5 hr. period. The loss of sodium via the urine from animals acclimatized to Cambridge tap water, calculated on the basis of a 40 mg. animal, a urine flow of 71 % of the body weight per day and a urine concentration of 83 mM./l., is 2·4μM./day or 0·10 μM./hr. This calculated value for urinary sodium loss is equivalent to 13 and 60 % respectively of the total losses recorded by Shaw & Sutcliffe for animals in 2% sea water and in 0·25 mM./l. NaCl. Now the urine produced by Gammarus is hypotonic to the blood at these concentrations, but it is clear that if it were isotonic the urinary salt loss would form a considerably larger proportion of the total loss than is observed for Eriocheir and Potomon.

The rate of loss of sodium to distilled water is comparatively high and this suggests that rapid changes will occur in the blood concentration following a change in the concentration of the medium. G. duebeni may be readily acclimatized to 175 % sea water and withstands direct transference from this medium to fresh water. Following transference from 110 to 160% sea water to fresh water there is, as expected, an initial rapid fall in the blood concentration. The urine is isotonic with the blood initially but usually after hr. the urine becomes hypotonic (Fig. 4). The urine becomes hypotonic despite the fact that at this time the blood may be up to twice as concentrated as the highest blood concentration at which animals fully adapted to their medium produce a hypotonic urine. Three further repetitions confirmed this effect (Fig. 5 b). Five G. duebeni transferred from 160 to 190% sea water to 50% sea water, at both of which concentrations isotonic urine would be expected in fully acclimatized animals, all failed to form hypotonic urine within the next hr. (Fig. 5 a). Two of these animals re-tested after 24 hr. were still producing isotonic urine.

The blood concentration falls in a logarithmic manner in animals which do not form hypotonic uriné. In those that do form hypotonic urine the blood concentration falls somewhat less rapidly (Fig. 5 a).

On the basis solely of studies on different species of crayfish it has long been accepted that an antennary gland with a long excretory tubule is potentially able to produce hypotonic urine. Additional support for this contention is now provided by the fact that hypotonic urine is formed by the two gammarids, G. duebeni and G. pulex, both of which have been shown by Hynes (1954) to have excretory tubules which are long in comparison with those of the more marine species G. locusta.

The fact that G. duebeni produces hypotonic urine when in brackish water also illustrates the fact that the capacity to form hypotonic urine is not the prerogative of fresh-water species. Furthermore, since an essentially brackish-water species forms hypotonic urine this ability cannot be regarded as being one of the later evolutionary stages in the physiological adaptation of fresh-water species to their environment as was suggested by Beadle (1943).

It has been postulated that the production of hypotonic urine by animals in fresh water results in some saving in the energy expended in osmoregulation (Potts, 1954) though the basis of this view has recently been criticized by Croghan (1961). Potts also produced evidence to show that the production of hypotonic urine in brackish water would result in little energy saving. It may be presumed therefore that hypotonic urine formation by G. duebeni has some other functional significance. It is suggested here that the production of hypotonic urine may subserve two important functions in G. duebeni.

  1. Ions already within the body are retained and hence less burden is placed on the mechanisms reponsible for active uptake at the body surface.

  2. The formation of hypotonic urine will tend to diminish the rate of fall in the blood concentration following any sudden dilution of the medium. Time is thereby gained for the osmotic adjustments which must be made by the general body cells when the blood concentration is changed.

These two functions may be expected to be of particular importance in small Species maintaining themselves markedly hypertonic to their medium, or rather in hypertonic species whose urine volume per day represents a large proportion of the body volume. (Small species will naturally have a greater urine volume/total volume ratio than larger forms of similar blood concentration and surface permeability.)

It is calculated from the half-time of exchange of tritiated water that the urine flow of G. duebeni, when in fresh water, is approximately equivalent to 70 % of the total body water per day. Urine is formed from the blood and, in most Crustacea so far studied, the blood accounts for some 25−50 % of the total water of the animal. Thus if an isotonic urine were to be formed by G. duebeni in fresh water, the daily salt loss by this route would be at least equivalent to the entire salt content of the blood and might be more than twice this amount. Correspondingly smaller amounts of salt would be lost in brackish water. Production of hypotonic urine will diminish this loss. Crabs such as Eriocheir sinensis and Potomon. niloticus which produce isotonic urine when they are in fresh water have comparatively low rates of urine production and their rate of salt loss via this route is small relative to their total loss. Selection pressure to form a hypotonic urine is therefore presumably less intense in these forms. Beadle & Cragg (1940) found that G. duebeni maintained its blood increasingly more hypertonic to the medium in the range 2−50 % sea water but was effectively isotonic at concentrations above 50−60% sea water. In the present study hypotonic urine formation has not been observed in animals fully adapted to media more concentrated than 50% sea water. Fig. 2 illustrates that the urine is progressively more dilute as the blood concentration is lowered.

