32 CELLULARIID^. 



structure there is tlie closest agreement between it and 

 Scj'upocellaria. The avicularia occupy the same positions 

 in both, and though the vibracula are more highly deve- 

 loped in Caberea, they are essentially the same in structure 

 as those of the other genus. We have also a link between 

 the two forms in the Canda arachnoides, Busk, an articu- 

 lated species, in which the vibracular cells are of large 

 size, and stretch diagonally across two thirds of the length 

 of the zooecium. 



The non-articulated condition of the zoarium does not 

 appear to be a point of so much significance as to warrant 

 the separation of Caberea from forms with which it has 

 such a close general affinity. 



I retain the genera Cellularia, Menipea, and Scrupocel- 

 laria as defined by Busk and Wyville Thomson, instead 

 of uniting them, with Smitt, in a single division. The 

 differences between them lie chiefly in the presence or 

 absence and in the disposition of the appendicular organs 

 (avicularia and vibracula) . As there is undoubtedly much 

 variability in these particulars, as the avicularia for in- 

 stance are often present on some cells and absent on 

 others within the limits of one and the same colony, 

 and in some species are now present and now altogether 

 wanting, it may be urged that these distinctions are 

 purely arbitrary*. But it must be remembered that, with 

 our present knowledge of the Polyzoa, a perfectly natural 

 arrangement is unattainable, and that these groups are 

 probably as good for their purpose as any that can be sub- 

 stituted for them. They represent some of the principal 

 modifications of the family type ; and in point of fact each 

 of them has, on the whole, a distinct and characteristic 

 fades of its own. 



* Vide " Floridan Bryozoa," by Prof. Smitt, part i. p. 1() (Kongl. 

 STenska Veteusk.-Akad. Handl. Band x. No. 11). 



