After the first half of fœtal life has passed, the yolk sac is removed from the inside of the chorion, and its place is supplied by another vesicular membrane coming from the fœtus, which gradually expanding, entirely encompasses the foetus, and on which the blood is now exposed, in order that it may undergo the necessary respiratory changes. The sac of the yolk still remains, however, covered by its network of vessels, which probably serve to absorb the substance of the yolk for the nourishment of the fœtus, or being now less exposed to the air, perform some subordinate part in respiration (Fig. 17, x y).

The vesicular membrane which supplies the place of the yolk sac as a respiratory organ, is the same part in a more expanded condition, which becomes developed in the batrachian tribes at the commencement of their aërial existence. According to the observations of the authors already quoted, the allantois of the Lacerta agilis appears, at the earliest period at which it has been seen, in the form of a small vesicle hanging from the lower part of the abdomen, projecting like a process from the cloacal part of the intestine, and resembling in its form and position the urinary bladder (See Fig. 19, Part I.). At this time the allantois is covered by a network of minute vessels, which become more obvious as the vesicle expands. Continuing to enlarge, it insinuates itself between the fœtus and the covering of the ovum (See Diagram, Fig. 18), and its farther progress outwards being impeded, it expands laterally till it envelopes the fœtus in its amnios, and the yolk, with a double layer of a vascular membrane (Fig. 17, zv). The vessels distributed on this membrane are the umbilical arteries and veins; the outermost layer is the most vascular, and is applied close to the inner side of the envelope of the egg, where it receives the full influence of the air transmitted to it from without. The allantois continues to act as a respiratory organ till the foetus breaks the shell, or tears the covering of the egg, and comes out fitted for receiving air into its lungs, and respiring solely by these organs. The greater part of the allantois remains in the egg, in the Lacerta agilis, and the urinary bladder of the adu

formed by a part at its root separated from it by

ch side of which the umbilical vessels previgain the expanded part of the vesicle.

The respiration, therefore, of the foetus of the lizard, as well as that of all oviparous animals higher in the scale, is entirely performed by the two membranes to which we have just alluded, viz. the sac of the yolk and the prolongation of the urinary bladder or allantois *.

Many lizards and serpents, however, are not truly oviparous, but retain their ova till the development of the foetus has proceeded some way, in general till the allantois becomes sufficiently expanded to be fitted to carry on respiration; and some serpents, such as the Coluber berus, are almost entirely ovo-viviparous. In this last it is curious to observe that the arterialization of the foetal blood is effected, though by a simpler apparatus, in nearly the same manner as that of mammalia, or truly viviparous animals. The allantois of the viper, after expanding so as to enclose the fœtus and yolk, comes into contact with a vascular lining of the oviduct, and is closely united with it, so that the venous blood of the fœtus is exposed to the influence of the oxygenized arterial blood of the parent.

It is an interesting fact also, that in some of the Testudines, the allantois, or at least a part of it, remains permanently in the adult, and that the umbilical vein continues as in the batrachia to carry off its blood to the liver. In the Testudo orbicularis, according to Townson, water is introduced into this sac, and it appears not improbable that the large urinary bladder, or permanent allantois of the turtles and of some serpents, serves as an auxiliary in the function of respiration during the whole of life †.

Although respiration by means of gills is rendered unnecessary, by the perfect state of the allantois in these reptiles, yet it appears, from some late observations, that at a period of the development of these, as well as of all the more perfect animals which have been examined, corresponding with that at which the branchiæ of fishes and batrachia begin to be formed, the

• The connexion of the sac of the yolk with the intestine has been demonstrated also in the Coluber natrix by Bojanus, Journal de Physique, 1829; and Dutrochet has shown very clearly the mode of development of the allantoid in the viper, in the Memoires de la Societé Medicale d'Emulation, tom. viii. Several preparations in the College of Surgeons' Museum, London, illustrate these facts extremely well, and some the sac of the yolk of the turtle.

+ See Carus' Compar. Anat., vol. ii. p. 249.

existence of gills in a rudimentary state, is indicated by the structure of the pharynx and surrounding parts. In the early periods of development, the distribution of the arteries especially, which take their origin from the bulb of the aorta, bears a striking resemblance to the primitive simple state in which these vessels have been observed to exist in fishes and batrachia before their gills are formed. In the embryo of the Lacerta agilis, while the cavities of the heart are yet single, Baer has observed that the aorta is divided into five pairs of vascular arches, which, after winding round the sides of the pharynx, reunite with one another above it to form the descending aorta. Baer and Rathke have also observed, that while these subdivisions of the aortawhich may very properly be called branchial arteries-exist, the sides of the pharynx are penetrated by transverse fissures, on each side of which runs one of the vessels rising from the bulb of the aorta. The same appearances have been observed by Baer in the foetus of the Coluber natrix, at a corresponding period of its development.