It is patent that if the permeability of the body surface remains unchanged the urine flow will increase with the gradient of osmotic concentration maintained between blood and medium, Since the most dilute urine is formed when the urine flow is high, it is clear that the concentration of the urine is not directly related to the time taken for the urine to pass down the excretory tubule. Some form of active control must therefore regulate the urine concentration. It has repeatedly been shown that a lowering of the blood concentration brings about an activation of the processes responsible for the uptake of ions at the body surface of fresh-water and brackish-water animals (Shaw, 1958, 1959a, 1960, 1961; Shaw & Sutcliffe, 1961; Lockwood, 1960). The relation between urine and blood concentrations (Fig. 2) suggests that some similar mechanism may regulate the urine concentration. However, the urine concentration cannot be linked solely to the blood concentration. This follows from the fact that when specimens of G. duebeni are acclimatized to a high salinity and are then transferred to fresh water they start to produce hypotonic urine when the blood concentration has fallen only to a level equivalent to 100% sea water (Fig. 3), twice the concentration at which hypotonic urine formation would be expected in animals fully acclimatized to their medium. Fig. 2 also illustrates that this is a general feature in animals exposed for only short periods to dilute media, the urine usually being markedly more dilute in these animals than would be expected in fully acclimatized animals having the same blood concentration. Specimens of G. duebeni transferred from 160 to 190% sea water to 50% sea water did not produce hypotonic urine. It cannot therefore be postulated that the stimulus causing animals to produce hypotonic urine when transferred from 110 to 160 % sea water to fresh water is merely a rapid fall in the blood concentration. The precise nature of the stimulus is far from clear and requires further investigation.

Comparison of the falling blood concentrations of animals transferred from 110 to 160% sea water to fresh water and from 160 to 190% sea water to 50% sea water illustrates that the production of hypotonic urine by the former decreases the rate of change of blood concentration (Fig. 5). The fall in blood concentration of the animals with isotonic urine approximates to a logarithmic curve, as would be expected if the permeability remained constant.

Shaw & Sutcliffe (1961) have investigated the rate of loss of sodium in G. duebeni acclimatized to different media at the lower end of their salinity tolerance range. They find that animals fully acclimatized to 2 % sea water have a higher rate of sodium loss from the body than have animals acclimatized to 0·25 mM./l. NaCl. They calculate that the difference in the rates of loss could be accounted for if the animals in 2 % sea water produced isotonic urine and if those in 0·25 mM./l. NaCl produced very dilute urine, provided that the urine flow was equivalent to approximately 7·4% of the body weight per hour. The present results indicate that the urine is already hypotonic in 2 % sea water and that the urine flow is about 3 % body water/hr. in fresh water. Some indirect support seems therefore to be provided for an alternative suggestion of Shaw & Sutcliffe that the permeability of the body surface may be varied in animals acclimatized to very dilute media. A critical study of the water fluxes of animals adapted to dilute media should indicate whether or not the regulation of water entry plays any part in this process.

The fresh-water species G. pulex is not normally exposed to saline media and it lacks the capacity to vary the urine concentration when that of the medium is raised. In consequence, in media more saline than about 20−30 mM./l. the urine is hypotonic not only to the blood but also to the medium ; it is possible that as in the case of the crayfish (Bryan 1960b) the urine concentration may rise at very high concentrations of the medium. Urine formation can therefore play little part in the restriction of changes in the blood concentration of G. pulex in saline media. Though the production of urine hypotonic to the medium will not in itself result in a rise in the concentration of the blood as long as the animal is hypertonic to its medium, the maintenance of the blood concentration will depend to a large extent on the degree to which the active uptake of ions at the body surface can be suppressed.

The ability of G. duebeni to vary the concentration of the urine gives this animal a twofold means of regulating its blood concentration, which is presumably of some importance in the fluctuating salinities inhabited by this species.

If the excretory organ is to play any effective part in regulating the blood concentration following a change in the medium, the urinary salt loss must constitute an appreciable proportion of the total loss and it must be possible for the animal to effect rapid changes in the urine concentration. It is not surprising therefore, that changes in the gradient of concentration between urine and blood in G. duebeni can be varied markedly within 2 hr. of a change in the concentration of the medium.

It is hoped to extend the present study to cover urine production in G. locustris, G. zaddachi, G. salinus and G. locusta. Preliminary studies indicate that the first three of these species produce hypotonic urine when they are in dilute media.

