216 TRANSACTIONS LIVERPOOL BIOLOGICAL SOCIETY. 



The microscleres are palmate anisochelse, 0"023 mm. in 

 length, which are not arranged in rosettes, and toxa, 0"112 

 by 0"002 mm. Most of the microscleres are found towards 

 the dorsal pole. Sigmata do not seem to occur in the 

 embryo. The embryos are enclosed in a loose membran- 

 ous capsule which apparently is derived from the parent 

 sponge. 



From the foregoing it will be seen that the structure of 

 the embryos of Esperella sordida, B, agrees almost com- 

 pletely with what Ridley and Dendy* state in regard to 

 the Halichondrina generally: "the embryo consists of a 

 sac of small, probably prismatic, ciliated cells, enclosing 

 a central mass of mesodermal (?) tissue containing the 



developing spicules. In some cases, perhaps in all " 



"the ectoderm appears to be absent from one pole at 

 which the mesoderm comes to the surface." But although 

 this description agrees very well with what I found in E. 

 so?xlida, still Ridley and Dendy' s figures show only a 

 very slight similarity with mine. These authors had 

 examined especially the embryos of Esperella biserialis, 

 Esperella mammiformis and Mijxilla nobilis, and they 

 state that similar embryos had been found by Keller in 

 Chalinula fertilis, by Carter in Halichondria simulans and 

 Esperella cegagropila, by Marshall in Beniera filigrana 

 and by Schulze in Euspongia officinalis adriatica. I have 

 only been able to see Carter's! paper in which he figures 

 living embryos with cilia dorsally and laterally and with 

 a circle of flagella near the ventral pole. The embryo of 

 Halichondria simulans is in shape somewhat similar to 

 that of Esperella sordida, but differs from it in its quite 



* Ridley and Dendy, Report on the Monaxonida collected by H. M. S. 

 " Challenger," p. liii. 



t Carter, Annals and Magazine of Natural History, 4th ser., vol xiv., 

 pis. 21 and 22, 





