The ductus pneumaticus exists in the eel, sturgeon, amia, erythrinus, lepidosteus, lepido siren, polypterus . It is remarkable that in these fishes the vaso-ganglion is not deve loped. " Under all diversities of structure and function the homology of the swim bladder with the lungs is clearly traceable ; and finally, in those orders of fishes which lead more directly to the reptilia —as, for example, the salamandroid ganoidei and pro topteri—those further modifications are super induced by which it becomes also analogous in function to the lungs of the air-breathing amphibia." * The Lungs in the Batrachia.—In the ichthioid arnphibia there exist two long membranous pul monary sacs, extending, like the air-bladder of fishes, far backwards into the cavity of the abdomen, above the other viscera, but freely moveable in the cavity of the peritoneum, and invested with this serous membrane. They consist of smooth plane-walled sacs, and communicate with the pharynx by means of their membranous ducti pneumatici or tra chem. This simple condition of the lungs occurs in a permanent form in the salaman dridm. Each sac is provided with a pulmo nary artery, which runs in a straight course along the outer side of the organ. From this vessel, branches proceed with great regu larity at right angles and at definite distances. From the midpoint of the space between the arterioles, a venule arises to run round the opposite semi-cylinder of the organ into the chief trunk of the pulmonary vein. In con sequence of this regularity in the distribution of the arteries and veins, the true capillary interspaces present a regularity of area. It follows from this arrangement that each drop of blood, in its passage from the extreme artery to the extreme vein, undergoes in every part of the lung the same crila7ii2012. of aeration. It is commonly supeosed, by com parative anatomists, that the simple lungs of the salanzandrida present a perfectly smooth and uniformly plane surface internally, such that every spot participates with equal acti vity in the office of aerating the blood. This, however, is not the case. The septa which, in the case of the frogs and toads, divide the internal superficies into cells, exist in a rudi mentary state, but unquestionably in the lung of the newt. They are indicated by inter secting lines of vibratile cilia. They coincide chiefly with the principal branches of blood vessels. Bundles of elastic fibres also run parallel with the vascular trunks, which confer upon these delicate organs an uncommon amount of elasticity. To the next point in the minute structure of the lungs especial 4' R. Owen, Cat. of Phys. Ser. of Coll. Sur. 4to, 1832-40 ; Muller, J., Vergleichende Anatomie der Myxinoiden ; Abhand. Akad der Wissenschaften zu Berlin, 1834 ; Agassiz, Hist. des Poissons Fos Biles, 1833-45 ; Cuvier et Valenciennes, Histoire Nat. des Poissons 1845 ; De Blainville, Annales des Sciences Naturelles, 1837; Bojanus, Versuch einer Deutung der Knochen in Kopfe der Fische, in Oken's Isis, 1818 ; Yarrell on British Fishes, 8vo, 1836 ; Paley, Nat. Theol. 8vo, ed. 10. 1805 ; Bre scher, Recherches sur l'Organe de l'Ouie des Pois sons, 1838 ; Monro, The Structure and Physiology of Fishes explained and compared with those of Man and other Animals, fol. 1785 ; Scarpa, De Au ditn et Olfactu, 1789; Hunter, Obs. on the Ani mal Economy, Palmer's ed., 1837; Ratke, in Die Physiologie von Burdach, 8vo, i. 1826 ; Allen Thompson, Jameson's Journal, 1830-31 ; Duvernoy, Sur le Mecanisme de la Respiration dans les Pois sons, in Ann. des Sc. Nat. 1839; De la Roche. Obs. sur la Vessie A6rienne des Poissons, Ann. du Mu thin], t. xiv. 1809. , attention is invited. The conclusions after to be drawn will be found opposed to the views of Mr. Rainey. This excellent observer* has affirmed the principle that, on the true breathing portions, or capillary seg ments of the lungs, there literally exists no epithelial lining of any description whatever, the vessels being as literally naked. To this " principle," deliberately enunciated and sup ported by elaborate " proof" by an acute and truthful observer, many anatomists have yielded implicit assent. First, it is here objected that such a " principle" violates directly all the lessons of analogy. Analogy! Is not demon stration better than analogy ? In the