1. The effects of different dilutions of Locke solution on the electrical activity of the isolated pedal ganglion of the slug can be reproduced by adding different concentrations of glucose or mannitol to a given concentration of Locke.

  2. This indicates that certain cells in the pedal ganglion are sensitive to the osmotic pressure of the solution and not its ionic concentration.

  3. The preparation is sensitive to slow changes in the concentration of the bathing medium. The cells increased their activity when the bathing solution was slowly changed from 0-7 Locke to o-6 Locke, the change taking 43 min. This corresponds approximately to a change of 1 % of the body fluid concentration over 4 min. Such rates of change are found in the normal intact animal.

  4. The sensitivity of the preparation compares well with that of the mammalian osmoreceptors.

The blood in the living slug can have a wide range of concentrations, depending on whether the animal is in a dehydrated or hydrated condition. In an earlier paper by Hughes & Kerkut (1956) it was shown that one could make a quantitative study of the spontaneous activity of a single unit in the isolated pedal ganglion of the slug. It was found that this preparation was sensitive to changes in the concentration of the bathing solution; the cells were more active in dilute Locke solution and less active in concentrated Locke solutions. Furthermore, the preparation was sensitive to experimental changes of concentration one-twentieth those that can occur in the blood of the living animal. However, it was not clear whether this effect was due to the osmotic pressure of the solution or whether it was due to the change in the ionic concentration, i.e. the effective concentration of Na+, K+, Ca2+, Mg2+. The experiments described in this paper will show the effect of solutions of identical ionic composition but differing osmotic pressure on the activity of the pedal ganglion cells.

In the earlier account the changes found to occur in the isolated pedal ganglion following concentration or dilution of the Locke solution were postulated as being similar to those that would occur in the intact animal following dilution or concentration of the blood, but there was a major difficulty to this interpretation. We had studied the effect of sudden changes in the concentration of the bathing medium, but the changes that take place in blood concentration following hydration or dehydration of the slug are gradual. It was therefore necessary to determine the effect of similar gradual changes on the activity of the isolated pedal ganglion before one could apply these results to the living animal.

The apparatus used in these experiments is similar to that described by Hughes & Kerkut (1956). The slug brain was dissected and immersed in earthed Locke solution. A fine tungsten wire electrode was inserted into the ganglion and the impulses led to a Leak amplifier and then to a Cossor double-beam oscilloscope. The impulses were monitored on the oscilloscope and then counted on a dekatron scaler (Kerkut, 1955). Identical results were obtained from both Agriolimax reticulus and Arion ater.

(i) Effects of solutions of differing osmotic pressure on the pedal ganglion

The effect of solutions of identical ionic composition but different osmotic pressure is shown in Fig. 1. Locke solution was diluted to half strength (0 5 Locke) and glucose added to it; 17·71 g. of glucose were added to each litre of 0·5 Locke, and this solution was referred to as 0·5 G. The preparation showed an activity of approximately 10 impulses/half-minute in 0·7 Locke, and the mechanical act of changing the solution as illustrated by a change from 0·7 Locke to 0·7 Locke had no effect on the activity.

When the solution was changed from 0·7 to 0·5 Locke there was a marked increase in the activity from 10 to 45 impulses/half minute. When the bathing solution was now changed from 0·5 Locke to 0-5 Locke plus glucose (0·5 G), the activity fell from 45 to 14 impulses/half minute. The preparation regained its activity to 0·5 Locke and fell off again in 0·7 Locke. It is interesting to note that there was little or no change in the activity of the preparation when the solution was changed from 0·7 Locke to 0·5 G. Thus 0·5 G behaved like 0·7 Locke, even though it had a different ionic concentration.

The glucose did not have any noticeable deleterious effect on the preparation, since after treatment with 0·5 G the ganglia behaved quite normally to subsequent treatment with 0·5, 0·7 and 0·9 Locke solutions.

