394 Mr. H. J. Carter on the 



compressa to 12 by 8-1800tlis of an inch (fig. 18) ; and this 

 will bring it nearer to the size of the two specimens found in 

 the preceding stages (figs. 16 & 17), although then the latter 

 will a})pear very large. Still, that they do represent the earlier 

 stages of this period of devel()j)ment in Grantia compressa, 

 their composition demonstrated beyond doubt. 



Hence, with the exception of following the active embryo of 

 Grantia compressa to its state of fixation and the withdrawal 

 of the cilia, which, from its microscopic minuteness, would be 

 very difficult if possible, the whole of the development of the 

 embryo of Grantia compressa, from the earliest recognizable 

 appearance of the ovum to the fully developed sponge, has 

 thus been described and illustrated. 



Observations. 



When we compare the embryo of Grantia compressa with 

 that of Ilalicliondria simulans at their exit from the parent 

 respectively, we can hardly come to any other conclusion than 

 that their development into the true form of the respective 

 parents must follow a similar course — since, although this 

 form may difl:er in these two particular sponges, it does not so 

 generally ; for there are tubular and sessile spreading forms of 

 both calcareous and siliceous sponges, and in some instances 

 even the same species of either may appear under both these 

 forms. 



In the first place, the shape of the embryo in the sponges 

 just mentioned is conical, with a pointed and a truncated end ; 

 the body is covered by a layer of minute monociliated cells 

 (the ectoderm), whose cilia slope in opposite directions from 

 the apex, and are more or less inclined backwards towards the 

 obtuse end, where there is a bunch of larger cells uncovered 

 by cilia ; the embryo progresses Avith its pointed end fore- 

 most, and rotates from left to right ; and both embryos attach 

 themselves to the bottom of the glass vessel and foreign objects 

 respectively by the posterior extremity or bunch of large cells, 

 which thus appear to possess a plastic prehensile property. 



On the other hand, internally, the body is composed of 

 sareode charged with cells (of dififerent sizes) and granules, 

 among which the largest of the former far exceed in size 

 the dermal cells ; while the bunch of large cells at the 

 posterior end may be seen in the embryo of Grantia compressa 

 to originate from its centre, thus having the body-substance 

 (which appears to be the endoderm or substance in wdiich the 

 spicules, horny skeleton, and ampullaceous sacs are developed) 

 between it and the ectodermal layer (PI. XX. figs. 13 & 15). 



Under these circumstances we cannot help concluding that 



