OBSERVATIONS ON POLYZOA, 99 



tractors) of the great retractors, although neither saw fit- 

 to give them different names. 



The gastric retractors keep the stomach in its proper 

 position in the axis of the tube (PL 7, fig. 4, PL 8. fig. 2, 

 PL 9, figs. 12, 13, PL 14, tig. 1, M). While the polypido 

 is evaginated these muscles move the digestive organs to 

 correspond with the different positions of the lophophore, 

 always maintaining them in the most favorable posture 

 for the performance of their functions, and guarding against 

 any abrupt curvatures that might otherwise occur during 

 the movements of the polypide. It must be held in mind, 

 also, that they are absolutely necessary in the majority of 

 the Phylactoltemata which grow inverted upon the lower 

 surfaces of logs, or other submerged objects. There is 

 no provision whatever in the muscular organization of the 

 alimentary canal enabling it to hold an upright position fa- 

 vorable for the performance of the peristaltic functions. 

 The gastric retractors supply this deficiency, and, also, 

 prevent the stomach when loaded with food, from bearing 

 its whole weight upon the oesophagus and intestine. 

 They, also, send off a few branches to the intestine, 

 but these are rarely seen ; and, although probably pres- 

 ent in all the genera, I have observed them only in Fred- 

 ericella and Cristatella (PL 7, fig. 4, PL 14, fig. 1). 



The part played by the gastric retractors in disposing 

 of the alimentary canal, varies in the different genera. 

 In Fredericella, where the end of the stomach seldom 

 passes the anterior terminations of these muscles, they 

 continue the act of retraction ; but, in Plumatella, they 

 frequently relax again after the stomach reaches their bases, 

 in order that it may pass still further into the ccenoeciurn. 

 In Pectinatella, on the contrary, they do not relax when the 

 invagination becomes more complete, but bind the end of 

 the stomach down between the bases of the retractors. In 

 Cristatella they relax again, as in Fredericella and Plu- 

 matella, permitting the stomach to pass in a dorsal direc- 

 tion some distance beyond their bases. 



The lowest bundles, attached near the origin of the 

 funiculus, were mistaken by Hancock, for reproductive or- 



