The Respiration of Freshwater Snails

The chief aim of this study was to investigate the oxygen consumption of some Danish freshwater snails under varying oxygen concentrations in order to elucidate, if possible, the capacity of the species to exist in nature under unfavourable respiratory conditions.

The first step was to investigate the relation between respiration and starvation; knowledge of this is a prerequisite for an evaluation of oxygen consumption at various times after collection of the animals.

Furthermore, the respiration of the species at varying temperatures has been investigated in order to make possible a comparison between results obtained at different temperatures of the water.

The oxygen consumption of the snails varies with the weight of the individuals. In the case of the limpet Ancylus fluviatilis it was found earlier that the oxygen consumption is proportional to about to^0-73 where w is the live weight (Berg, Lumbye & Ockelmann, 1958). The relation of oxygen consumption to weight for the individuals of each species was also investigated.

After the preliminary studies on the relation of respiration to starvation, varying temperature and weight of individuals, the oxygen consumption in relation to oxygen content of the water was studied.

All experiments were carried out in 1957. The animals were not narcotized and they were all collected from their localities just before the experiments. The observed rate of respiration must, therefore, be regarded as about the same as that in nature under similar conditions. This is not a basal or standard respiration, but an active respiration. The snails move only slowly in the respiratory bottles, but still it is an active oxygen consumption whose relation to environmental conditions has been studied.

A few words, which characterize the localities from where the experimental animals were collected, will be appropriate.

Physa fontinalis (L.), Myxas glutinosa (0. F. Müller) and Lymnaea auricularia (L.). From the sandy shore of the clear, slightly eutrophic lake, Sláenso (Jutland), at a depth of 0*2—1 m.; the first mentioned in April, the others in May.

Lymnaea pereger (O. F. Müller) and L. palustris (O. F. Müller). From stones on the shore of the eutrophic lake, Esrom So (Sealand) at a depth of 0-0·2 m.; June. Also in August among algae washed ashore.

Bithyma tentaculata (L.). From a eutrophic, slightly humic pond in a beech wood, mostly on branches and in the vegetation, to a depth of 1 m.; July, August and October.

Valvata pisdnaHs (O. F. Miiller). In the littoral zone of Esrom So, at a depth of 0-5-2 m. on sandy and gravelly bottom; September and October.

Bithyma leachi (Sheppard). From Esrom So, at a depth of 1-2 m., partly on gravelly bottom at the shore and partly on mud smelling of H₂S in a yachting harbour; October and November.

Concerning the localities of species studied earlier the following data are given:

Acroloxus lacustris (L.). From two eutrophic, slightly humic localities, Karlso and Torkeriso (Sealand), on stems of plants and in cavities below withered leaves. May-June and September-October (Berg, 1952).

Theodoxus fluviatilis (L.). From fresh water: On stones at a depth of c. 0·5 m. in the eutrophic lake Borreso (Jutland); June. From brackish water: On stones at a depth of 0·3-0·5 m.-in Ringkobing inlet (Jutland), salinity c. 9-n%₀; July (Lumbye, 1958).

Potamopyrgus jenkinsi (Smith). From fresh water: In a marlpit at Kolstrup (Jutland) with a freshwater fauna without any brackish-water elements; July. From brackish water: The same locality Ringkobing inlet, as mentioned above (Lumbye, 1958).

The oxygen determinations were made by a polarometric method developed by Bartels and used earlier for similar purposes (cf. Berg, 1953; Berg et al. 1958; Lumbye, 1958).

Where not otherwise indicated the experiments were carried out as follows. The animals were collected in nature, put in a Dewar vessel with water from the locality and brought to the laboratory. The journey lasted c. $12$ hr. The animals were then placed in aerated water in a thermostatic bath, and the experiments started almost at once.

The experiments were made in closed respiration chambers, that is, in bottles containing 4-12 ml. of aerated water, the volumes of which were known exactly. The oxygen content of the water was measured before the experiment. During the experiment the bottles were placed in darkness in the water bath. The duration of an experiment was usually 1 hr. After the experiment the oxygen content of the water was measured again and this was usually 60-70 % of the oxygen content at the beginning. The difference between the two oxygen determinations is the oxygen consumed.

