in POEIFEEA 43 



jelly makes its appearance, which is the real stiffening element in 

 the sponge-wall. 



The formation of inhalent pores is now hegun. Individual cells 

 of the outer or dermal layer extend inwards through the jelly, and 

 press asunder adjacent cells of the inner or gastral layer. These cells 

 then hecome hollowed out, converted into drain-pipes, as one might 

 term it, and the action of the flagella draws in water through them. 

 Other cells migrate from the outer layer into the jelly, and form 

 the characteristic calcareous needles or spicules. The first type of 

 cells are called porocytes, the second scleroblasts. After the pores 

 have been acting for some time, an exhalent opening is formed at 

 the distal end of the cylinder. The fol-mation of this osculum seems 

 to be due in part to the hydrostatic pressure caused by the action of 

 the pores. 



The tiny sponge is now quite comparable to the type of adult 

 sponge exemplified by the genus Zeucosolenia. Its transformation 

 into the adult Sycon is an affair of slow growth, and the process has 

 not been observed in this Grantia, but there is no reason to doubt 

 that it is essentially similar to what occurs in Sycandra raphanus, 

 in which it has been described by Maas (1900). 



In Sycandra, pouches grow out horizontally from the cylinder 

 which forms the body of the young sponge ; they are formed gradually, 

 not all at once. As the pouches are formed the flagellated cells are 

 taken up into them, and the dermal cells migrate inwards from the 

 outside, pressing the flagellated cells asunder, and constitute the 

 epithelium lining the central cavity of the sponge or "paragaster." 

 The interspaces between the openings of the horizontal pouches, that 

 is to say, the niches left between the outer surfaces of these pouches, 

 constitute the inhalent system of canals. In this sponge the re- 

 productive cells seem also to be formed from the dermal layer ; in 

 their undifferentiated form they are full of yolk, and are known as 

 archaeocytes. 



OTHER SPONGES 



To Maas (1898) we owe the demonstration that all sponge larvae 

 are modifications of the type just described. Of the development of 

 the Hexactinellida nothing is known ; larvae, it is true, have been 

 observed which seem to originate from unfertilized eggs, and which 

 resemble the larvae of other siliceous sponges, but their history has 

 not been followed. 



When we turn to the Demospongiae in which the spicules are 

 arranged in cords, and which constitute the vast majority of sponges, 

 we can trace a complete series from a development more primi- 

 tive than that of Grantia to the most modified form. Beginning 

 with Oscarella, which, although devoid of a skeleton, has its affinities 

 with the Demospongiae, Maas shows that the embryo is hatched as 

 an oval blastula, consisting of a uniform layer of flagellated cells. 

 During the course of its free hfe as a larva, the cells of the posterior 



