Respiration of the Amphibious Fishes Periophthalmus Cantonensis and Boleophthalmus Chinensis in Water and On Land

Measurements of the oxygen uptake from water and air have been reported for the amphibious fishes Periophthalmus sobrinus and Sicyaces sanguineus by Gordon et al. (1968), Anabas testudinius by Hughes & Singh (1970 a, b), Ciarias bairachus by Singh & Hughes (1971), Saccobranchus fossilis by Hughes & Singh (1971) and Amia calva by Johansen, Hanson & Lenfant (1970). Recent reviews are given by Graham (1976) and Singh (1976). The partitioning of the oxygen uptake by specimens which were freely allowed to breathe from water and air was determined in the studies with C. batrachus and S. fossilis, but the relative amounts of oxygen taken up by the gill and skin of amphibious fishes in water or air have not been compared.

In this paper, the oxygen uptake and relative roles of the skin and gills in respiration of two Japanese amphibious fishes, Periophthalmus cantonensis (Osbeck) and Boleo-phthabnus chinensis (Osbeck), in aquatic and terrestrial habitats were studied. A comparison was made between specimens of the two species with respect to oxygen uptake by the gill or the skin alone. Adaptations of the two species to terrestrial life were also studied in relation to their respiration.

Ranges in body weight and length of the specimens used are given in Table 1. They were sampled from the estuary of the Honmyo River along Ariake Bay. The amphibious fishes were kept in a circulation containing 50 % sea water (specific gravity 1·011-1·013). The fishes were acclimated for more than one day at the same temperature as that used in the experiment. Two to four individuals of the amphibious fishes were used for every experiment, but one individual each for estimations of oxygen uptake by the gill or skin alone. Oxygen consumption was estimated either (a) by a revised version of Morooka’s method (Tamura & Morooka, 1973) in open aquaria of 2 1 capacity beginning measurements when the water was almost fully saturated with dissolved oxygen, or (b) directly from the decrease in dissolved oxygen content in a closed vessel. The oxygen uptake from air or water alone was measured in a closed perspex cylinder in which the head and body regions of the fish were separated by a thin rubber septum. From measurements of the rate of oxygen uptake under three conditions (Fig. 1 A, B, C), it was possible to calculate values for the uptake by the gills and skin (Fig. 1). Cutaneous respiration was also measured by sealing the mouth and the gill openings using an adhesive paste, Aron Alpha. Oxygen uptake at each of the respiratory surfaces in fish that were free to breathe either water or air was measured in a closed two litre vessel containing 1·5 1 of water and 0·5 1 of air at atmospheric pressure (Fig. 2). In this experiment the estimation of oxygen consumption in air was corrected for the volume of oxygen dissolved during the experiment. Since the time for measurement was always 1 h, the carbon dioxide produced by the fish during this period was considered to have a negligible effect on the oxygen measurements. The experiments were carried out at 20 ± 1 °C. For the calculation of Q₁₀, the oxygen consumption was measured at 5 °C intervals for a specimen acclimated to each temperature for 1 day.

Routine oxygen consumption of fish confined to water increased geometrically with ambient temperature, whereas when they were confined to air the increase was logarithmic. In both species the opercular movements stopped in water at 5 °C but they started again at 10 °C in P. cantonensis and at 15 °C in B. chinensis (Fig. 3) (Table 2)

The oxygen consumption, estimated by the revised Morooka’s method, of both species suddenly doubled following an increase in light intensity during a period of 15-30 min, but it gradually returned to the initial value after a period of about 1 h (Fig. 4).

At temperatures of 20 °C and above, the oxygen uptake of both species was greater in water than in air, when fish were confined to each habitat : in Periophthalmus the oxygen uptake from air was about 60% of that from water and in Boleophthalmus about 70 % at 20 °C (Fig. 5). When these two species were able to select freely between air-breathing and water-breathing it was found that Periophthalmus took 66 % of its oxygen from water whereas Boleophthalmus obtained 70 % of its total oxygen requirement from water (Table 3). Thus under these conditions both species depend more on water than on air. The total uptake was 236 110 ml/kg.hr in Periophthalmus and no ml/kg.hr in Boleophthalmus at 20 ± 1 °C (Fig. 2, Table 3). When the latter was restricted to water-or air-breathing, consumption was less than that in the freely-selected conditions, namely 65 % when confined to water and 43 % when confined to air (Fig. 1).

In Periophthalmus confined to water, the gills and skin were used to an almost equal extent: 52 ± 12 % of oxygen uptake at the gills and 48 ± 12 % at the skin. In air the corresponding figures were 27±7% and 76 + 10% respectively. For Boleophthalmu the corresponding figures were slightly different namely 59±17% and 36±8% in water, and 52! 13% and 43 +11 % in air. If the oxygen uptake of each organ in animals restricted to water breathing is taken as a standard, the proportion entering through the gill of Periophthalmus in air is 31 % of that in water. Thus the ability of the gills to take up oxygen from the air seems remarkably small. On the contrary, the oxygen uptake via the skin in air is 96 % of that in water so that the skin is utilized to almost the same extent under both conditions. On the other hand, for Boleophthalmus, oxygen uptake via the gill in air is 58 % of that in water, and the proportion entering via the skin in air is 78 % of that in water. Thus in this species the ability of the skin for oxygen uptake in air is smaller, and that of the gill for oxygen uptake in air is greater, than in Periophthalmus.

