3 i6 THE TANGANYIKA PROBLEM. 



than the Spongillidse are usually supposed to have. As a 

 matter of fact, it is difficult to make out what structural 

 reasons there are for retaining the family Spongillidse. It is 

 not at all improbable that, when they are more carefully 

 studied, they will be distributed among the several genera 

 of the Homorrhaphidse. But as our knowledge has not yet 

 attained a stage which will enable us to do this, it is deemed 

 advisable for the present to place this new species among 

 the Spongillidse, and to retain that assemblage of sponges 

 as a family, however artificial it may be. 



Spongilla moorei appears to be more closely related to 

 Spongilla aspinosa (Potts) than to any other species of the 

 Spongillinae. Both species agree in possessing spicules 

 which are smooth, straight or curved, and for the most part 

 rather abruptly pointed. Malformed spicules, as they are 

 described by Potts, are found in both, but they appear to be 

 more numerous and more complicated in Spongilla moorei 

 than in Spongilla aspinosa. Further, both species produce 

 gemmules which are small in size, spherical in shape, and 

 supplied with a thin crust which is not protected by spicules 

 characteristic of the gemmule, but by the ordinary skeleton 

 spicules. Though the gemmules are few in Spongilla 

 aspinosa, they are more numerous than in Spongilla moorei, 

 a feature which may be explained either by the lesser 

 importance and consequent scarcity of the gemmule in the 

 latter species, or simply by the season at which the material 

 was collected. 



Spongilla aspinosa differs from Spongilla moorei in that it 

 possesses small flesh spicules, which lie on the dermal mem- 

 brane and among the smooth, slender skeleton spicules. 

 These small spicules are not found in Spongilla moorei, unless 

 they are represented by those which are drawn in Fig. 2 — b., 

 which is probably the case. However, it must be admitted, 



