756 PROF. A, DENDY AND MR. R. W. H. ROW ON 



Section B. With large, usually radially arranged oxea, 

 bub without microxea. 



Although the author's description does not conform to the 

 above diagnosis, we include Polejaeff"s Leucilla connexiva in this 

 section of the genus, for the figures given by him show oxea 

 present, though no reference is made to them in the text, and 

 our own examination of the type specimen revealed the presence 

 of occasional trichoxea. 



4. y. CONNEXIVA Polejaeff. 



Leucilla conriewiva Polejaeff [1883]. 



5. V. CYATHUS Verrill. 



Leucandra cyathus Verrill [1873]. 



6. Y. DENDYI Row. 



Vosmaerojasis dendyi Row [1913 MS.]. 



7. Y. PRiMiTivA Roto. 



Vosmaeropsis primitiva Row [1913 MS.]. 



8. Y. SERiCATUM Ridley. 



Aphroceras sericatiom Ridley [1884]. 

 This species has been placed in Vosmaeropsis as a result of an 

 examination of the type specimen made by us at the Natural 

 History Department of the British Museum, which revealed the 

 existence of typical subdermal pseudosagittal trii-adiates, though 

 the author's original description does not mention them. 



Genus 22. Grantilla Row [1909] (emend.). 



Diagnosis. Canal system sy conoid. Tubar skeleton ( ? always) in- 

 articulate, composed of subdermal pseudosagittal triradiates 

 and subgastral sagittal triradiates, supplemented by 

 subdermal quadriradiates. No colossal longitudinal oxea. 



For illustrations of this genus see Row [1909]. 



This gen\is was originally proposed by Row for two species, 

 G. qiiadriradiata and G. hastifera, which were supposed to possess 

 certain features that necessitated the provision of a new family, 

 Grantillidfe. We now consider, however, that the characters in 

 question do not represent any fundamental peculiarities of 

 structure, and we have therefore abandoned the family, as already 

 stated in the Introduction. 



One of the two species originally assigned to Grantilla, G. 

 qundriradlata, however, presents an association of subdermal 

 quadriradiates with subdermal pseudosagittal triradiates, which is 

 not known in any other species of calcareous sponge, and we 

 therefore I'etain the name Grantilla for this species with an 

 emended diagnosis. The development of subdermal quadri- 

 radiates has evidently taken place repeatedly in the phylogeny of 