Similar results were obtained by adding mannitol to Locke solution. Results of some of these experiments are shown in Figs. 2 and 3. Fig. 2 shows the effect of adding different concentrations of mannitol to 0·5 Locke. The preparation showed 13 impulses/half minute in 0·5 Locke + 17·7 g. mannitol/1.; 4 impulses/half minute in 0-5 Locke +35-42 g. mannitol/1.; 27 impulses/half minute in 0-5 Locke.

The effect of mannitol is further demonstrated in Fig. 3. The rate of activity was constant in 0·5 Locke and greatly increased if the solution was changed to 0·25 Locke. If, however, the preparation was placed in solutions of 0·5 Locke which had various concentrations of mannitol added, then a series of curves was obtained which showed that the activity was highest in those solutions that contained the least mannitol, and lowest in those solutions that contained the most mannitol. The rate of activity was thus inversely proportional to the concentration of mannitol in the solution, even though all these solutions had the same ionic composition. These experiments indicate that the pedal ganglion is probably responding to changes in the osmotic pressure of the solution and not to the total ionic concentration.

(2) Gradual changes in the concentration of the bathing medium

As mentioned in the introduction, the changes in blood concentration following hydration or dehydration of the living animal are probably gradual. It was therefore necessary to decide whether the pedal ganglion in the intact animal would show changes in activity following gradual changes in the concentration of the bathing solution.

We were lucky in our investigation of this effect in that one of the first preparations so studied showed peculiar properties. This preparation was active in 0·7 Locke and more dilute solutions, but was totally inactive in 0·8 Locke and more concentrated solutions. We therefore immersed the ganglion in 0·8 Locke and gradually diluted the medium. A burette containing dilute Locke or distilled water was placed to one side of the bath and the rate of flow controlled by the burette tap.

By knowing the initial volume of fluid in the bath and the volume of fluid added, one could calculate the dilution. The preparation was protected from the burette by a series of baffles, and the solution was thoroughly mixed by an aerator. The first change took 10 min. to complete, the change being from 0·8 Locke to 0·4 Locke. It will be seen from Fig. 4 that the activity of the preparation increased after dilution. Fig. 5 shows the effect of a slower dilution, the change from 0·8 Locke to 0·4 Locke taking 50 min. The times taken for the solution to reach the various dilutions intermediate between 0·8 Locke and 0·4 Locke are shown on the graph.

The effect of smaller gradients (i.e. smaller dilutions over a longer time) is shown in Fig. 6. In Fig. 6 the effect of a gradual change from 0·7 Locke to 0·6 Locke over 43 min. is shown. Here we have plotted the running mean of the activity against time, this method having the advantage of smoothing out the curve and showing the general trend more clearly. Though the record shows some perturbations there is a gradual increase in the level of activity.

Fig. 7 shows the effect of diluting the bathing medium from 0·5 Locke to 0·4 Locke, the change taking place over 44 min. The effect of gradual dilution is to increase the activity of the preparation, though it will be noted that there was a certain degree of adaptation, i.e. after some time the activity slowly fell off. Even so the activity at the end of the dilution was clearly greater than that observed at the beginning of the dilution.

We chose to examine the effect of gradual changes by diluting rather than by concentrating the medium, since the normal trend of a preparation remaining in a given solution is gradually to diminish its rate of activity. Thus we might have observed a slow decrease in activity following the gradual concentration of the solution, but would have been unable to decide whether this was due to ageing of the preparation or the effect of the more concentrated solution.

The experiments described in the previous section raise three points: (a) the relative importance of ionic and osmotic changes ; (i) whether the results from the in vitro experiments can be applied to the intact animal; (0) the agreement of the present results with those previously described for other osmoreceptors.

One can see from the experimental results that the important factor in changing the solution is the change in osmotic pressure of the solution. Diluting the solution and so lowering its osmotic pressure makes the pedal ganglion more active, whilst concentrating the solution and so raising its osmotic pressure makes the preparation less active. The effects of dilute and concentrated solutions can be obtained by using solutions containing the same ionic concentration but differing amounts of glucose or mannitol.

