Mucous membrane of the branchi,c.— The gills of fishes are lubricated and defended by a glary viscid secretion : this is the product of the epithelium. This latter, therefore, even on these parts enacts a secernent office. On the plane superficies of the leaflets, the epi thelium constitutes a single stratum, resting immediately on the expanse of the rete mira bile. In this situation, as in all others, the epithelial layer is supported by a limitary mem brane. Froin its extreme attenuation, how ever, it does not easily admit of separate defini tion. Neither cytoblasts, granules, or any other immature particles, mingle with or underlie this layer of adult epithelium. It is difficult to conceive the mechanisin of their renewal. In the interior of its component cells, how ever, the eye clearly distinguishes a nucleus and a few pellucid granules. It has already been proved that these latter parts are almost suppressed in the epithelium which lines the active capillary segments of the true lung. Of the branchial epithelium of fishes it may be said that it unites the glandular to the me chanical function of the pulmonary epithelium, that its office is exclusively mechanical. The halitus of the air-cells is not a secreted pro duct. It is a transpiration.
Viewed with reference to the principles of exosmoses, a layer of amorphous granules, the cytoblasts of future epithelia, interposed between the blood-vessels and the super ficial strata of adult cells, would obviously render the partition to be traversed by the gases engaged in respiration very inconve niently dense. The structure presented by the epithelium of the branchim implies, that in the organs ofaquatic respiration the blood is brought less directly into relation with the external medium than in those of atmospheric breathing.
The cells investing the branchial capillaries are not structureless, hyaline, flattened scales, devoid of nucleus and granules, as though the principle aimed at were merely the me chanical one of thinning to the extremest practicable limit all structures between the blood and the outer medium. They constitute irregularly oval bodies, carrying a nucleus, and provided with a few pellucid molecules. No cilia exist in the gills of fishes. In their fixed condition, as in the cyprinoid families, in their concealed situation, as in the shark tribe, or their exposed and free position, as exhibited in the higher osseous orders,—forms enough diversified, these organs are charac terised alike by the complete and uniform absence of vibratile epithelium.
The Vascular System of the Branchice.— In fishes, the whole force of the heart and bulbous aorta is expended upon the branchial circula tion. The power of the heart is materially reinforced by the resilient structure which composes the parietes of the aortic bulb. This structure is remarkably contractile. By all standard authorities on comparative ana tomy the muscular nature of the walls of the bulbous aorta is admitted without question.
The interior of this " bulb" is strengthened by carnice columme. It is a fact of no common interest that the colour of the muscular struc ture of the ventricle is higher or redder in ground-fishes than in those species of which the habitat is superficial in the water, and which are gifted with the power of active locomotion. These facts are indicative respectively of supe rior and inferior degrees of muscular irritability, and well shown to conspire with other proofs to determine a difference in amount between the respiration of deep and surface-swimming fishes. The valves which guard the proximal and distal portions of the bulbus arteriosus vary in situation, number and size, in different families. The movement of contraction which takes place in this vessel is not instantaneous in duration, like that of the ventricle : it is slow and vermicular. The pressure which is thus exerted upon the column of the blood is continuous. The near proximity of the delicate capillary structures of the branchim to these powerful centres of force, demands the graduated manner in which the aorta re acts upon the stream of the blood. This mode of action also explains the purposes subserved by the " valves," w hich in some instances occupy an advanced situation in the vessel. The abdominal aorta, which results from the confluence of the branchial veins, differs con siderably in structure from the pre-branchial division of the vessel : the parietes of the former resemble those of a vein. Ho pulsa tions are detectible in the abdominal aorta of fishes. The pulsatile movement of the blood, derived from the systole of the ventricle, ceases at the branchial capillaries. In fish, therefore, arterial pulsations exist in no other part of the circulating system than in that liinited segment which intervenes between the branchial network and the cardiac ven tricle. Neither the head nor any other part of the body receives blood directly from the cardiac ventricle. The carotids, the homo logues of the subclavian, the hyo-opercular and orbito-nasal arteries proceed from the abdominal aorta at the point of confluence of the branchial veins. In these vessels, there fore, the blood is arterialised, while its move ment is impulsatile or venous. The pro pulsive agents, under the form of diminutive lymphatic hearts, which Dr. M. Hall*, M. Fohman -I-, and J. Milner I have shown to guard the several points of communication between the absorbent and venous systems, probably renders as great assistance in cir culating the contents of the latter as those of the fornier orders of vessels. In the white bait (Clupea alba of Yarrell) these micro scopic hearts can be most perfectly observed. From a consideration of the preceding traits distinctive of the circulation of fishes, it will be afterwards shown that the laws of aquatic respiration are destined to receive new eluci dation.