DEVELOPMENT OF SPONGES FROM DISSOCIATED TISSUE CELLS. 23 



ridges or walls i to 2 mm. high, or arches may be formed which rest upon the substratum 

 only at the ends. 



By the next day concentration, viz, the aggregation of tissue toward certain nodules 

 or bars and the transformation of coarse reticula into compact masses, had gone farther. 

 It is evident that the masses of tissue were too thick to flatten and spread, and thus 

 the opposite tendency, a tendency to separate from the substratum and contract into 

 massive shape, came into activity. In such concentration the edge of a plate or reticular 

 expansion often curls up, as in g, figure 23. 



On August 3 the plasmodial tissue was still in the shape of networks, plates, and 

 masses attached to the cloth. The cloth, which was fastened to wire netting, was now 

 hung out in the live box, to give the plasmodial tissue a chance to metamorphose. Two 

 such pieces of cloth, each about 4 inches in diameter, were hung out. There was probably 

 some unintentional difference in the handling, for on one piece all the tissue died, while 

 on the other much of it had by August 10 metamorphosed completely into incrusting 

 sponges with oscula, canals, etc. 



On August 3 eight compact small masses, some about i mm. in diameter, others 2 

 to 4 mm. long and about i mm. thick, were hung out in bolting-cloth bags. These 

 did not do well. The bags silted up. The sponge masses flattened and spread to 

 some degree over the cloth, but did not metamori)hose. 



Experiment record 11, Augjtst 24, igoj. — Result: In this experiment the attach- 

 ment of the tissue to the substratum was forcibly interfered with. But it sometimes 

 happens that when no such interference has been made the tissue nevertheless contracts 

 into massive aggregates. It may be said in general that in the history of the early 

 formed plasmodial masses a critical moment arrives when the masses will either flatten 

 out and begin metamorphosis or go on contracting and uniting to form massive 

 aggregates. Such massive aggregates will not transform in confinement. The formation 

 of massive aggregates is furthered by strewing the tissue thickly and by breaking the 

 early attachment to the substratum. The attachment to the substratum is strength- 

 ened, I believe, by the use of flat surfaces, such as slides. When the slide or similar 

 body is drained the tissue sinks closer to it and is mechanically somewhat flattened. 

 This aids attachment. 



Microciona tissue was pressed out through bolting cloth into a saucer. Bottom 

 of the saucer (50 mm. in diameter) was covered with the tissue. Saucer was left to 

 stand 30 minutes, by which time the tissue had attached in some degree. Water was 

 now poured off and fresh sea water added. The tissue was then dislodged with pipette 

 from the bottom and concentrated toward the center. Saucer now immersed in a 

 large dish of water. Fusion quickly takes place and bottom becomes covered with a 

 fine reticulum and small masses. 



Local contraction goes on and in some hours the tissue appears in the shape of 

 coarse reticula, perforated membranes, or isolated compact masses (about as in fig. 23). 

 To hasten or insure the formation of small compact masses it is only necessary to cut 

 off a small part of a coarse reticulum or plate. When so freed from the substratum, the 