For Periophthalmus in air, the uptake of oxygen per unit area of skin was almost the same (96%) as that in water, and for Boleophthalmus, the oxygen uptake per unit area of skin in air was 78 % of that in water (Table 4).

Individuals of Periophthalmus at Ariake Bay begin to move onto the dried mud of the estuaries from March onwards (Kobayashi et al. 1971), the dominant individuals leave the water for the land at temperatures greater than 18 °C. Corresponding specimens of Boleophthalmus start to move onto the dried mud from May onwards and the dominant individuals migrate to land at temperatures above 22 °C (Tamura, in press). They become active for feeding on land at these temperatures. Since the temperature variations are rapid and radiant heat is severe in air, these fishes often dive into puddles during the day time.

The effect of light is probably caused by changes in intensity and not the intensity itself. Effects due to the increase in body temperature caused by the radiant heat of sunlight are probably not involved in the results shown in Fig. 4.

In measurements of oxygen uptake in closed or open vessels, the activity of both species differs in relation to temperature. Periophthalmus is more active than Boleophthalmus at 20 °C, the former move their pectoral or caudal fin several times or change the body position a little, and in the dark the activity is routine but not very marked. On the contrary Boleophthalmus hardly move at all and activity is near to standard metabolism. Therefore the difference between the amount of oxygen uptake of both species when showing the same activity may not be so great as indicated by Tables 3 and 5. The oxygen uptake of Japanese amphibious fishes confined to water was greater than when they were restricted to air breathing. In freely selected habitats, the oxygen uptake was also greater from water than from air in both species. Therefore with respect to their respiration these amphibious species appear to be less well adapted to a terrestrial habit than other species of tropical amphibious fishes that have been previously investigated (Tables 5, 6).

The respiratory organs of amphibious fishes show a number of functional changes including reduction in gill area, a change in the number of gill filaments and total filament length, reduction in size of the gill opening and the degree of vascular development of blood capillaries of the suprabranchial wall or skin etc. (Schöttle, 1931; Carter, 1957; Stebbins & Kalk, 1961; Tamura, in press). A number of authors have discussed the relative roles of the gill and skin in oxygen uptake, for example, Hughes & Singh (1971) reported that in Ciarias batrachus and Saccobranchus fossilis confined to water breathing, 83-9 and 83-0 % of the oxygen was absorbed through the gill and i6-6 and 17% through the skin. Berg & Steen (1965) found that for Anguilla vulgaris in water the shares of the gill and skin in the total oxygen uptake were 85-90 and 15-10% respectively. Thus for these three species the gills are responsible for much greater proportions of the total oxygen uptake than in the Japanese species. Teal & Carey (1967) determined the proportions of the total oxygen uptake through the gill and skin in both habitats for Periophthalmus sobrinus and obtained values of 45%.’55% in water and 37%: 63% in air, relationships which are close to those found for Periophthalmus cantonensis in the present study.

As mentioned above, Periophthalmus cantonensis showed greater cutaneous respiration in air per unit of body weight and surface area than did Boleophthalmus, therefore it is suggested that the former is more adapted to air breathing. In a histological study, Matsuo (unpublished, Table 7) has shown that the water-to-blood distance for the skin of Periophthalmus is 9 µm on average, whereas for Boleophthalmus it averaged 17 µm. Thus the respiratory efficiency for these two species would differ for fish of the same surface area. The observation that oxygen uptake is reduced in air should be confirmed by further experiment, especially in high humidity and temperature, because the oxygen uptake by skin in air may be largely determined by the ambient humidity and temperature.

Some indication of the terrestrial adaptation of these two species is given by changes in the proportion of the total oxygen uptake which enters through the gill when the fish is in water or on land. For Periophthalmus, gill oxygen uptake when confined to air decreased by 48% (52-27%) (Fig. 6) relative to that when it was confined to water; whereas in Boleophthalmus, gill oxygen uptake decreased by only 12% (5952 %) relative to that in water. Accordingly, the gill of Boleophthalmus can be regarded as being more adapted to a terrestrial life than that of Periophthalmus. The ratios of the gill area to body surface area were 0-38 in Periophthalmus and 0-56 in Boleophthalmus (Tamura, in press). These results showing a relationship between the size of the relative gill dimensions and the amount of oxygen uptake via gills in air can account for the terrestrial adaptation of Boleophthalmus. Further, adaptation of the gills in such fish usually involves an increase in stoutness of the gill as a whole which restricts their collapse in air. The gills are usually thicker and secondary lamellae are farther apart. In fact, in Boleophthalmus the secondary lamellae are more widely spaced than in Periophthalmus (Tamura, in press). It may be concluded that these differences in respiratory function are associated with an adaptation to air-breathing in the gill of Boleophthalmus, whereas the skin of Periophthalmus is more adapted to terrestrial life.

This work was supported in part by a Japanese Government Subsidy for Aiding Scientific Research. We are grateful to Dr Morooka for his help in our study, and especially to Professor Hughes for his advice and comments on this paper.