science of organised beings, the connected reasoning founded on analogy cannot be despised. The closest scrutiny in individual instances may miss the truth. The manifold illusions of the microscope may readily mislead. Analogy supposes a mass of cumulative evidence. The general law neutralises particular errors. In no instance whatever, either in the vertebrate or invertebrate kingdom, has it been proved, in the course of the present extended inquiry, that the vessels of a real breathing organ can exiNt under a perfectly " naked form." What is true of invertebrate animals as an organic law cannot be untrue of the vertebrate. The gills of fishes are furnished with a very marked epithelial covering. The temporary branchim of the amphibia are clothed with epidermis. The air-bladder is provided with an epithelial lining, the cells of which admit of ready and conclusive demonstration. Why should it not exist in the case of the true pulmonary structures ? No reason can be imagined; but the fact that it does not has been affirmed by Mr. Rainey. Mr. Rainey's observations were instituted upon injected preparations. This is the source of the error into which he has fallen. If the lung of the newt be carefully, but quickly, laid open, co vered, but not pressed, with a thin slip of glass, and examined under the microscope, it will be found that the vibratile cilia have a limited distribution. Under the favourable oppor tunities afforded by such a preparation it is perfectly easy to follow with the eye the continuation of the epithelial cells (c, c, fig. 232.) beyond the limits of the ciliary areas into the true capillary or active breathing segnzents. The ciliated portions of the epi thelium (over the vessels b, b) exhibit a flocculent character, precisely as shown in the preparations of Mr. Rainey ; while the areas immediately adjacent appear smooth or naked. But under the use of a higher power and a steady gaze the polygonal outlines of the epi thelial scales can be distinctly discerned most readily between the islets of parenchyma (d, d). If this covering consist of " basement mem brane," then basement membrane is composed of scaliform parts; but it is not. It is a true and real and unbroken continuation of the tracheal and bronchial mucous membrane. It is only the ciliary appendages to the cells that cease at a certain limit ; the cells them selves continue to invest the whole super ficies of the lungs. It is full of interest also to note that the epithelial scales which cover the capillary areas of the lung of the newt (parts which coincide with air-cells of the mammalian lung) lose not only the external appendages (cilia), but also their internal parts (nucleus and granules). This succes sive reduction leaves nothing but a hyaline involucrum enclosing a pellucid fluid. This is the real structure, supported indeed by a hypothetical basement membrane, by which the capillary areas of the pulmonary organs are invested. It finds a parallel in the trans parent scales which cover the cornea.
In these ribless amphibia the operation of breathing resolves itself into an act of " swal lowing" air. The glottidean chink is em braced by two minute semilunar pieces of cartilages and furnished with muscles for opening and closing the orifice. In the pa rietes of the lungs no trace of muscular fibres can be discovered ; but elastic fibres are present everywhere among the vessels. It is
by the agency of this elastic tissue, aided by the abdominal parietes, that the act of expira tion is performed. The exterior of the lung is lined by peritoneum, the scales of which are much attenuated compared with those of other parts of the same membrane, as those of the internal lining. It is a curious fact that the exterior of the lung should be desti tute of cilia, while they should be present on that of the liver in the newt. They are, how ever, on this last organ, limited to the margin. Nothing is more easy- than to exhibit the living circulation in the lung of the newt. The which forms a thin pavement coating. This surface is destitute of cilia. In the frog, as in internal bore of the vessel viewed by trans mitted light is much greater than the long diameter of the red corpuscles. The meshes (a, a) are mere points. The scene is one thick, rich, surpassingly beautiful network of moving blood.
