down the distinction between " Monoceratina" and "Hexaceratina," or "Dictyoceratiua" 

 and " Dendroceratina." The Spongeliidae, like the Aplysillidse, have large sac-shaped 

 flagellate chambers, simple canal-system, and clear transparent ground-substance. 

 In fact, they differ from the Aplysillidse only in their reticulate skeleton and in their 

 habit of taking foreign bodies into the fibre. The so-called " pith " in the fibre is 

 also less obvious, but this is an extremely variable character, and one upon which we 

 cannot place very much reliance for purposes of classification. In Megalopastas 

 pulvillus, for example, one and the same section may show great differences in this 

 respect, some fibres showing a strongly marked pith, differentiated by its darker 

 colour, and others apparently having no pith at all (Plate XV., fig. 3), the difference 

 apparently depending upon differences in local conditions at the time when the fibre 

 is growing, which give rise to a more or less distinct lamination analogous to the 

 annual rings in a tree trunk. 



Moreover, when we remember that SCHULZE has described (71), under the name 

 Spongelia xpinifera, a species in which the arenaceous fibres do not form a network 

 at all, but are arranged in a tree-like manner, as in the genus Aplysilla, we see at 

 once that the distinction between the Spongeliidae and Aplysillidse is purely arbitrary, 

 though, as a matter of convenience, it may, perhaps, still be maintained. From the 

 Spongeliidse the transition to the Spongiidse, by complication of the canal-system, 

 reduction in the size of the flagellate chambers and granulation of the ground 

 substance between them, is very simple. 



I therefore conclude that the Euceratosa are a natural group descended from the 

 Myxospongidse, that their evolution starts with the Aplysillida3 and ends with the 

 Spongiidse, between which the Spongeliida3 occupy an intermediate position, and that 

 the reticulate skeleton of the higher types has been independently evolved from a 

 more primitive dendritic skeleton. 



FAMILY: APLYSILLIDSE. 



Euceratosa with a dendritic or reticulate skeleton composed of spongin-fibres 

 containing a more or less distinct pith, but usually without foreign inclusions ; 

 sometimes also with isolated spicules of spongin ; with a lacunar canal-system 

 and large sac-shaped flagellate chambers opening by wide mouths direct into 

 wide exhalant lacunae. 



Darwinella, MULLER. 



Aplysillidse with a dendritic skeleton and with isolated spicules of spongin. 



Four species of this remarkable genus have been described, viz., D. aurea, 

 MULLER (67) ; D. auslraliensix, CARTER (18) ; D. joyeuxi, TOPSENT (89) ; and 

 D. simplex, TOPSENT (84) ; but it appears to me somewhat doubtful whether they 

 should all l)e regarded as specifically distinct from one another. 



2 D 2 



