FEEDING REACTIONS IN CORAL POLYPS. 427 



filaments. Afterwards usually the skeleton falls into pieces but 

 not before almost the whole of the internal organs are dissolved. 

 After about three hours the copepod is nearly completely devoid 

 of its soft parts and the more or less disjointed skeletal elements 

 are removed from between the mesenterial filaments. 



This experiment, which can be easily followed under the micro- 

 scope from the beginning till the end, gives a strong evidence for 

 the view that besides intracellular digestion there is in Astrangia a 

 secretion of an enzyme which dissolves the food in an extracellular 

 way. Whilst the external surface of the copepod still is com- 

 pletely intact the greater part of the internal organs are already 

 dissolved and ingested by the cells of the digestive zone of the 

 mesenterial filaments. This disintegration of the soft parts of the 

 copepod, as long as it does not yet fall into pieces, is only possible 

 when a digestive fluid penetrates into it. 



Probably in Astrangia a number of different digestive enzymes 

 may be demonstrated as in the case of other ccelenterates (cf. 

 Bodansky and Rose, 1922). The most effective of these enzymes 

 is undoubtedly one which is comparable to the trypsin of other 

 animals. This we may already expect in advance as the digestive 

 vacuoles during the later period have an alkaline reaction. More- 

 over in all lower animals in which the proteolytic enzymes are 

 studied they have a trypsin-like function (cf. Jordan, 19076). I 

 have made no elaborate experiments on the nature of the enzymes 

 in Astrangia, but the few enzyme preparations tested showed that 

 a trypsin-like enzyme is the chief factor for the disintegration of 

 the food. Owing to the polyps being small the mesenterial fila- 

 ments cannot easily be separated from the other parts of the 

 polyps. The suspensions were made by pounding the tissues 

 (chiefly consisting of the mesenterial filaments, extracted from a 

 number of polyps) with sand to a mash. The latter was diluted 

 with sea water and preserved with a few drops of chloroform. 

 To equal parts of tissue suspension a piece of crab meat (previ- 

 ously boiled to destroy the blood enzymes of the crab it might 

 contain) was added. Tests, containing crab meat in sea water 

 with chloroform, were prepared in the same time. After four 

 days the meat was partially dissolved. No positive results were 

 obtained with the biuret reaction; the reaction with ninhydnn, 



