Marine Sponges from Japan. 399 



for the fibre being- glossy and the spicules hexactinellid, while 

 the fibre of Phahellia ventilabrum is keratose and the spicules 

 " monactinellid," the two are almost identical, further illus- 

 trating the fact that there is a unity of plan in the Spongida 

 as regards general form, whatever the nature of the material 

 of construction may be. 



On account of the addition of the central sexradiate to the 

 quadrilateral areas of the dermal reticulation (fig. 9), the 

 interstices of the latter become divided into triangular spaces 

 very much like those of Esperia, which, together with the 

 poriferous tympanizing sarcode, thus gives it a similar beauty ; 

 while the presence of the sarcode with the pores still existing 

 in it, although dry now, shows that the specimen was taken 

 alive and preserved with much care afterwards. 



I have often thought with reference to the open state of the 

 pores under such circumstances that — after having shown in 

 my " Ultimate Structure of Spongilla " ( l Annals,' 1857, 

 vol. xx. p. 25, pi. i. figs. 6 and 7), that the "investing mem- 

 brane " or dermal sarcode is composed of plastic nucleated 

 cells, and that it is by their partial separation and alteration 

 in form that the pores are alternately produced and closed — 

 it might be asked how the pores are maintained in an 

 open state during desiccation. In reply to which it may be 

 stated that these plastic units, which in fact are the epithelial 

 cells, are, like Amoeba?, so exceedingly slow in their move- 

 ments that, under desiccation, they pass into a homogeneous 

 dry membrane before they have time to go very far in the 

 alteration of their form, and thus the pores remain open ; thus 

 too the cilia of the spongozoa is often preserved. 



The fact of the pores in this species opening directly into 

 large channels of the excretory canal-system whose vents 

 are immediately opposite, i. e. on the other side of the wall 

 (PI. XIII. figs. 1, 2, and 9), recalls to mind what occurs in 

 Teichonella labyrinthica ('Annals,' 1885, vol. xv. p. 119, 

 pi. iv. fig. 7) in this respect, that is the nutritive particles 

 which pass through the pores do not go on to the spongozoa 

 (Geisselzellen) of the ampullaceous sacs, direct. 



The presence of the smooth acerates, no. 3 (fig. 4, e, /), 

 is, so far as my knowledge extends, a peculiarity in the Hex- 

 actinellida • at the same time they appear to me to be very 

 like the smooth early form of the barbula, which also occurs 

 in bundles of this form at an early period, as well as after- 

 wards in their fully developed state. The wncapitate rays of 

 the scopuliform spicule furnished for the most part with only 

 two is also new to me in situ ; while the only other case in 

 which 1 have met with this form is in my mountings of the dust 