The snails were dried on filter paper, weighed and killed in boiling water; then-shells were then dried and weighed. The difference between the two weights is the live weight of the animals.

The oxygen consumption found in the experiments was calculated in μl. per hr. per individual of a particular live weight, the standard individuals. This is an individual of about the mean of the weight variation of the particular species. This seems more correct than calculating the oxygen consumption per gram, as such a figure may vary according to the size of the animal, large ones often having a lower oxygen consumption per gram than smaller ones.

Every experiment on the oxygen consumption of a standard individual was based upon a series of 5-6 separate determinations of the type described above (cf. fig. 1, p. 46, Berg et al. 1958; and Berg, 1953; Mann, 1956; Lumbye, 1958). The animals selected for experiments might vary from a few to several hundred milligrams, but all individuals placed in the same respiration bottle were carefully selected so as to be of the same size. Several small, or some few medium-sized, or one or two large snails were put in every respiration bottle. Thus often twenty to thirty specimens were used in every experiment.

All the puhnonates which were studied had their lungs filled with water.

Accuracy of measurements. On the basis of several series of experiments of the same type as previously described for the limpet Ancylus fluviatiUs the accuracy of the measurements was computed (Berg et al. 1958, pp. 47-49). It was found that the standard deviation for the respiration of a standard individual is ± 2 %. Thus there is a 95 % probability that a new experiment of the same type would give a result which would at most deviate c. ± 4 % from the average of a series of experiments. A somewhat similar accuracy must be expected of the measurements in this paper. But because it was also found in the experiments with A. fluviatiUs that the cause of the variation of the single determinations is essentially of a physiological nature, not so much an experimental error, it may be supposed that in experiments under the same conditions with other snails the results may vary somewhat more than ± 4 % from the average.

The purpose of these experiments was to see if the oxygen consumption decreased after collection of the animals in nature, owing to partial or total starvation. If so, it might be necessary to pay attention to this in the evaluation of the results of other experiments carried out over a period of several hours after collection. In the case of A. fluviatiUs it was found that starvation markedly influenced respiration (Berg et al. 1958).

The animals were collected and carried to the laboratory in vacuum flasks. They were transferred to respiration bottles, usually in the water from which they were collected, and kept in these bottles in a thermostatic bath at the same temperature (± 1° C.) as that of the locality from which they had been collected. The oxygen consumption was determined at intervals over 8 or more hours and between measurements the water was aerated.

The results are shown in Fig. 1, where the oxygen consumption per five weight of the standard individuals is indicated along the ordinate. The abscissa shows the time elapsed after collection of the animals. According to the different sizes of the species, the live weight of the standard individuals varies from species to species.

Fig. 1 shows that Lymnaea pereger, Myxas glutinosa, Bithynia tentaculata, Valvata piscinaUs and possibly Physa fontinalis and Lymnaea auricularia (only two determinations) show only a very little or no decrease of oxygen consumption in the period investigated. But in Lymnaea palustris and Bithynia leachi a distinct decrease has been found, and this may be taken into account when the relation of respiration to temperature is discussed.

It was found earlier that the oxygen consumption of Ancylus fluviatilis during a gradual increase of temperature from 11° to 18° C. followed Krogh’s curve in the main, but not always (Berg et al. 1958).

The relation of the oxygen consumption to the temperature was studied during the present experiments, the temperature being increased c. 1° per hr. from the temperature at which the animals were collected in nature. In a few cases the oxygen consumption was also investigated during a gradual decrease of the temperature. When a particular change of temperature had been completed, a respiration experiment was carried out at a constant temperature and in the usual way.

The results are shown in Fig. 2 and for comparison a Krogh’s curve drawn on a logarithmic scale is also shown. Besides the observed values of Bithyma leachi, values are given computed after correction for starvation in accordance with the results from Fig. 1. For the other species no correction has been made. (No experiment of this type was carried out with Lymnaea palustris.)

According to Krogh’s curve about a 100% increase of oxygen consumption follows an increase of the temperature from n° to 18° C. For this temperature increase the species investigated show, according to Fig. 2, the following increases of oxygen consumption:

When the temperature changes from 13° to 21° C. Valvata piscinalis shows an increase of 80 % of that shown by the corresponding part of Krogh’s curve.

