4. The chief factor in the temporary closure of the originally open esophagus is the change in shape of the esophagus from a tube approxi mately circular in cross-section to a structure of wide rectangular form with at first a slit-like lumen and finally a minute central aperture. The cause of the change in shape, upon which the obliteration of the lumen largely depends, is the combination of growth within the esoph agus in opposition to the denser lateral mesenchymal plates, by the invasion and medial fusion of which the laryngo-tracheal groove becomes converted into a tube and incidentally separated from the esophagus distally. This process is assisted, as concerns the oblitera tion of the esophageal lumen, by the active cell proliferation in the dorsal wall of the esophagus.
5. In the 16-day embryo the atresia of the esophagus extends through about 1,500 microns. Beyond the oral end vacuoles begin to form in the lining epithelium. These represent dilated "intercellular" spaces chiefly within the central syncytial plug of tissue. They in crease in number, and enlarge caudally, where they become confluent. During succeeding stages this process of vacuolization continues, until at the 32-day stage only the extreme oral end of the esophagus remains closed.
6. Both the closure and the reestablishment of the lumen of the embryonic esophagus involve mechanical as well as growth processes, but are normal for a certain stage of the embryonic development. The closure is not largely dependent upon intrinsic cell-division, and the fenestration process involves no tissue degeneration or resorption. The level of initial closure and the level of final perforation are approxi mately the same—namely, the laryngeal level of the esophagus.
7. In the process of vacuolization upon which the opening of the temporarily stenosed esophagus depends, the larger spherical vacuoles are drawn into irregular spaces as if through traction exerted from without. This traction no doubt inheres in the growing and expanding
periphery of the esophagus. The esophagus now has a fenestrated appearance in section; its lumen is spanned by more or less delicate nucleated septa which may anastomose, giving to the whole the appear ance of a wide-meshed syncytium. Ultimately the trabeculle are drawn into the lining epithelium and their nuclei incorporated among the entodermal cells of the mucous lining.
8. The temporary atresia of the esophagus in the Caretta embryo would appear to be a device for the protection of the lung during its development against yolk material from the gut, which material could not be digested, but would interfere with normal development of the lung.
9. This hypothesis can comprehend and correlate conditions in embryos of forms with meroblastic, holoblastic telolecithal, and aleci thal eggs. Where yolk is very abundant, as in the meroblastic eggs of fishes, reptiles, and birds, the atresia is relatively extensive and of longer duration; in amphibia the closure is largely of the nature of a stenosis in which yolk-globules are involved, probably in process of digestion, while their forward progress is delayed by reason of the con etricted lumen. In most mammals and in man such mechanism is functionally superfluous, and consequently absent except in slight and variable degree. As such it may persist or become accentuated, and produce congenital Atreus or stenosis of the esophagus. The relatively frequent embryonic and congenital occlusions of the duodenum just cau dal to the pancreatic anlages may perhaps likewise find their explana tion in terms of a phylogenetic protective device against yolk invasion.