In the frogs and toads the lungs consist of two large, short, and broad, slightly cancellated shining bags. They are situated on either side of the spine, at the roof of the abdominal cavity. They are remarkably elastic, like those of the newt. They are capable of slowly expelling their contents even after the removal of the abdominal walls, and of draw ing themselves up into little hard balls on either side of the pharynx. They exhibit well the living circulation. The glottidean aperture communicates directly with the interior of the organ. There is, therefore, no trachea. The orifice of the glottis is surrounded by rudimentary cartilages somewhat further de veloped. The mechanism of breathing is the same in the frog and toad, in which, like the newt, the thoracic ribs are wanting, as in the salamanders. The steps of the process are, however, better studied in the frog. The outer surface of the lung in the frog is closely invested with peritoneum, the epithelium of the newt, the edges of the liver are fringed with motile cilia. The pulinonary artery (see art. CIRCULATION), derived from the aorta, pro ceeds along the outer side of the lung. It lies immediately underneath the peritoneal epi thelium. The very reverse course is taken by the large venous trunk on the opposite side. This lies in immediate contact with the internal or mucous surface. By this ar rangement the contact of the blood with the air is prolonged. The contributory branches of the vein course along the free internal edges of the septa bounding the cells. The branches of the artery occupy the opposed fixed borders of the same septa. The flat surfaces, or sides of the cells, being the areas dividing the arteries and veins, are the scenes of the capillary segments. To this rule, of course, the eye, by close scrutiny, may detect rnany exceptions. By this distribution of parts, every spot of the internal superficies is functionally utilised. The ciliary epithelium is limited, in its distribution, to the margins of the cells and the lines of the larger vessels. The true capillary areas whereon alone respi ration actively proceeds are covered only by a hyaline epithelium, the cells of which can only be distinguished by their outlines (fig. 232. c, c). The ciliated tracts, according to the manner already described in the lung of the newt, terminate by abrupt borders. The epi thelial cell only is continued over the capillary areas. There prevails an average uniformity in the dimensions of these areas. Each particle of blood, therefore, in its transit from the artery to the vein across this area, is exposed, for the same period of time, to the influence of the air. In the lung of the frog and toad the septa support two layers of reticulate vessels, one on either side of a fibrous partition. A plane of vessels disposed in such a manner can only receive the influence of the aerating elernent on cme side. This fact constitutes a real anatomical distinction between the lung of a reptile and that of a mammal. In this latter case the partitions of the cells are composed only of a single stratum, both sides of which are exposed to the air. By this simple mechanical provision the amount of the respiratory agency is everywhere doubled. In the structure of the reptilian lung the elastic fibre forms a predominant element. It is a substitute for ribs and other accessory apparatus of breathing. The lungs of the frog, relatively to the cubic capacity of their interior, present a much more extensive active surface than those of the Salaman dridm. Thus the purpose of " the cells" is fulfilled, of multiplying the operative surface. The " septa" project from the sides into the interior of the organ. In this respect they may be likened to the gills of fishes ; for, like the latter, " they penetrate the surround ing medium." The lungs of ophidian rep tiles are generally composed of true unsym metrical, long cylindrical or fusiform sacs, extending from the pharynx far into the cavity of the abdomen, above the other viscera, and surrounded with the serous lining of that cavity. They are capable of con taining a considerable quantity of air, which, when driven out with force, produces the " hiss" peculiar to the serpent. In some ge nera, as the coluber, typhlops and vipera, the lung of one side only is developed ; in others, as the boa and python, the two lungs are sym metrically or equally developed. The lungs, in these families, eonununicate, by means of a long and narrow trachea, surrounded by incomplete cartilaginous " rings," with the back part of the tongue. In all ophidia, the lungs display internally, but only on the anterior and upper parts, an elaborate system of alveoli or cells. more like secondary lungs than air-cells. The posterior two thirds of the internal superficies are almost plane, or devoid of" cells," like the lung of the newt. The alveoli, traced from before backwards, become shallower and shallower, until at length they disappear. It thus appears that the serpent may store up in its lung a con siderable volume of air which, slowly passing out over the vascular air-cells, prevents the carbonic acid, the effete product of the pro cess, from contaminating the whole contents of the organ. Each " alveolus," separately examined by vertical section, is found to communicate by a single opening with the general chamber of the lung. Traced inwards, it divides and subdivides into secondary and tertiary tiers of " alveoli," each cell being isolated by dissepiments of which the struc ture is identical with those of the frog's lung already described. Each cell is a separate cavity. It does not communicate with those adjacent by openings in the septa. These septa are utilised in the outspreading of the vascular rete. Each septum, as in the frog, carries two layers of capillary, blood-vessels, separated from each other and supported by an intermediate stratum of elastic tissue. In the mechanism ot breathing, this tissue enacts an important office. Over the interior of the ophidian lung, like the batrachian, the ciliated epithelium is limitedly distributed. The true capillary areas which chiefly co incide with the flat sides and bottoms of the cells are clothed only with hyaline epithelium.' Everywhere throughout the interior of the lung, along the courses of the larger vessels, the borders of cells, or along lines of thick ened tissue, the phenomenon of ciliary vibra tion may be readily detected. It is thus evi dent that the office of cilia is mechanical, if not to cause determinate currents in the air, at least in the halitus and fluid which, by accumulating, rnay obstruct the respiratory function of the capillary areas.