Thus during gradual increase of temperature the snails increased their oxygen consumption by c. 65-90 % of the increase according to Krogh’s curve. In the case of Myxas glutinosa and Physa fontinalis the increase of rate of respiration must be regarded as nearly the same as that found by Krogh and applying to other animals, because the uncertainty of the two curves must be taken into consideration.

Deviations from Krogh’s curve may be caused by a change of the state of activity in the experimental animals, and here it should be remembered that oxygen consumption was not determined under conditions of basal metabolism.

These results make it possible to compare the oxygen consumptions found at various temperatures, but the differences found between the snail species can hardly at the present moment be connected with particular ecological conditions.

The experiments on the relation of oxygen consumption to the weight of the snails showed that in the same species this relation may differ at various seasons. Examples of such intraspecific variation are shown in Fig. 4. In this figure the regression lines are, however, drawn on a graphical estimate, as described above. This gives satisfactory determinations of the oxygen consumption of the standard individuals. But in order to show that there exists a significant variation in the relation of oxygen consumption to live weight two series of experiments with Lymnaea palustris and L. pereger respectively were computed statistically and compared (Fig. 3).

Lymnaea palustris. The first series of experiments was carried out in June and included three experiments of the usual type, each of which was made up of five determinations of the oxygen consumption of individuals of varying sizes. The equation of the common regression line of these experiments is computed to be log y =—0·2065 + 0·7609 log x, and the slope of the regression line, b = 0·761, has the standard deviation r = ±0·088.

The second series of experiments with L. palustris was carried out in August and included four experiments of the usual type, each of which comprised six determinations. The equation of the common regression line of these experimental results is log y = +0·2421+0·4516 log x, and the slope of the line, b = 0·452, has the standard deviation s = ± 0·070.

The two regression lines are shown in Fig. 3 (unbroken line). The difference between their slopes, 0·761—0·452 = 0·309, is significant (97·5 % < P < 99·0%, where P is the probability), as shown by the above-mentioned standard deviations.

L. pereger. The first series of experiments with this species was carried out in June and included four experiments, each of which was made up of six determinations. The equation of the common regression line of these experiments is found to be logy =—0·4905 +0·9379 log x, and the slope b = 0·938 has the standard deviation s = ±0·036.

The second series of experiments was carried out in August and included five experiments, each of which comprised six determinations. In this case the equation of the common regression line is log y = 0·0807 + 0·5856 log x, and the slope b = 0·586 has the standard deviation s = ±0·105.

The last-mentioned two regression lines are also drawn in Fig. 3 (broken). The difference between their slopes, 0·352, is highly significant (P > 99·95 %), as shown by their standard deviations.

All in all the experiments with L. palustris and L. pereger have shown that the dependence of the oxygen consumption on the size (live weight) of the individuals varies seasonally. Thus the relation, oxygen consumption to body size, is not a fixed, unchangeable quantity characteristic of all species as supposed by Bertalanffy (1957). He thinks that there are three metabolic types. In the first type the metabolic rate is proportional to the surface or the 2/3 power of the weight, in the second type the rate is proportional to the weight itself, and in his third type the metabolic rates are intermediate between proportionality to weight and proportionality to surface area. Bertalanffy thinks that only one of the three possible relations mentioned is characteristic of any one species. After a survey of the available observations he draws the conclusion (loc. cit. p. 220) that in general it can be said that the ‘metabolic type’, i.e. the relation of metabolic rate to body size, is a physiological characteristic of the species or group concerned.

Furthermore Bertalanffy finds that as there are different metabolic types, so also there are different growth types, which are distinguished by the mode of growth as expressed in growth curves. These show the growth rate of the various species, and Bertalanfiy thinks the growth types may be correlated with his three metabolic types. If this were true, then the observed seasonal variation in metabolic type would imply a seasonal variation in the type of growth rate.

The seasonal variation of the relation between oxygen consumption and body size cannot be caused by the inclusion of both juvenile and sexually mature individuals in the experimental animals, since all of them were beyond the juvenile stage. The variation, however, is not in accordance with Bertalanffy’s conception of three metabolic types.

With reference to the observation that species in the course of their ontogenetic development pass through successive stages characterized by regularly changing values of b (= slope of regression line), Zeuthen (1955) has already surmised that there are many more ‘metabolic types’ which should be related to phenomena of growth than the three (surface, weight, or intermediate) suggested by Bertalanffy.

