HARG1TT. 



FIG. 6. 



fourth set, although the upper half of the bell was removed, the 

 edges contracted quite rapidly and in 36-48 hours had fused. In 

 this last set the typical bell'shape of the medusa could not be as- 

 sumed because of the large amount of the bell removed. In con- 

 tracting the margin was drawn up and when the edges had fused 



the medusa was flattened and the bell 

 very shallow. Sometimes the center of 

 the bell was depressed (Fig. 6), though 

 thisjdisappeared later. Occasionally the 

 bell was^twisted somewhat, the margin 

 turned under and a tentacle thrust 

 through the hole in the top of the bell. 

 This of course delayed the healing of the 

 wound. In the four sets of this experiment, after the cut edges 

 had fused no scar or sign of the fusion was present. 



By the time the wound in the bell had closed, or very soon 

 after, the free ends of the radial canals had also met and fused. 

 The manubrium developed slowly. In 4880 hours after the 

 canals had fused a gastric pouch had usually formed. This is 

 simply a large pouch into which the canals lead, and at first has 

 no opening to the exterior. Fig. 7, a, shows 

 the gastric pouch before the development 

 of the manubrium. From these pouches 

 the manubria develop. In some cases they 

 had completely formed in four days after 

 the operation. In other cases a week or 

 ten days was necessary for their complete 

 development. Quite common was the for- 

 mation of double or bifurcated manubria. 

 In the four sets of this experiment 21 per 

 cent, developed double manubria. The for- 

 mation of the double manubrium was fol- 

 lowed out in detail in one specimen. Fig. 

 7, a, shows the large gastric pouch as it 

 first appeared, with no mouth opening. A 

 small diverticulum from this pouch was at first taken for the 

 beginning of a new radial canal, but it developed into another 

 pouch as shown in Fig. 7, b. From each of these pouches a 



FIG. 7. 