In these animals, however, the transverse plates of the neck, and the branchial arteries passing along them, do not undergo farther subdivision, or become more fully developed, like the gills of fishes or batrachia; on the contrary, the plates are gradually incorporated with the parietes of the pharynx, the apertures are entirely closed, and the vascular arches are converted, by the obliteration of some parts and the enlargement of others, into the arteries, which, in the adult animal, take their origin from the right and left ventricles or cells of the heart.

Unfortunately, the manner in which the farther development of these parts takes place has not been observed, owing to the difficulty of procuring the ova of the saurian and ophidian reptiles at many different stages of fœtal life. From the many varieties in the distribution of the arteries rising from the heart in these reptiles, it appears probable that this would prove a very interesting subject for investigation; and it may be remarked, that no animals are better suited for observation, from the length of time that the blood continues to circulate in the foetus after it has been taken from the ovum.

Two of the branchial arches appear to remain permanent, in order to form the right and left roots of the aorta in the adult

lizard and tortoise. The anterior parts only of three other pairs of branchial arches also remain to convey blood into the carotid, subclavian and pulmonary arteries (See Fig. 19).

The development of the lungs has been little attended to in the higher reptiles. In lizards and tortoises the lungs are double, and the pulmonary arteries are derived from the pulmonary arch on each side (See Fig. 19). In serpents, on the other hand, the lungs, as well as the pulmonary artery, are generally single; but in the Anguis fragilis, in the adult of which the lungs are double, Rathke has observed, that in the foetus they are at first single; they consist of a mass of dense gelatinous substance, in which a cavity is gradually formed; in this animal the left is gradually separated from the right lung in the progress of development, and the cavities of the two lungs continue to communicate freely with one another for some time.

Development of the Respiratory Organs in Birds.

For the development of the ova of birds, the application of external heat, as well as the direct agency of air, is required,.. and a respiratory change of the blood of the fœtus, comparatively as extensive as that in the adult bird, seems to be necessary. This change is produced by means of the same membranes, viz. the sac of the yolk and allantois, as those on which the blood of the fœtal lizard is exposed, and though these parts are considerably more developed in birds, their relations, as well as the mode in which they are supplied with vessels, are nearly the same as in the higher reptiles. But the facts relative to the structure of these membranes appear to be so well known, as to render it unnecessary for me to enter into any detailed account of them. While the respiration of the fœtal bird is almost entirely carried on by the membranes of the yolk and allantois, the structure of the parietes of the neck and pharynx, as well as of the adjoining arteries, exhibits certain traces of branchiæ similar to those already alluded to in the embryo of the lizard. These appearances were first discovered by Rathke, and an account of them was published by him in the Isis, in 1825; it is since this time that the observations of this author have been extended, and the same appearances discovered in the lizards, serpents, and mammalia by himself, and by Huschke, Baer and others; by

which a series of analogies of the most interesting nature has been established between air-breathing animals, and those which are aquatic during the whole or some part of their existence *.

From the manner in which the rudimentary intestinal canal is formed, it has been shewn that the anterior and posterior parts of the intestine constitute at first two shut sacs, into which the only entrance is by a large opening in the middle between them, by which they communicate with the sac of the yolk. Neither mouth nor anus, therefore, at first exist; but both these apertures are afterwards formed by a wasting away or absorption of the substance of the germinal membrane, at the two extremities of the rudimentary intestine. The opening of the mouth (Fig. 21, e) appears towards the end of the second day of incubation, some days before that of the anus is perforated. The mouth, or anterior opening into the intestinal canal +, has at first the appearance of a transverse slit, or cleft, on the lower part of the head. On the third day, the part of the intestine into which this opening leads becomes wider anteriorly, and assumes the form of a cone, the apex of which is directed towards the tail of the embryo. The walls of this cavity, which corresponds in many respects with the branchial cavity of cartilaginous fishes, and with part of what is afterwards converted into the pharynx of the bird, become thicker and of a firmer consistence at the same period. Towards the end of the third day six clefts, or transverse slits, make their appearance behind the mouth, three on each side of the intestine. The foremost pair of these clefts appears first, and the second and third after it gradually. The wall of the pharyngeal cavity projects slightly at the parts between the clefts on each side; it is here of a firmer consistence than elsewhere, and has the appearance of being formed of transverse bands, united anteriorly on the mesial line, like the branchial

See the Memoirs of Rathke in the Répert. Génér. d'Anat. et de Physiol., tom. vii, în the Edinburgh Medical and Surgical Journal, 1830; and in the Isis, 1823, No. 6, and 1828, No. 1; those of Huschke in the Isis, 1827, No. 1, ard 18.8 No 2; and those of Baer in Meckel's Archiv. vol. ii., No. 4; and in Breschet's Repertoire, 1829; also Baer, de ovi mammalium et hominis ergosi Epistola.

✦ The opening alluded to can scarcely with correctness be called the mouth at this period, this cavity being afterwards formed before it, by the gods of the superior and inferior maxillæ.