THE EMBRYOLOGY OF THE OPOSSUM 161 



fourth and fifth gill pouches are also detached and considera- 

 ble migration has occurred. As there is no histological differ- 

 entiation between the derivatives of the different pouches at 

 this time, identification becomes complicated. This much can 

 be established with certainty: 1) the thyroid has migrated 

 caudad until it is just ventral to the ultimobranchials (fifth 

 pair) ; and 2) the third and fourth pairs, including both dorsal 

 and ventral components, have migrated further caudad. The 

 fact that the dorsal (parathyroid) elements of the third and 

 fourth descend with the ventral (thymus) elements into the 

 mediastinum is clear. The only point about which there is 

 any uncertainty is which are the third derivatives and which 

 are the fourth. As both sets have a similar history the 

 question is not important. If, as in other mammals, the third 

 gill pouches migrate most caudad of all, then the four epi- 

 thelial bodies caudal to the aortic arches and just cranial to 

 the pericardium are probably the two ventral and the two 

 dorsal elements of the third pair. Correspondingly, the four 

 units situated around the common carotid arteries just an- 

 terior to the fourth aortic arches are probably the dorsal and 

 ventral elements of the fourth pair. At any rate, in later 

 stages all eight derivatives of the third and fourth branchial 

 pouches come together in the mediastinum to form the thymus 

 and the parathyroids. 



It has been known for some time that in the adult opossum 

 no parathyroids can be found in or near the thyroids. This is 

 due to the fact just mentioned, that all four parathyroids 

 become incorporated into the thymus in the mediastinum. 



During stage 33 the ultimobranchials come into contact with 

 the dorsal walls of the wings of the thyroid (17151) and fuse 

 with the latter just as the isthmus of the thyroid disappears. 

 These are the so-called 'lateral thyroids' of older authors. 

 Whether they contribute any real thyroid tissue or not, I am 

 unable to say, as they seem to become completely lost in the 

 gland before differentiation of colloid vesicles occurs. 



The allantois at its height. The principal facts in the de- 

 velopment of the allantois have been described in connection 



