MUSEUM OF COMPARATIVE ZOOLOGY. Ill 



nation and remains always in communication with the surrounding 

 medium. In Cristatella the lumen is formed in the bud at the time 

 when its diameter perpendicular to the roof of the colony slightly ex- 

 ceeds that parallel to it. As Figure 14 (Plate II.) shows, this cavity 

 {lu. gm.) first makes its appearance in the distal part of the central 

 mass of cells. There are always cells lying above the lumen, and thus 

 cutting off the ectoderm from contact with it. The two layers from 

 which, according to my view, all of the cells of the adult polypide are 

 derived, are now completely established ; and the cavity has already 

 appeared which, by enlargement, out-pocketing, and the concrescence of 

 its walls, gives rise to the atrium, and the lumina of the alimentary 

 tract and supra-cesophageal ganglion. 



The bud elongates, and often at this time, preparatory to giving rise to 

 a new bud from its upper marginal angle, becomes bent or curved, the 

 concavity always being next the daughter bud (see Figs. 3, 5, 11, and 

 22). By this change in form the bud becomes bilaterally symmetrical. 



2. Origin of the Alimentary Tract. — The first organ derived from the 

 two-layered sac is the alimentary tract. Nitsche ('75, p. 356) described 

 the process of its formation in a very clear manner ; but I believe he is 

 in error. The original lumen of the bud represents, he says, the atrium 

 and the lumen of the alimentary tract. The part lying nearest the 

 attached end of the bud gives rise to the former ; the latter is derived 

 from the lower part of the lumen. These two regions become separated 

 by the invagination of the two layers of the bud along a furrow on each 

 side of the bud ; just as though the walls of a two-layered hollow rubber 

 ball were pressed together by a finger of each hand acting at opposite 

 sides until the points of the fingers should be separated by the four 

 layers of the ball only. By this process, mouth, anus, and the entire 

 gut, would of course be formed at one time. Reinhard ('80% p. 212) 

 appears to agree with Nitsche as to the method of origin of the alimen- 

 tary tract. Braem ('89 b , pp. 677, 678) describes and figures diagram- 

 matically this process in the statoblast of Cristatella. In the median 

 plane of the bud there is an out-pocketing from the anal side of the 

 atrium which involves both layers of the bud ; it assumes the form of a 

 comma, its blind end curving forward to meet the blind end of a lesser 

 evagination of the oral part of the atrium, — the oesophagus. The blind 

 ends of these two pockets meet, and by the breaking through of the 

 intervening tissue their lumina freely communicate with each other, 

 thus completing the alimentary tract. He adds : " Audi bei den Poly- 

 piden der fertigen Colonie der Darm durch Verwachsung eines analen 



