730 PROF. A. DENDY AND MR. R. W. H. ROW ON 



Flagellate chambers ranging from long and possibly branched, 

 with a tendency to radial arrangement round the exhalant 

 canals, to small, approximately spherical, and scattered. 

 With a distinct and independent dermal membrane (or 

 cortex) pierced by true dermal pores. Skeleton consisting 

 mainly of equiangular and equiradiate spicules, which may 

 become sagittal at the oscular margins. Radiates of the 

 chamber' ayer without definite arrangement, but irregularly 

 scattered in the walls of the elongated chambers, or between 

 the small, scattered chambers. No subgastral sagittal 

 radiates. Nuclei of collared cells probably always basal. 



This family was provided by Deudy [1892 B] for the reception 

 of the genus Leucascus with its two species, L. simplex and 

 L. clavatus. Minchin [1900] refused to recognise either the 

 genus or the family, and included the two species in his Clathrina, 

 apparently ignoring the fact that none of the known species of 

 Glathrina, or indeed any other homocoel sponge, possess an 

 independent dermal membrane or cortex. It is true that many 

 Clathrinas develop a pseudoderm, but this is invariably formed 

 from the outermost tubes of the reticulation, and therefore 

 includes a layer of endoderm (gastral layer). In Leucascus, on 

 the other hand, the dermal membrane is formed exclusively of 

 ectoderm and mesogloea (dermal layer), and does not consist 

 merely of the outer tubes of the reticulation. Moreover, the 

 radiate and non-reticulate arrangement of the elongated chambers 

 in Leitcascus indicates a relationship with Dendya leather than 

 with the Olathrinoid Leucosolenias. 



We here extend our conception of the family Leucascidse to 

 include, not only the genus Leucascus and allied genei-a with a 

 similar type of canal system, but also a number of species with 

 a leuconoid tj'pe of canal system, which we have hitherto assigned 

 to the genera Leucandra and Leucilla. The skeleton of these 

 species, in the absence of all ti-aces of syconoid ancestry such as 

 subgastral sagittal trii'adiates, cleai-ly indicates a wide phylo- 

 genetic separation from the typical Leucandras and Leucillas, 

 such as Leucandra aspera and Leucilla amphora, and closely 

 resembles that of Leucascus. 



It will be remembered that Haeckel, in " Die Kalkschwamme " 

 (vol. ii. p. 122, [1872]), placed his Leucetta primigenia (one of the 

 species which was formerly assigned to Leucandra, but now placed 

 in the Leucascidse) at the beginning of his Leucones, which he 

 derived directly from a,n Ascon ancestry. He says : ^^ Leucetta 

 primigenia, als die wahrscheinliche Stainmform der Leuconen, 

 steht in ihre Skeletbildung der gemeinsamen Stammform aller 

 Kalkschwamme, der Ascetta primordialis, so nahe, dass man sie 

 unmittelbar von der letzteren ableiten kann." We agree with 

 these vioAvs so far as the relationship -to the "A scones "is con- 

 cerned, but we can no longer agree that the more advanced types 

 of "Leucones" (such as Leucandra) have had a leucettid ancestry. 



