Axcret')!-!/ Pruct^sci< in l'i>nfi'ra. 487 



advaiicei] a sta-^e of " traii^t'orinatioii " as others, and some of 

 these former still retain tlieir ecll-outliiie. 



By selectin;::^ suitable examj)les we can compile a complete 

 series showing the ^railual Iraiist'ormatinn of a choaiioeyte to 

 tilt' ama'boicl coiulitioii. Such a series is represented in tig. 3. 



Fig. 2. — Transverse section of Uvantia, showing two radial chambers and 



nephrocytes. 

 Fig. o. — Series A, 13, C, D, showing transformation of a choanocyte; 



camera, high power. 



References. 



a. Radial chamber, with progi'essiug nc. Nephrocytes. 



transformation. v. Vacuoles aronnd food. 



b. Radial chamber, with amcebiform 



choanocytes. 



Fig. .^ a is a choanocyte (drawn by caraei*a under hioh 

 power) showing the contained carmine particles. This is the 

 flagellated ingestive phase. In h we And that the cell-outline 

 lias become less distinct, and this is especially so at one end, 

 which is the inner end or the end away from the flagellum. 

 This condition becomes more pronounced in the phase drawn 

 at c, and, finally, at d we have the cell in the amoeboid 

 digestive phase, practically unrecognizable from an amoeboid 

 mesodermal cell. 



Concurrently with this change the cell leaves the flagellated 

 chamber-wall and migrates into the mesoderm. What ulti- 



33* 