As mentioned above, the slopes b of the regression lines of Lymnaea palustris and L. pereger were greater in June than in August. In Ancylus fluviatilis it was found earlier that the oxygen consumption itself is greater during the period of reproduction than during other seasons (Berg et al. 1958). It is reasonable, therefore, to suppose that the same applies to the two Lymnaea species as well. If this is the case the seasonal variation of b may be caused by a comparatively greater increase during the season of reproduction (June) of the oxygen consumption of the larger individuals with a more vigorous reproduction than of the smaller ones, which, though mature, produce fewer eggs. In August, when reproduction declines, this difference of the oxygen consumption is not so clearly manifested and b therefore decreases.

In Fig. 4 the oxygen consumption at 18° C. of the prosobranchs and pulmonates is shown in relation to the live weight of the species. The regression lines of all experiments are drawn graphically as they are found in experiments similar to those depicted in Fig. 3; the respiration of the standard individuals is shown by means of crosses and circles, respectively. The scale of the top abscissa (of pulmonates) is ten times as large as the scale of the bottom abscissa (of prosobranchs) in order to separate the two groups. The projection of the regression lines on the abscissa indicates the size variation of the experimental animals concerned.

Fig. 4, in addition to new experiments, shows the results of earlier experiments on Acroloxus lacustris (Berg, 1952, fig. 8b corrected to 18° C. from the experimental temperature 16° C.), experiments on Ancylus fluviatilis (Berg, 1953), and also on Theodoxus fluviatilis and Potamopyrgus jenkinsi from fresh and brackish water according to Lumbye (1958), who carried out his experiments in our laboratory and with the same methods. The species were taken from ecologically very different biotopes (cf. p. 690). It will be noted that the live weight of the standard individuals varies from 2-5 to 500 mg.

The following slopes b of the regression lines were found for prosobranchs:

Thus it is seen that some values of b have nearly the value 0·67 required by the surface law, but others are markedly higher, up to 0·95.

According to Krywienczyk (1952 a, cf. Bertalanffy, 1957) the prosobranchs should have an oxygen consumption proportional to the surface, i.e. proportional to w$23$ where w is the weight of the species. Our experiments have shown that in the case of Bithynia leachi, and in Potamopyrgus jenkinsi this may nearly be so. But in other species oxygen consumption may vary much more in relation to weight, being sometimes nearly proportional to weight

In pulmonates the oxygen consumption in relation to weight (Fig. 4, at the top) also varies. The slope of the regression line, b, is found to be in the case of

Thus the oxygen consumption of pulmonates in relation to weight varies from b = c. 0·45 to b = c. 1·00, i.e. between less than proportional to surface and proportional to weight.

According to Bertalanffy (1957) the relation between metabolic rate and body size for pulmonates varies in such a way that the respiration in some cases is proportional to the body surface (according to experiments by Brand, Nolan & Mann, 1948) and in some other cases is intermediate, i.e. proportional to more than 2/3 but less than 3/3 power of the weight. The last-mentioned instances include species of Lymnaea, Planorbis and Isidora according to experiments by Bertalanffy & Muller (1943), Fiisser & Kruger (1951) and Krywienczyk (19526). Thus the experiments reported in this paper are in agreement with the observations of other workers except where they have shown respiration to be proportional to body weight. Experiments with Lymnaea auricularia by Krywienczyk (1952 b) have possibly also shown respiration proportional to body weight.

Fig. 4 also calls for comment on the oxygen consumption of the freshwater snails regarded as a group, a unity. It will be seen that the oxygen consumption of all the standard individuals together is depicted as a belt showing only a slight dispersion. The width of the belt, the dispersion, includes a seasonal variation of some of the species. In spite of this it is characteristic of the belt, formed by the respiration values of the standard individuals, that it is narrow, i.e. the freshwater snails examined have a fairly uniform respiration.

The relation of the oxygen consumption to the size of the freshwater snails as a group seems, according to Fig. 4, to be expressed by a regression line with a slope just below 1·0 and at any rate greater than 0·75. This fact is of course not inconsistent with the above-mentioned result, that most of the species have different values for the constant b. Furthermore, there does not seem to be any difference between the two groups investigated, pulmonates and prosobranchs, with regard to the slopes.