I am most grateful to Mr T. Warwick who collected many of the animals for me and who loaned me some specimens of Gammarus species varified by Mr D. M. Reid.

I am also indebted to Drs Shaw and Sutcliffe for permission to examine the proofs of their paper on G. duebeni and G. pulex.

Beadle
,
L. C.
(
1943
).
Osmotic regulation and the faunas of inland waters
.
Biol. Rev
.
18
,
172
83
.
Beadle
,
L. C.
&
Cragg
,
J. B.
(
1940
).
Studies on adaptation to salinity in Gammarus sp. I. Regulation of blood and tissues and the problem of adaptation to fresh water
.
J. Exp. Biol
.
17
,
153
63
.
Bryan
,
G. W.
(
1960a
).
Sodium regulation in the crayfish Astacus fluviatilis. I. The normal animal
.
J. Exp. Biol
.
37
,
83
99
.
Bryan
,
G. W.
(
1960b
).
Sodium regulation in the crayfish Astacus fluviatilis. II. Experiments with NaCl loaded animals
.
J. Exp. Biol
.
37
,
113
28
.
Croghan
,
P. C.
(
1961
).
Competition and mechanisms of osmotic regulation
.
Symp. Soc. Exp. Biol
.
15
. (In the Press.)
Hynes
,
H. B. N.
(
1954
).
The ecology of Gammarus duebeni Lilljeborg and its occurrence in fresh water in Western Britain
.
J. Anim. Ecol
.
23
,
38
84
.
Kinne
,
O.
(
1959
).
Ecological data on the Amphipod Gammarus duebeni. A monograph
.
Ver. Inst. Meeres. Bremer
.
6
,
177
202
.
Lockwood
,
A. P. M.
(
1960
).
Some effects of temperature and concentration of the medium on the ionic regulation of the isopod Asellus aquaticus (L
.).
J. Exp. Biol
,
37
,
614
30
.
Potts
,
W. T. W.
(
1954
).
The energetics of osmoregulation in brackish and fresh-water animals
.
J. Exp. Biol
.
31
,
618
30
.,
Ramsay
,
J. A.
&
Brown
,
R. H. J.
(
1955
).
Simplified apparatus and procedure for freezing-point determinations upon small volumes of fluid
.
J. Sci. Instrum
.
32
,
372
.
Reid
,
D. M.
(
1944
).
Gammaridae (Amphipoda) ; with key to the families of British Gammaridea
.
The Linnean Society of London Synopsis of the British Fauna
, no.
3
,
1
33
.
Scholles
,
W.
(
1933
).
Über die Mineralregulation wasserlebender Evertebraten
.
Z. vergl. Physiol
.
19
,
522
54
.
Schwabe
,
E.
(
1933
).
Über die osmoregulation verschiedener krebse (Malacostracen)
.
Z. vergl. Physiol
.
19
,
183
236
.
Segerstrâle
,
S. G.
(
1959
).
Synopsis of Data on the Crustaceans Gammarus locusta, Gammarus oceanicus, Pontoporeia affinis, and Corophium volutator (Amphipoda Gammaridea)
.
Soc. Sci. Fennica Com. Biol
.
20
, (
5
).
3
23
.
Shaw
,
J.
(
1958
).
Sodium uptake by the crayfish
.
Nature, Lond
.,
182
,
1105
6
.
Shaw
,
J.
(
1959a
).
The absorption of sodium ions by the crayfish Astacus pallipes (Lereboullet). I. The effect of external and internal sodium concentrations
.
J. Exp. Biol
.
36
,
126
45
.
Shaw
,
J.
(
1959b
).
Salt and water balance in the East African fresh-water crab Potamon niloticus (M. Edw
.).
J. Exp. Biol
.
36
,
157
76
.
Shaw
,
J.
(
1960
).
The absorption of sodium ions by the crayfish Astacus pallipes (Lereboullet)
.
J. Exp. Biol
.
37
,
534
47
.
Shaw
,
J.
(
1961
).
Sodium balance in Eriocheir sinensis (M. Edw.). The adaptation of the Crustacea to-fresh water
.
J. Exp, Biol
.
38
,
153
62
.
Shaw
,
J.
&
Sutcliffe
,
D. W.
(
1961
).
Studies on sodium balance in Gammarus duebeni (Lilljeborg) and G. pulex pulex (L
.).
J. Exp. Biol
.
38
,
1
16
.
Vinogradov
,
A. P.
(
1953
).
The Elementary Chemical Composition of Marine Organisms
, pp.
1
647
.
Sears Foundation for Marine Research
.
*

This method for the collection of urine from gammarus was first devised and used by Mr T. D. lies.