This is not to say that the ionic concentrations are of no importance to the animal, though they are most clearly shown when the relative concentration of different ions such as calcium or magnesium is changed. These experiments are for the most part artificial, since such conditions do not often occur in nature. However, Lustig, Ernst & Reuss (1937) have shown that the hibernating snail has a higher concentration of magnesium in its blood than has the active animal. Experiments on the effects of changes in ionic balance will be presented in a later paper. In the living animal the effect of hydration or dehydration is presumably to alter the total concentration of the ions and thus the effective osmotic pressure of the blood; there is no evidence that the relative concentrations of ions are altered. Thus we may conclude that the important factor in the intact animal is the effective osmotic pressure of the blood.

It is now necessary to show that the osmotic pressure changes in our experiments on the isolated pedal ganglia are similar to the changes that can occur in the intact animal. In the live desiccated slug, the blood is isotonic with a 1·4 Locke solution. In a live hydrated animal the blood is isotonic with a 0·4 Locke solution. The isolated pedal ganglion is sensitive to a sudden change from 0·7 Locke to one of 0·65 Locke, i.e. a change one-twentieth the intensity of that which can occur in the living animal. However, this comparison is not strictly valid, since in the living animal the changes are probably gradual. Thus we must consider the rate of change and not only the absolute change itself.

Experiments show that a slug can increase its body weight by 30% in 1 hr. hydration. The increase in weight is mostly due to the uptake of water into the blood, the blood being the most sensitive of the tissues to hydration or desiccation (Pusswald, 1948). If we take the standard osmotic pressure of slug’s blood as isotonic with 0·7 Locke, then a 30% dilution of the blood would make it approximately isotonic with 0·5 Locke. The change in the slug’s blood concentration would be 0·7·0·5 over 1 hr. We have shown that the isolated preparation is sensitive to a slower rate of change than this, i.e. 0·7 Locke to 0·6 Locke in 43 min. Thus the experimental conditions are if anything less vigorous than those that can occur in nature.

In studying the effect of sudden changes of concentration of the bathing solution on the activity of the pedal ganglion, Hughes & Kerkut (1956) found that some preparations showed adaptation. That is, after the solution had been diluted the activity of the preparation increased, reached a peak after some 20 min. and then fell off. This adaptation might be of considerable importance in the living animal. Thus the normal animal in the rain would have water slowly entering its body and diluting its blood. If the rate of dilution was equal to the rate of adaptation, the pedal ganglion would be unaffected by the change in the blood’s concentration. In fact, experiments on the effect of gradual changes in concentration show that unless the rate of hydration or dehydration is extremely low there is a detectable effect on the activity of the preparation. Though some adaptation may occur following dilution, all preparations show that the activity after dilution is greater than the initial activity.

The reactions of osmoreceptors have been studied in other animals. Verney (1947) showed that the mammalian receptors were a series of nerve cells lying in the hypothalamus. If the blood was diluted, these cells became vacuolated (Jewell, 1953), and it is thought that in some way they affect the secretion of the anti-diuretic hormone.

Verney showed that the mammalian osmoreceptors were sensitive to a 2 % change in the osmotic pressure of the blood. If we take blood as being isotonic with Locke solution, then the mammalian receptor is sensitive to a sudden change of 2 % Locke, whilst the slug pedal ganglia are sensitive to a sudden change of 5 % Locke. On the other hand, the slug can detect a change of 10% Locke taking place over 43 min.i.e. a change of the order of 1 % in 4 min.

C. von Euler (1953) showed that the injection of 2% NaCl (Locke is approximately 1 % NaCl) into the hypothalamic region of a cat led to a slow potential (recorded between the supraoptic region of the hypothalamus and the frontal air sinus) lasting approximately half a minute. A similar potential was evoked by injecting 10% glucose solution, whilst injection of tap water brought about a slow potential of opposite sign.

There are certain references in the literature to the sensitivity of nerve cells to osmotic changes. Libet & Gerard (1938, 1939), studying the waves of electrical activity in the isolated frog brain, found that the potentials were in some cases sensitive to changes in the osmotic pressure of the solution. Addition of glucose increased the amplitude of the waves in some cases from 45 to 120 μ V. The wavelength was slightly increased from 0·2 to 0·25 sec., and thus the frequency was slightly lower. This then agrees with our findings in the slug that increasing the osmotic pressure results in a lower frequency of discharge, though Libet & Gerard were here studying slow waves and not the activity of single units.

