CARPENTER. — FEEDING REACTIONS OF THE ROSE CORAL. 155 



many copepods, larvoc of various kinds, Protozoa, and other minute ma- 

 rine organisms. Almost immediately the tentacles of the polyp under 

 observation began to turn in various directions. Most of the forms in 

 the plankton were too small to be followed by the eye under the low- 

 power lenses used, but I actually saw one copepod struck by and afhxed 

 to the knobbed end of a tentacle. Many organisms doubtless struck 

 against the oral surface of the polyp. In a very few seconds the stomo- 

 dieum became everted, mesenterial filaments were protruded, the disk 

 sank, and the edge zone commenced to fold in toward the center. The 

 latter movement continued until the tentacles and oral disk were com- 

 pletely roofed over and concealed from sight. In this condition the 

 polyp remained for some time. 



What takes place within the contracted polyp during this period of 

 quiescence cannot be directly observed. But it seems to me highly 

 probable that most of the organisms caught and held by the tenta- 

 cles, or trapped by the overfolded edge zone, do not reach the gastro- 

 coelomic cavity. I believe they are retained in the superficial chamber 

 bounded by the oral disk below and the overfolded edge zone above, and 

 here digested through contact with the extruded mesenterial filaments. 

 The latter are known to be the digestive organs of the Actinozoa. 

 They act upon proteid food by means of a tryptic ferment contained in 

 the secretions of their gland cells, and then ingest or absorb the par- 

 tially or wholly digested material (Jordan, : 07 ; Pratt, : 05). The mesen- 

 terial filaments can easily reach to all parts of the superficial chamber, 

 and so be brought into contact with the food held by the tentacles. 

 The attachment of the organisms to the tentacles appears from experi- 

 ment to be so secure that it is difficult to see how the captured bodies 

 can be discharged except by being digested away. 



The mesenterial filaments are themselves well supplied with nemato- 

 cysts, and these may be useful in securing any free organisms met with 

 in the upper chamber. A few such organisms which are still at liberty 

 to move may make their way into the diminished coelenteron, although 

 the cilia of the everted stomodaeum are no longer in a position to pro- 

 duce inhalent currents ; but it seems probable that the greater part of 

 the plankton captured by the polyp undergoes extra-coelenteric digestion 

 in the way described. 



When, under natural conditions, the polyp is feeding on small organ- 

 isms, it is probable that a number of these must be captured by the 

 tentacles, or come into contact with the oral disk, before a cumulative 

 stimulus is obtained sufficient to cause the contraction of the sphincter 

 and mesenterial muscles, and the extrusion of the mesenterial filaments. 

 A few small carmine grains soaked in meat extract and dropped on the 



