12 THE VOYAGE OF H.M.S. CHALLENGER. 



corresponding individuals of sponges (the flagello-chambers), we find similar relations 

 between the associated persons and their large community. The variable manner in 

 which the persons or zooids are connected and arranged, and in which the common 

 canal-system of the community (the ccenenchyma) is developed, often shows a striking 

 similarity in both cases. 



The theory that the true primordial form of all sponges was an Olynthus, and that 

 all the other forms were developed from an Olynthus stage (Olynthula), was first stated 

 in 1872 in my Monograph of the Calcispongise (in connection with the Gastrsea theory), 

 and has now been accepted by most modern spongiologists, by F. E. Schulze, Sollas, 

 Lendenfeld, Vosmaer, Marshall, Keller, and others. Many of these recent observers 

 have demonstrated the existence of a homologous larval form (Olynthula) in 

 the ontogeny of very different sponges. But ripe and fully-developed sponges, which 

 persist in the primitive Olynthus form, and produce in this form eggs and sperm, were 

 known hitherto only among the Calcarea (Calcolynthus). 1 It is therefore a fact of 

 general interest that among the Deep-sea Keratosa collected by the Challenger, there 

 occurs a small ripe sponge (with eggs) which seems to be a true horny Olynthus (or, 

 more strictly speaking, an arenaceous Olynthus), the remarkable Ammolynthus figured 

 in PL VIII. figs. ] , 2. Unfortunately the delicate soft parts of the tissues, in this as well 

 as in all the other Deep-sea Keratosa, were very badly preserved, so that the histological 

 evidence of its true nature could not be demonstrated with all the desirable certainty. 



Eegarding the Olynthus as the simple sponge individual or zooid (Gastrsea), and as 

 equivalent or homologous to a single tubular branch in the Homoccela, and to a single 

 flagello-chamber in the Heteroccela, we must regard all other sponge-forms as corms, 

 composed of numerous Olynths. They exhibit the same relation to Olynthus as the 

 various Hydroid corms bear to Hydra. The external form of these corms or stocks in 

 the Deep-sea Keratosa has the same variability and wide divergence as in the other 

 groups of the class. This is especially the case in the smaller forms, the Ammoconidse and 

 Psamminidse, where we find irregular, crusty, and massive corms, flat discs, tuberose 

 lumps, branched and reticular stocks, &c. Among the larger Keratosa of the deep sea 

 the prevailing and most common form is that of a thin flabelliform leaf (Stannophyllum, 

 PL I.; Psammophyllum, Pis. IV., V.). This form (rarely occurring in other sponges) is 

 remarkable for its perfect bilateral symmetry (or, strictly speaking, the amphithect 

 ground form). The regular symmetry is especially striking in those forms in which 

 branched ribs are disposed on both sides (PL I. figs. 3, 4 ; PL IV. fig. 5). The two 

 flat sides of the reniform leaf exhibit constantly the same structure ; it is therefore 

 probable that these flabelliform sponges arise vertically from the sea-bottom, attached by 

 the slender pedicle, which is inserted in the middle of the basal concave margin. 



1 See pi. i., pi. vi. fig. 1, pi. xi. figs. 6-9, pi. xiii., &c, in my Monograph. 