It seems noteworthy that the slope of the regression line of the freshwater snails as a group does not fit in with the common regression line of poikilotherms and homiotherms described by, e.g. Hemmingsen(i95o, p. n) and Zeuthen (1953, p. 3). The difference is supposed to be important for the understanding of the phylogeny of the group in relation to physiology. In this connexion it should be added that the results found by us so far only apply to mature animals, and among these the usual phenomenon of larger species having a lower rate of respiration per unit weight has been observed only to a very small extent. If immature stages were investigated the results might be different.

It may be mentioned incidentally that Pisidium sp. has an oxygen consumption of c. 0·4 μl./hr./individual of 2 mg. at 18° C., and this value falls also within the belt formed by the respiration values of the freshwater snails.

(?) The main purpose of this study was to see whether or not the freshwater snails were able to maintain their oxygen consumption with decreasing oxygen content of the water. If the oxygen consumption falls, it is interesting to see if the fall sets in as soon as the oxygen concentration decreases, or only after the oxygen concentration has reached a low level.

As before, the oxygen consumption here recorded was an active respiratory rate. Concerning this Fry (1957) writes as follows with reference to fishes: ‘Any reduction of the oxygen content below the level where the active metabolic rate begins to be restricted is probably unfavourable to the species concerned. From the ecological point of view this “incipient limiting level” (the critical level under conditions of activity) can be taken as the point where the oxygen content begins to be unsuitable. The level of the beginning of respiratory dependence as an index of water quality was probably first proposed by Lindroth (1940) and formalized by Fry (1947). …’

The series of experiments reported below may serve to demonstrate the incipient limiting or critical level of oxygen supply for freshwater gastropods.

All experiments on the oxygen consumption at a certain oxygen content of the water were carried out in the way described by means of five to six determinations (p. 691). The first experiment of a series began immediately after return to the laboratory with the animals in lake or pond water saturated with atmospheric air. In the following experiments water in equilibrium with nitrogen mixtures containing, e.g. 18·7 %, 16·2 %, 12·8 %, 10·0%, 7·4%, 4·7% and 2·7% of oxygen was used. The experimental time was most often 1 hr., and the experiments were carried out one after another during the same day. During each experiment the oxygen concentration of the water decreased from the above-mentioned oxygen percentages to about two-thirds of these.

In Fig. 5 a series of consecutive experiments on the oxygen consumption in relation to varying oxygen supply is shown; initial and final oxygen concentrations of each experiment are indicated by means of arrows, and the average concentration by means of a cross. A curve through the mean values shows the result. The mean of an initial and final oxygen concentration is accepted as a useful approximation to the concentration of oxygen in which the animals have respired. The curves in Figs. 6 and 7 are drawn in a similar way through the means of the oxygen concentrations, leaving out the initial and final values of concentration.

In Fig. 6 the standard live weights of the species, the experimental temperatures and the season are indicated after the names of the snails.

(c) The results of the series of experiments shown in Fig. 6 give occasion for the following remarks on the various species.

Lymnaea auricularia. The characteristic feature is that the species is nearly able to maintain its oxygen consumption in relation to decreasing oxygen concentration of the water down to about 11%, but at lower concentrations the uptake decreases distinctly.

Myxas glutinosa. The usual oxygen consumption is maintained down to a concentration of about 12% oxygen, but the consumption decreases at lower oxygen concentrations, especially below a content of c. 6 % O₂.

Lymnaea pereger. Immediately after decrease of the oxygen content of the water the oxygen uptake decreases; below a content of c. 8 % of oxygen the decrease of uptake is marked.

L. palustris. The oxygen consumption decreases at once with declining oxygen content, but seems to increase again; below 12-13 % of oxygen content of the water the decrease of uptake is fairly regular.

Bithynia tentaculata. The oxygen consumption decreases distinctly as soon as the oxygen supply declines.

Physa fontinalis. Even if the oxygen consumption first decreases a little the species is able to maintain the normal uptake at a concentration of 13-14% of oxygen; at a lower oxygen concentration the uptake decreases slightly and below a concentration of c. 6 % of oxygen it decreases distinctly.

Valvata piscinalis. It is nearly able to maintain its oxygen consumption with declining oxygen content of the water down to 9-10 %; after that a distinct decrease in uptake is found.

