DEVELOPMENT OF SPONGES FROM TISSUE CELLS. I269 



easily seen with the unaided eye scattered through the interior or over the sur- 

 face of the remains of the original sponge. They are frequently spheroidal, but 

 often of an irregular shape, and have the power of slow amoeboid movement. 

 In successful cases of treatment these masses, varying from a fraction of a 

 millimeter to a few millimeters in diameter, are exceedingly abundant. The 

 smaller ones of more regular shape at once call to mind the gemmules that are 

 normally formed in such sponges as Spongilla. Experiment shows them to be 

 physiologically like such gemmules in that thev have the power to transform 

 into perfect sponges. To bring about this transformation it was only necessary 

 to remove the regenerative masses to the open water of the harbor at Beaufort, 

 where they were kept in small bolting cloth bags suspended in a floating live 

 box. The sponge especially worked on was a silicious form, a species of 

 Stylotella^ 



The second of the two methods, a description of which may be found in the 

 Journal of Experimental Zoology,* is the more interesting and important. It 

 should be said that the method succeeds best with sponges in which there is a 

 considerable development of homy skeletal fiber. The form especially used in 

 my work has been Microciona prolifcra Verrill, and it has proved practically 

 necessary to use always the large bushy specimens. The procedure is as follows: 



Cut the sponge into small pieces and put them on a square of bolting cloth. 

 Gather the cloth round the sponge fragments in the shape of a bag. Holding 

 the upper end closed with the fingers, compress the bag repeatedly with small 

 dissecting forceps. The bag meantime remains immersed in a little dish of sea 

 water. The sponge cells are squeezed free of the skeleton and are strained 

 through the pores of the bolting cloth. They fall like a fine sediment on the 

 bottom of the dish. Collect the sediment with a small pipette and strew it over 

 glass plates or shells immersed in sea water. The originally separate cells 

 quickly combine with one another, exhibiting amoeboid phenomena. The masses 

 so formed go on fusing with one another through the formation of peripheral 

 pseudopodia, and thus the whole surface of the glass slide (or other object used) 

 may become covered with a network of plasmodial masses and cords, which 

 adhere to it with some firmness. After perhaps half an hour the plate should be 

 lifted from the water and cautiously drained. The sponge plasmodia are thus 

 flattened out somewhat and their attachment to the plate is strengthened. 

 Return the plate at once to a dish of fresh sea water, where it should be left 

 with two or three changes of water for a day. By this time the network of plas- 



1 Otto Maas has independently discovered that the cells ot calcareous sponges under the influence 

 of reagents exhibit a behavior essentially like that above described. See my account in Science, loc. cit. 



b Wilson, H. v.: On some phenomena of coalescence and regeneration in sponges. Journal of 

 Experimental Zoology, vol. v, no. 2, December, 1907. 