Fatt & Katz (1952) found a different response to osmotic change. They studied the spontaneous subthreshold potentials in the motor end-plate of the frog. These were affected by changes in the osmotic pressure of the bathing solution. A 50% increase in the osmotic pressure (by addition of sucrose) led to an increase in the frequency of the potentials from 2 to 90/sec. Further experiments showed that raising the osmotic pressure 30% increased the rate from 15 to 150/sec., whilst decreasing the osmotic pressure by 50% reduced the frequency from 28 to 0·9/sec. Thus in the frog subthreshold end-plate potentials, the frequency is increased. by increasing the osmotic pressure of the solution, whereas in the slug the frequency of the potentials is increased by decreasing the osmotic pressure of the solution.

Alanis & Matthews (1952) showed that the nerve cell bodies in the ventral horn of the frog spinal cord were sensitive to mechanical pressure. Pressure on the cell body tended to depolarize the membrane and facilitate the development of action potentials. We have noticed in the slug that pressure on the ganglion tends to elicit a higher frequency of action potentials.

If the neurone behaves like a single osmometer it should swell in dilute solutions and so distend the membrane. This would have an effect similar to that of mechanical pressure on the membrane and so facilitate the development of action potentials. On the other hand, it would also be necessary to postulate some means of adaptation, since animal cells, unlike plant cells, are not surrounded by a thick cellulose sheath which prevents them from swelling. The pedal ganglion is surrounded by a collagen sheath, but this appears to play little part in the sensitivity of the preparation to osmotic changes, since the preparation is still sensitive to osmotic changes even if the sheath is slashed. It is more probable that the nerve membrane either alters its permeability to the substances outside the nerve, or that the membrane in some as yet unknown way changes its sensitivity.

Alanis
,
J.
&
Matthews
,
B. H. C.
(
1952
).
The mechano-receptor properties of central neurones
.
J. Physiol
.
117
,
59P
.
Von Euler
,
C.
(
1953
).
A preliminary note on slow hypothalamic potentials
.
Acta phyriol. scand
.
29
,
133
6
.
Fatt
,
P.
&
Katz
,
B.
(
1952
).
Spontaneous subthreshold activity at motor nerve endings
.
J. Physiol
117
,
109
28
.
Hughes
,
G. M.
&
Kerkut
,
G. A.
(
1956
).
The effect of variations in concentration of Locke solution on the electrical activity of the isolated pedal ganglion of the slug
.
J. Exp. Biol, (in the Press)
.
Jewell
,
P. A.
(
1953
).
The occurrence of vesiculated neurones in the hypothalamus of the dog
.
J. Physiol
.
121
,
167
81
.
Kerkut
,
G. A.
(
1955
).
An inexpensive Dekatron Scaler
.
Electron. Engng
,
27
,
378
80
.
Libet
,
B.
&
Gerard
,
R. W.
(
1938
).
Chemical control of the isolated frog brain
.
Amer. J. Physiol
.
123
,
128
8
.
Libet
,
B.
&
Gerard
,
R. W.
(
1939
).
Control of the potential rhythm of the isolated frog brain
.
J. Neurophyriol
.
2
,
153
69
.
Lustig
,
B.
,
Ernst
,
T.
&
Reuss
,
E.
(
1937
).
Die Zusammensetzung der Blutes von Helix pomatia bei Sommer und Wintertieren
.
Biochem. Z
.
290
,
95
8
.
Pusswald
,
A. H.
(
1948
).
Beitrage rum wasserhaushalt der Pulmonaten
.
Z. vergl. Physiol
.
31
,
227
48
.
Verney
,
E. B.
(
1947
).
The antidiuretic hormone and the factors which determine its release
.
Proc. Roy. Soc. B
,
135
,
25
106
.