(3 Art. 1. — N. Annandale and T. Kawamura : 



leaves of water-plants or of an irregular network of considerable 

 relative thicknesss but mainly horizental. though often highly 

 convex, in direction. This form is sometimes lightly attached to 

 weeds with small leaves, but also occurs on stones. In the thin 

 form the upper parts of the gemmules is sometimes devoid of 

 spicules and has the pneumatic layer thin, while the foraminal 

 tubule always lies on the lower surface, by which the gemmule 

 adheres to the leaf. In more robust sponges, however, the 

 coats of the gemmule are developed uniformly, the gemmule lies 

 free in the sponge and the tubule does not always point in the 

 same direction. 



The species is only known from the Main Island of Japan. 

 It was described from Kasumi-ga-Ura on the Pacific coast and is 

 not uncommon in L. Biwa, especially in the South Lake. We 

 found a specimen in the small lake at Komatsu also. 



Spongil/a aspinosa PoTTS. 



Pi. II, tig. i. 



1887, Potts, Proc. Acad. Nat. Sc. Philadelphia, p. 184, PI. VIII, fig. vi. 



This species is also related to S. lacustris, from which it may 

 be at once distinguished, when gemmules are present, by the fact 

 that each gemmule is enclosed in an irregular but dense network 

 or cage of skeleton-spicules. These spicules are sometimes smaller 

 than those that take part in the formation of the skeleton in other 

 regions, but this is not always the case. The gemmules have no 

 special spicules but possess a well-developed coat of "granular" 

 air-spaces and a foramen protected by a horny tubule. The micro- 

 scleres are confined or practically confined to the dermal 

 membrane. They were described by Potts as being quite smooth, 

 but in the Japanese specimen which we have seen they bear shorl 

 irregular spines (PI. II, tig. la). Potts' s specimens were green, 

 but the one from Lake Biwa was yellowish. Forms of S. lacustris 

 are known in which the gemmule-spicules are very few or altogether 

 lacking, but this is exceptional and when it occurs the pneumatic 

 coat is usually degenerate; the gemmules are never enclosed in 



