276 RECORD OF CURRENT RESEARCHES RELATING TO 



either ojigtis directly into an efferent canal, or first unites with the 

 ducts from several contiguous chambers, a racemose arrangement being 

 thus produced. 



A layer of flat epithelial cells can be made out, by the employment 

 of silver nitrate, lining both afferent and efferent canals. On the 

 surface of the sponge such a layer is more difficult to demonstrate, 

 and here its cells are often covered with a delicate cuticle, as in 

 Euspongia and Cacospongia* 



The connective layer has also, in all essential respects, the same 

 characters as in Euspongia ; the contractile fibre cells are also similar. 

 A special accumulation of connective tissue takes place round the 

 developing egg, producing a capsialar investment. 



The skeleton resembles that of Cacospongia scalaris. The primary 

 radial fibres include foreign bodies, are usually terete, and are fre- 

 quently perforated so as to present the ajjpearance of an irregular 

 network. The secondary connecting fibres may or may not contain 

 sand-grains, siliceous spicules, &c. Although no definite layer of 

 sj)ongoblasts could be made out, Schulze has no doubt that the fibres 

 are cuticular structures. The collar cells differ in no respect from 

 those of Euspongia and Cacospongia. 



Eggs were found in all stages of development, from irregular 

 rounded cells resembling the amoeboid corpuscles of the connective 

 layer, to large ripe eggs, quite opaque from the great number of 

 highly refracting yolk-granules contained in them. All stages of 

 cleavage were also observed, but no sperm aggregations were met with 

 in this species. Probably they are found only in a small number of 

 individuals. 



The author confirms Oscar Schmidt's discovery of the skipping-rope 

 shape of the fibrill^ef so characteristic of this genus. They are 4-8 

 mm. long, 6 /a thick in the middle, and 3 ;«, at the ends ; the 

 globular or pear-shaped knobs which terminate them are 6-10 fx in 

 diameter. For isolating them the sponges were macerated for some 

 weeks in water mixed with ammonia, washed thoroughly with a stream 

 of water, cut into pieces almost the size of a walnut, and placed in 

 hydrochloric acid 10 per cent, for some days, and then washed in 

 running water for some hours. By this means cobweb-like balls of 

 intertwined filaments are pressed out from the cut surfaces. These 

 are then placed in a watch-glass of water, and the individual filaments 

 can then be isolated with needles. 



The filaments consist of three layers : a smooth, membranous, 

 highly refracting sheath, offering great resistance to the action of 

 reagents ; a medullary mass, swollen by acids ; and a fine, rounded, 

 somewhat granular axial cord. The terminal knob consists chiefly of 

 medulla ; it contains granules, but nothing answering to a cell- 

 nucleus. 



Fibrils of equal thickness to these, but not more than 0-9 mm. 

 long, are also found. The filaments show no cellulose reaction, and 

 probably agree in composition with the spongiolin fibres of the 

 skeleton. 



* This Journal, ante, p. 102. f Ibid., ii. (1879) p. 49. 



