CH^TOPODA. 283 



at first lined by hypoblast, the epiblast soon becomes inflected so as to 

 line the oesophageal region. The splanchnic mesoblast of the cephalic 

 region subsequently invests the oesophagus, and some of its cells 

 penetrating between the adjoining epiblast cells give rise to a thick 

 wall for this part of the alimentary tract; the original epiblast cells 

 being reduced to a thin membrane. This mesoblastic wall is sharply 

 separated from the muscular wall outside, which is also formed of 

 splanchnic mesoblast. 



The anus is a late formation. 



Alternations of generations. 



Amongst Chsetopoda a considerable number of forms exhibit the 

 phenomenon of alternations of generations, which in the same general 

 way as in the case of the Ccelenterata, is secondarily caused by 

 budding or fission. 



The process of fission essentially consists in the division of a parent 

 form into two zooids by the formation of a zone of fission between 

 two old rings, which becomes differentiated (1) into an anal zone 

 in front which forms the anal region of the anterior zooid, and (2) 

 into a cephalic zone behind which forms the head and some of the 

 succeeding segments of the posterior zooid. The anal zone is capable, 

 by growth and successive segmentation, of giving rise to an indefinite 

 number of fresh segments. 



In Protula Dysteri, as shewn by Huxley, there is a simple 

 fission into two in the way described. Sexual reproduction does not 

 take place at the same time as reproduction by fission, but both 

 zooids produced are quite similar and multiply sexually. 



In the freshwater forms Nais and Chffitogaster a more or less 

 similar phenomenon takes place. By a continual process of growth 

 in the anal zones, and the formation of fresh zones of fission when- 

 ever four or five segments are added in front of an anal zone, com- 

 plicated chains of adhering zooids are produced, each with a small 

 number of segments. As long as the process of fission continues 

 sexual products are not developed, but eventually the chains break 

 up, the individuals derived from them cease to go on budding, and, 

 after developing a considerably greater number of segments than in 

 the asexual state, reproduce themselves sexually. The forms de- 

 veloped from the ovum then repeat again the phenomenon of budding, 

 etc., and so the cycle is continued 1 . 



The phenomena so far can hardly be described as cases of 

 alternation of generations. The process is however in certain types 

 further differentiated. In Syllis (Quatrefages) fission takes place, 

 the parent form dividing into two, of which only the posterior after 

 its detachment develops sexual organs. The anterior asexual zooid 

 continues to produce fresh sexual zooids by fission. In Myrianida 



1 Keproduction by budding and formation of the sexual products to some extent 

 overlap. 



