REPORT ON THE DEEP-SEA KERATOSA. 57 



of the dry fibrillar was filled with air. The difference between the thin axial thread 

 and its thick spongin-envelope is the same as in the thinner forms of the so-called 

 homogeneous horny fibres of the Euspongidas, Spongelidrc, &c. On the other hand, 

 they recall also the peculiar so-called " filaments " of the Hircinidse, and this leads us 

 to the question of the true nature of the fibrillse. 



Nature of the Fibrillse. — The first cpiestion arising out of the examination of the 

 peculiar fibrillar of Stannomidse, and also of the similar " filaments " of the Hircinidse, is 

 this: Are they produced by the sponge itself? or are they foreign organisms which live 

 in the sponge as parasites or symbiontes ? As is well known, this question is not yet 

 decided in the case of the Hircinidse. Polejaeff, in his Eeport on the Keratosa, 1 

 discusses the nature and the systematic value of the filaments of the Hircinidae, and the 

 majority of modern spongiologists agree with him when he says that " their nature as 

 independent organisms is clearly established." But of what nature are these "independent 

 organisms ? " No zoologist will accept them as animals, no protistologist can regard 

 them as neutral protists, no botanist will acknowledge their vegetable nature ! All 

 botanists who have thoroughly examined the filaments of the Hircinidse, and among 

 these are some great authorities, mainly fungologists, declare decidedly that they are not 

 fungi, and not plants at all. Indeed, neither their chemical nature nor their anatomical 

 structure is that of any fungus or alga, and, although many observers have examined 

 them for a long time and in all possible directions, no one has been able to discover 

 their fructification and development. 



Polejaeff's principal argument in favour of the parasitical nature of the filaments 

 found in the Hircinidas is as -follows : — " F. E. Schulze made out the structure of sponges 

 characterised by the presence of filaments, and found that anatomically and histologically 

 they do not differ from sponges which, like Euspongia, have never been found with 

 filaments." 2 This argument, in my opinion, has no decisive value. If we apply it to the 

 Chondrosidse, we might arrive at the following conclusion : — " Chondrilla, characterised 

 by the presence of sphero-stellate sdiceous spicules, does not differ anatomically and 

 histologically from Cliondrosia, which has never been found with these spicules. There- 

 fore these spicules are not produced by the sponge itself, but are independent organisms." 

 On the other hand, the fact that the fibrillse of the Hircinidse are not in direct connection 

 with the reticulate horny skeleton of these Keratosa has also no decisive value. For 

 Darwinella possesses numerous radiate horny spicules imbedded in the mesoderm, 

 without connection with the ramified tree of the Keratose skeleton ; so also have many 

 Halichondrinse siliceous " flesh-spicules " imbedded in the connective tissue, without 

 connection with the main skeleton. 



I am therefore inclined to regard the filaments of the Hircinidas, and also the similar 

 fibrillar of the Stannomidse, as true skeletal fibres, comparable to the elastic fibres in the 



1 Zool. Chall. Exp., part xxxi. pp. 11-16. 2 hoc. cit., p. 13. 



(ZOOL. CHALL. EXP. PART LX3XII. 1889.) NllM! 8 



