600 SUMMARY OF CURRENT RESEARCHES RELATING TO 



Cnidaria) they are sharply distinguished ; while the Coelenterates may be 

 called Epithelaria, the Sponges are Mesodermalia ; the latter may have, the 

 former never have, a branching canal system. With regard to their tissues, 

 the Mesodermalia have invariably simple ectodermal and endodermal epi- 

 thelia, the cells of which are always flat pavement-cells, and are never 

 converted into muscular, glandular, sexual, or sensitive elements; such 

 elements are in Sponges invariably modified cells of the mesogloea; the 

 Epithelaria, on the other hand, have a mesogloea, the cells of which remain 

 more or less amoeboid, and are not differentiated to any extent. The mus- 

 cular, glandular, sexual, and sensitive cells are produced in the epithelia, 

 sinking below the outer cell-layer with advancing development, and lying 

 on the surface of the mesogloea or supporting lamella. A phylogenetic 

 table and a systematic classification follow. 



With regard to the classification of Sponges, Dr. v. Lendenfeld regards 

 the composition of the skeleton as affording the first basis ; those in which 

 the skeleton is composed chiefly of carbonate of lime form the subclass 

 Calcarea, and those in which it is originally composed of siliceous spicules 

 are the Silicea ; in the former there is the single order of the Calcispongiae, 

 while the latter are divisible into (1) Hexactinellida, where the mesogloea 

 is soft, and the supporting skeleton is often strengthened with siliceous 

 cement ; the spicules are triaxial ; (2) Chondrospongiae, where the mesogloea 

 is hard, the toughness being due to the hardening of the ground-substance ; 

 the spicules are tetraxonial, monaxonial, anaxonial, or absent : in (3) the 

 Cornacuspongise, the mesogloea is soft, the supporting skeleton is strength- 

 ened by spougin cement or exclusively formed of spongin, with or without 

 foreign bodies ; spicules are monaxonial or absent. A systematic table is 

 followed by a key to the recent families of Sponges, and a compendious 

 bibliography completes the paper. 



Relationships of the Porifera.* — Dr. G. C. J. Vosmaer accepts Gray's 

 division of the Sponges into Calcarea and Non-Calcarea ; the latter class is 

 divided into Hyalospongia ( = Hexactinellidse of authors), Spiculispongiae, 

 and Cornacuspongise ; though we do not know of any direct transition from 

 the first to the other two orders, yet there are certain indications among 

 many already known facts, and the author states that he has found many 

 sponges in which there are a few rudimentary six-rayed spicules lying 

 among the normal ones. 



The hypothesis that Sponges are Coelenterates is not accepted ; argu- 

 ments against the degeneration-idea of Marshall are brought forward, and 

 it is urged that the central cavity of sponges is in no way a gastric cavity ; 

 the argument from development is distinctly against the Coelenterate affinity 

 of sponges. 



" If we accept a free-swimming form as the ancestor, and suppose further 

 structures become secreted in certain cells (thereby conferring an advantage 

 in rendering these delicate forms of life less subject to fall a prey to other 

 animals), then we must at the same time believe that in one group cal- 

 careous, and in another siliceous matter was developed. But this new 

 development led to the restriction, nay, finally, even to the complete pre- 

 vention of free movement, and thereby a higher animal development was 

 precluded. Sessile animals must develope in a special direction in order 

 to maintain the struggle for existence. Nutrition and respiration must be 

 assured ; hence, though the degree of development is a low one, yet a well- 

 developed canal-system has been formed." 



* Vosmaer, 'Porifera' in Bronn's Klaesen ii. Ordnungen, ii. pp. 472-81. See 

 Ann. and Mag. Nat. Hist., xix. (1887) pp. 249-60. 



