338 THE AMERICAN MONTHLY [Dec 



but in place of a simple tube with its one exhalent open- 

 ing, there are here hundreds of simple tubes, not leading 

 an independent existence but all connected with and open- 

 ing into one central cavity, which, in turn, opens to the 

 exterior by one os.culum. The water enters directly into 

 the flagellate chambers by means of prosopyles which oc- 

 cur freely in the part of the wall exposed to the sea ; then 

 it flows through the chamber and so into the common 

 central cavity, finally passing out of the sponge by the 

 osculum at the top. 



If sponges of this species are gathered in the summer, 

 scores of embryos will be found embedded in the gelatin- 

 ous material supporting the chambers or even in the cham- 

 bers themselves, and anyone wishing to study the embry- 

 ology of a sponge cannot do better than commence with 

 Sycon raphanus. For this reason I have figured and de- 

 scribed the various stages (see frontispiece) in the develop- 

 ment of such a sponge, nearly all of which can be follov/- 

 ed without much difficulty. The best methods to employ 

 in such investigation can be obtained from text books 

 which treat of special microscopic technique. 



The next sponge on our list is Sycon gelatinosum, often 

 known under the name given to it by Haeckel, as S. arbo- 

 rea. This is a very common species in Australia and is 

 very variable in form, being either colonial or solitary. 

 The solitary form, about an inch in height, is found at- 

 tached to the under surface of stones; whilst the colonial 

 form, which consists of tubes very richly branched, 

 reaches a height of 2 or 3 inches and is found growing 

 amongst the sea-grass on the ocean side of the reef at 

 Flinders. In color these sponges are a creamy white and 

 in some a fringe of spicules surrounds the oscula whilst 

 in others it is absent. 



A single tube or branch of the colonial form represents 

 a single individual of S. raphanus and if we examine a 

 transverse section of such a tube we notice that the cham- 



