The Physical Properties of the Isolated Swimbladder in Cyprinidae

The experiments described in previous papers (Alexander 1959 a, b) were done on intact fish. They provide information only about the swimbladder as a whole. This paper will report experiments on isolated swimbladders designed to provide information on the parts of the swimbladder.

Müller (1843) reports that the anterior sac of the swimbladder is more extensible than the posterior, but gives no quantitative data. The method described below has been used to determine the extensibility and size of each of the sacs of the swimbladder. A number of species was examined in an attempt to elucidate the difference in the overall extensibility of the swimbladder reported in Alexander (1959 a).

The apparatus is shown in Fig. 1. The hypodermic needle was of size 20 and was bound into a piece of rubber tubing with copper wire as shown in the inset It was blunted by rubbing its tip on emery paper to reduce the danger of puncturing the swimbladder. The capacity of the capillary was 0·0270 ml./cm. length. This apparatus was suitable for investigating the swimbladders of Cyprinidae weighing 50−120 g.

The experimental procedure was as follows. The fish was killed by a blow on the head, dried with a towel, and weighed. The swimbladder was removed with part of the pneumatic duct, great care being taken to avoid tearing the tunica externa. The hypodermic needle was then passed up the pneumatic duct so that its tip was in the posterior sac, and the duct wall was bound to this cannula with cotton thread.

The flask was filled with Ringer solution (Young, 1933; formula recommended for freshwater teleosts other than eels), the swimbladder on its cannula was inserted, and the bung pushed home. Sometimes a little air remained in the flask, but this was unimportant as the pressure in the flask was constant throughout the experiment ; only the pressure in the swimbladder was altered. Ringer was passed in or out through the funnel to move the meniscus to a convenient position in the capillary.

A series of readings of meniscus position was then taken with the pressure in the swimbladder alternately at 2 and 6 cm. Hg. As the meniscus did not come to rest until about 30 sec. after each change of pressure, readings were taken at 1 min. intervals, the pressure being changed immediately after each reading. The first two readings were usually discarded as they often showed a greater change than subsequent ones, which remained constant within reasonable limits for a series of cycles. The meniscus movements represented changes in the volume of the swimbladder.

The pressure was then returned to 2 cm. Hg and the cannulated swimbladder was taken out of the flask. A cotton thread was tied tightly round the ductus communicans^* to prevent the passage of gas between the two sacs. The swimbladder was returned to the flask and the readings were repeated. Whereas the first series of readings had given the extension of the whole swimbladder, the second, on account of the ligature, gave only that of the posterior sac.

The pressure was returned to 2 cm. Hg and the swimbladder removed from the flask. A half-hitch was tied round the pneumatic duct with a cotton thread, and the duct was slipped off the hypodermic needle while the knot was pulled tight. One or two more knots made the ligature secure.

The isolated swimbladder now had ligatures on the ductus pneumaticus and communicans, and an excess pressure of 2 cm. Hg in each sac. It remained to determine approximately the volume of each sac. This was done using a weighing bottle as a density bottle.

The whole swimbladder was weighed in Ringer in the bottle. It was necessary to insert the stopper slowly and wait some minutes for equilibration on account of the tendency to compress the swimbladder. The ductus communicans was then cut anterior to the ligature and the posterior sac alone weighed with Ringer in the bottle. The bottle was then weighed full of Ringer alone. From these three weighings, and the specific gravity of Ringer, it was possible to calculate the volume of each sac of the swimbladder. The weight of the swimbladder wall in Ringer was found to be negligible.

From these measurements of volume and volume change the extensibility of each sac of the swimbladder over the range 2−6 cm. Hg was calculated. The lower pressure, 2 cm. Hg, was selected so that the results would be readily comparable with the data on extension given in Alexander (1959 a). The higher pressure was chosen to give conveniently large volume changes.

The use of the isolated swimbladder makes this experiment a very rough one. It was designed only to detect gross differences between the two sacs and between species.

There are two quantities which can be calculated either from the results of the experiments described here or from the results of those described in Alexander (1959 a). These are the extensibility of the whole swimbladder over the range 2−6 cm. Hg excess pressure, and the volume of the swimbladder with 2 cm. Hg excess pressure. These data, as obtained from the two sets of experiments, are compared in Table 1.

