238 ALIMENTARY CANAL. 



stitute the permanent glands. The greater part of the original out- 

 growth remains as the unpaired duct of the two glands 1 . 



In the larva observed by Grenadier the anterior pair of salivary 

 glands originated from independent lateral outgrowths of the floor of 

 the mouth, close to the opening of the posterior salivary glands. 



The yolk sack of the Cephalopoda. The yolk, as has already been stated, 

 becomes at an early period completely enclosed in a membrane formed of 

 flattened cells, which constitutes a definite yolk sack. It is, in the more 

 typical forms of Cephalopoda, divided into an external and an internal 

 section, of which the former is probably a special differentiation of the 

 median part of the foot of other cephalophorons Mollusca (vide p. 225). At 

 no period does the yolk sack communicate with the alimentary tract. The 

 two sections of the yolk sack are at first not separated by a constriction. 

 In the second half of embryonic life the condition of the yolk sack under- 

 goes considerable changes. The internal part grows greatly in size at the 

 expense of the external, and the latter diminishes very rapidly and becomes 

 constricted off from the internal part of the sack, with which it remains 

 connected by a narrow vitelline duct. 



The internal yolk sack becomes divided into three sections : a dilated 

 section in the head, a narrow section in the neck, and an enormously 

 developed portion in the mantle region. It is the latter part which 

 mainly grows at the expense of the external yolk sack. It gives off at its 

 dorsal end two lobes, which pass round and embrace the lower part of the 

 oesophagus. The passage of the yolk from the external to the internal 

 yolk sack is probably largely due to the contractions of the former. 



The external yolk sack is not vascular, and probably the absorption of 

 the yolk for the nutrition of the embryo can only take place in the internal 

 yolk sack. The most remarkable feature of the Cephalopod yolk sack is the 

 fact that it lies on the opposite side of the alimentary tract to the yolk cells, 

 which form a rudimentary yolk sack in such Gasteropoda as Nassa and 

 Fusus. In these forms, the yolk sack is at first dorsal, but subsequently is 

 carried by the growth of the liver to the right side. In Cephalopoda on the 

 contrary, the yolk sack is placed on the ventral side of the body. 



What is known of the development of the alimentary tract in the 

 Polyplacophora has already been mentioned. 



In the Lamellibranchiata (Lankester, No. 239), the mesenteron 

 early grows out into two lateral lobes which form the liver, while the 

 part between them forms the stomach. 



In Pisidium the intestine is formed from the original pedicle 

 of invagination, which remains permanently attached to the epiblast. 

 The stomodseum is formed by the usual epiblastic invagination, and 

 becomes the mouth and oesophagus. The development of the crystalline 

 rod and its sack do. not appear to be known. In the adult the sack of 

 the crystalline rod opens into a part of the alimentary tract which ap- 

 pears to belong to the mesenteron. Were however the development 

 to shew them to be really derived from the stomodaBum they might be 

 interpreted as rudiments of the organ which constitutes the odontophore 



1 Ju Loligo only a single pair of salivary glands is present. 



