so as to resemble a frog's lungs; and the walls and partitions of these cells have many blood-vessels. These air-bladders are, in fact, more cellular and more vascular than the lungs of Menobranchus, or the hinder and larger portion of the lungs of serpents. And, in the light of the observations already recorded, there seems good reason for believing that pure air is inhaled and vitiated air exhaled whenever the fish rises to the surface.
It is worth noting, also, that both Amia and Lepidosteus are very tenacious of life, and endure removal from the water for a time much better than do the sturgeons, whose air-bladders are neither cellular nor vascular. The latter, also, are bottom-feeders, while the gars seem to keep near the surface of the water.
Why, then, are not these air-bladders lungs?
The most obvious objection is, that their openings are into the upper or dorsal side of the throat, while the glottis of batrachians, reptiles, birds, quadrupeds, and ourselves, communicates with the lower or ventral side.
This objection may be met in two ways. In the first place, if allowed, we should have to admit that all the so-called air-breathing vertebrates have organs (the lungs) which have no representative in the fishes, and that most of the latter have an organ (the air-bladder) which has no representative in the former.
It is true that some fishes have no air-bladder; but with some, as Amphioxtis, the lamprey-eels, the sharks, and the skates, we may infer that it has not yet become developed; while with others, as the flat fishes, the air-bladder may have been lost through what may be called a local retrograde metamorphosis.
It is important to note, also, that an air-bladder and lungs have never been found in one and the same animal; and since arms, front-legs, flippers, and wings, are all regarded as modifications of the same organs, anterior limbs; and since, in many other cases, organs of very different size, form, complexity, and function, are considered as homologous, we shall be following precedent in admitting a willingness to regard air-bladders and lungs as modifications of the same organ.
But the true argument against the objection is derived from the existence of transition forms, or links, between air-bladders and lungs, as to the position of the organs themselves, and their communications with the alimentary canal.
With Amia and Lepidosteus the air-bladder and the opening of the duct are both dorsal. With the Brazilian fish called Erythrinus (as first stated by Johannes Müller, and lately verified by the writer), the duct opens upon one side of the throat. In the lately-discovered Ceratodus of Australia, as described by Günther, the sac and duct are single, but the former is vascular, and the latter enters at the left
- Maclay has figured a rudimentary air-bladder in certain shark-embryos.