Bithynia leachi. The consumption is maintained, or it has even increased a little, till the oxygen concentration has declined to 13-14 %; but below this concentration the uptake decreases. The respiration at low oxygen percentages, however, is comparatively great, about two-thirds of the oxygen consumption in air-saturated water.

Summing up it may be said:

• A critical point of oxygen supply in the sense of Fry has been found in Lymnaea auricularia (c. n % O₂), Myxas glutinosa (c. 12% O₂), Physa fontinalis (13-14% O₂), Valvata piscinalis (9-10% O₂) and Bithynia leachi (14-15% Og). But the critical point is not very pronounced.

• In some other species oxygen consumption decreases immediately in response to a declining oxygen supply: Lymnaea pereger, L. palustris and Bithynia tenta-culata. Among these species Lymnaea pereger and L. palustris increase uptake once more at low oxygen levels, but not to the initial value. In these cases there seems to be some reaction, perhaps increased activity, to the reduction of oxygen consumption. A similar increase of oxygen consumption following a moderate decrease is also found in Myxas glutinosa and Physa fontinalis, and in these cases the increase is so great that the oxygen consumption attains almost the initial value.

• In some species the decrease in oxygen consumption in response to a decreasing oxygen supply is not gradual, but shows a steep fall below certain values of the oxygen content: For Myxasglutinosa at c. 6 % O₂, for Lymnaea pereger at c. 8 % O₂, and for Physa fontinalis at c. 6 % O₂.

• The only species able to maintain a comparatively high oxygen consumption at low oxygen supply is Bithynia leachi.

In order to make certain comparisons the results shown in Fig. 6 are computed as oxygen consumption per gram (instead of per standard individual) and at 13° C., which is the temperature midway between the experimental temperatures actually used. The values of oxygen consumption calculated in this way are depicted in Fig. 7.

Fig. 7 shows that when the oxygen supply is abundant the two small species Valvata piscinalis and Bithynia leachi have, as might have been expected, a great oxygen consumption. The same applies to Physa fontinalis, which also is fairly small and in addition is usually more active than the other species examined. Furthermore Fig. 7 shows that at a low oxygen content of the water, below c. 4-5 % O₂, Bithynia leachi is the only species able to maintain a great and steady oxygen consumption (c. 75 /xl./g./hr.); among the other species the oxygen uptake is lower under these circumstances and also decreases distinctly in consequence of declining oxygen supply.

The incipient limiting or critical point of oxygen supply is seen very clearly in Fig. 7 for Physa fontinalis (13-14% O₂), Valvata piscinalis (9-10% O₂), Bithynia leachi (14-15 % O₂), Myxas glutinosa (c. 12% O₂) and, less pronounced, Lymnaea auricularia (c. 11% O₂). Altogether the critical points of oxygen supply are found to occur from about 9 to 15 % O₂, that is from just under half to about threequarters of air-saturation.

If an organism is able to maintain its usual oxygen consumption until the oxygen supply falls to a certain low critical value, this must be favourable to the organism. If the oxygen consumption decreases as soon as the oxygen supply diminishes it must be unfavourable, because one or more physiological functions of the organism must then have been depressed. But the latter way of reaction need not be so decisive for the organism that it cannot for that reason exist in a given locality with a more or less bad oxygen supply. Perhaps the species merely does not thrive so well, e.g. its growth, the extent of its egg production, etc. are reduced. Thus the difference between two species, one of which has a critical level (point) of oxygen supply while the other has not, does not necessarily mean such a difference in an ecological respect that it has a decisive influence on their existence in localities poor in oxygen, i. e. on their distribution. There may be other differences, e.g. in food requirements, which are more important for the distribution. No absolute correlation between the observed respiratory characteristics of the freshwater snails and their distribution in nature can therefore be expected. Nor was such a correlation found in the case of the snails investigated.

On the other hand, respiratory curves with or without a critical point no doubt express an essential physiological difference between the species concerned. Under bad respiratory conditions the species having a critical point at a low oxygen supply must be expected to thrive better than the species showing a decreasing oxygen consumption with decreasing supply.