Agreement between the two methods is very rough. Differences may be attributed to the crude nature of the experiments on isolated swimbladders and to the small number of specimens used. Very considerable stretching of carp swimbladders occurred in these experiments, leading to results very different from those for intact fish.

The experiments on isolated swimbladders were designed to determine the extensibility and size of each of the sacs of the swimbladder. The mean values of these for each species are shown in Table 2; details of individual fish are given in the Appendix. Four swimbladders whose tunicae externae were torn, and one from which it was missing, are included in the means; the extensibilities of the anterior sacs of these are not appreciably different from those of intact anterior sacs of the same species.

From these data a number of conclusions may be drawn. Müller’s report of the difference in extensibility between the sacs is amply confirmed : in most species the anterior sac is about ten times as extensible as the posterior. As the anterior sac alone is involved in the operation of the Weberian ossicles, it is clearly advantageous to limit the extension of the swimbladder to this sac.

The more extensible anterior sac is usually rather smaller than the posterior sac, but in the carp it is very much larger than the posterior sac. The high overall extensibility of the carp swimbladder is due to this peculiarity, while that of the dace appears to be due to the unusual extensibility of a rather small anterior sac. The extensibility of the bream swimbladder only approaches its low asymptotic value at high pressures, and the experiments on isolated swimbladders were conducted over too low a range of pressures to show this exceptional inextensibility.

Guyénot (1909) tied a cannula into the posterior sac of an isolated tench swim-bladder and inflated it. He photographed it at two unspecified excess pressures. He measured his photographs, and reported that the anterior sac was considerably more extensible in length than in breadth. He interpreted this as an adaptation to efficient operation of the Weberian mechanism.

This experiment has been repeated, again using a tench. The tunica externa was not damaged. The excess pressure was increased slowly from zero to 18 cm. Hg over a period of 27 min., and a photograph was taken from a dorsal viewpoint after every cm. Hg increase. The camera was mounted at a fixed height above the swimbladder. The photographic lights were turned off between photographs to avoid drying the specimen.

The negatives were projected in an enlarger at constant magnification and their outlines drawn. In each outline the length of the anterior sac from the ductus communicans to the insertion of the ossicles and the width across the widest part were measured. Both length and width were found to increase linearly with pressure above 3 cm. Hg : below this pressure the increase was more rapid.

When the excess pressure was increased from 3 to 18 cm. Hg the length of the swimbladder increased from 4·80 to 5·47 units—an increase of 14%. For the same increase in pressure the width of the swimbladder increased from 2·93 to 3·32 units—an increase of 13%. There is thus no appreciable difference between the extensibilities in these two directions.

A similar but less numerous set of photographs was taken from a lateral viewpoint No marked difference between the extensibilities of the anterior sac in length and in depth was found.

My failure to find greater extensibility in length than in breadth in the tench anterior sac is difficult to reconcile with Guyénot’s result. His diagram illustrating his measurements, however, provides a clue (Guyénot, 1909, fig. 4). It shows his two measurements of breadth as having been taken at the same absolute distance from the ductus communicans. They should rather have been taken between the same points on the surface of the swimbladder, which extends in length as well as in breadth. My measurements were made at a fixed level on the swimbladder— its broadest, parallel-sided part, about one-quarter of the anterior sac length from its anterior end.

Guyénots diagram is reproduced, with some additions, as Fig. 2. His measurements of breadth were made at bb′, which is 0·24 of the anterior sac length from its posterior end at the lower pressure, but only 0-21 at the higher. The sharp slope of the walls here makes this misplacement of measurements important. I would have measured the swimbladder at ee′(lower pressure) and ff′ (higher presure).

The proportions of Guyénot’s figure agree so well with his published measurements that we may conclude that the figure was traced from the photographs, ff′ is 19% longer than ee′. This increase is not much less than the 24% increase in aa′, and is very much larger than the 9% increase in bb′.

It thus appears that proper treatment of Guyénot’s data leads to a conclusion similar to mine: that there is no marked difference between the extensibilities of the anterior sac in length and in breadth.

Thus one reported adaptation of the ossicle system is confirmed and another denied; it is found that the anterior sac is much more extensible than the posterior, but that it is not significantly more extensible in length than in other directions.

I wish to thank Dr K. E. Machin and Dr G. M. Hughes for much advice, and the Development Commission for financial support.

Details of results