454 REVISION OF THE AXINELLID^, i., 



Til the characters whicli it combines, the genus is a most 

 aiiomaluus one ; and the (juestion of its relationship affords 

 scopt^ for considerable speculation. The form of the spini- 

 spiruhe irresistibly suggests their derivation from spirasters ; 

 and this view of their origin receives strong support from the 

 fact that identically similar spicules — which undoubtedly are de- 

 rived from spirasters — occur in Spirastrella l^i) sjnnispirtdifera 

 (Carter) Dendy(7), and Spiradrdla (I) dUatata (Kieschnick) 

 Thiele(39), and from the fact that, in the latter species (which is 

 a|)parently unique in this respect among the Spirasti-ellidfe), the 

 megascleres are united into definite fibres by means of spongin. 

 Also in support of this view, is the fact of the presence of niicro- 

 strongyla. On the other hand, in structural features of the 

 skeleton, the genus conforms to a ty^kt which is characteristic ^^i 

 genera in which the microscleres are sigmata or ai'e such as are 

 known to occur in association with sigmata. Ti'achycladus, 

 therefore, ai)})ears to form a connecting link between the Sjjira- 

 strellidie and the signiatophorous section of the Monaxonida, and 

 provides ground for the view that these two groups are derived 

 from a common Monaxonid stem.* 



As the several species agree \'ery closely in by far the greater 

 number of their characters, a preliminary general account of them 

 is desirable in order to obviate to some extent the necessity of 

 repetition in their separate descriptions. 



With the possible exception of T. hacterhtm — which is described 

 by Leiidenfeld as " eiformig, niit schmaler Basis festgewachsen " 

 — all the species are of ramose habit, typically stipitate and more 

 or less arborescent, with branches which are circular or nearly 

 so in cross-section (occasionally somewhat compressed in T. 

 reteporonKs), and never of considerable stoutness; in T. piiatulosus 

 alone, the branches generally remain much abbreviated, closely 

 crowded, and more or less coalesced together proximally, thus 

 sometimes (through excessive reduction and fusion) producing a 



■^ In tliis connection, I may menti(jn that evidence is not wanting which 

 would justify the hypothesis that sigmata and clieht have originated from 

 spirasters, peiliaps independently; and it is even possible that the acan- 

 thoscleres of the Desmacidonidic aie similarly derived. 