At a low oxygen content of the water the oxygen consumption of all the species investigated, with the exception of Bithynia leacki (p. 704), is reduced very considerably. This physiological quality of B. leachi is probably of importance. In Esrom lake the specimens used were collected both in the littoral zone between vegetation and on gravel, where the water is undoubtedly rich in oxygen, and also in a yachting harbour on a muddy bottom, which smells of hydrogen sulphide and where the water must be very poor in oxygen; there the shells are dark with ferrous sulphide. Under these conditions the snails’ ability to maintain a great oxygen consumption is probably an advantage.

Freshwater molluscs as a whole seem to possess a considerably greater physiological adaptability than, e.g. marine molluscs. The comparatively small differences as to respiration of the freshwater snails examined are probably connected with their ability to live together in ecologically very different places. Nevertheless, for some of them the oxygen conditions in a certain locality may be of decisive importance.

From E. Frbmming, ‘Biologie der mitteleuropaeischen Süsswasserschnecken ‘(1956) it seems that small species do not have a shorter life span than most of the larger ones, i.e. the growth rate of the former must be considerably smaller than that of the latter. In localities where the oxygen content of the water is low during a shorter or longer period small species are especially common, i.e. those which have a small growth rate. Valvata and Bithynia, for instance, occur at fairly great depths in the eutrophic lakes and in other places with bad respiratory conditions, whereas the large species of Lymnaea occur especially near the surface of ponds or at the shores of lakes and rivers. A similar example is represented by the Danish Pisidium species, among which Pisidium amnicum (O. F. Müller) attains by far the greatest size. And just this species seems to be restricted to localities with fairly great water movements, i.e. localities having favourable oxygen conditions. The occurrence of the most suitable food must, of course, also be important for the growth and occurrence of the species.

It was shown earlier that the variation in the oxygen consumption of Ancylus fluviatilis in the course of the year is great. Measured at the same temperature the oxygen consumption is about 1-3 times to nearly twice as great in spring and summer as in autumn and winter. The increase of the oxygen consumption during the reproductive period is regarded as an expression of the sexual activity (Berg et al. J958).

A similar seasonal variation was shown for Acroloxus lacustris (Berg, 1952, fig. 8 b). If the demonstrated oxygen consumption is converted from the experimental tern-perature 16° to 18° C. by means of the relation also shown to exist between the respiration of this limpet and the temperature the following values are found:

May-June (1949): 1·44 μl. oxygen consumption/hr./individual of 5 mg.

September-October (1950): 0·97 μl. oxygen consumption/hr./individual of 5 mg.

Thus the oxygen consumption of A. lacustris in spring and early summer is found to be about 50 % greater than in autumn (the measurements were carried out with the same methods).

Some of the experiments mentioned in this paper indicate also the existence of a seasonal variation in some other species. Thus in the case of Lymnaea pereger the following values were found at 18° C.:

June: 35·2 μl. oxygen consumption/hr./individual of 120 mg.

June: 30·6 μl. oxygen consumption/hr./individual of 120 mg.

August: 25·0 μl. oxygen consumption/hr./individual of 120 mg.

The two values from June are about 40 % and 22 % greater than the value from August. It is therefore reasonable to suppose that there exists a seasonal variation in this species.

In Lymnaea pahistris the following values were found at 18° C:

June: 24·6 μl. oxygen consumption/hr./individual of 100 mg.

August: 17·2 μl. oxygen consumption/hr./individual of too mg.

Here also there is a distinct seasonal variation. The oxygen uptake in June is 43 % greater than in August.

In Bithynia tentaculata at 18°:

July: 7·6 μl. oxygen consumption/hr./individual of 80 mg.

August: 8·4 μl. oxygen consumption/hr./individual of 80 mg.

October: 6·8 μl. oxygen consumption/hr./individual of 80 mg.

The value from July is c. 12 % greater than the value from October, and thus this also may indicate the existence of a seasonal variation.

The experiments on the other snails were carried out in the same season of 1957 and not distributed over several months. Hence it is not possible to say anything about the seasonal variation of the respiration of these species. But on the basis of the above-mentioned examples it must be presumed that a seasonal variation in oxygen uptake is fairly common in Danish freshwater snails, and the variation may be very extensive, at least up to 100 % of the lowest value. Seasonal variation is, therefore, a quality which must be considered in intra-and interspecific comparisons of the physiology of snails.