on the Polyzoa, 159 



neighbouring cells by six short processes or extensions of its 

 wall, and these connecting bands, though not absolutely 

 isolated from one another, are separated by a deep depression. 

 So that we have a first step towards the retiforra condition 

 which is characteristic of such a form as DlacliofU^ Busk. 

 On the strength of this structural peculiarity MacGillivray, in 

 a recent part of his important* work on the Polyzoa of Vic- 

 toria, transfers this species to the genus Beania (which he 

 identifies with Diachoris). In support of this view he points 

 out that its avicularium is also transitional and marks an 

 advance towards the capitate form which distinguishes the 

 Bicellarian family. But M. radicifera, whatever its tran- 

 sitional tendencies, is still a characteristic Memhrani'pora. If 

 it shows us a possible road from the one type of structure to 

 the other, it has not itself joined the Bicellarian camp. Its 

 cells are still vmited in a solid zoarium ; its avicularium, 

 though suggestive of change, has not reached the articulate 

 goal, but remains fast soldered to its place and as incapable 

 of movement as the most rudimentary appendage. It is not, 

 I respectfully submit, permissible to sever it from the tribe to 

 which its actual characteristics ally it, and transfer it to 

 one towards which at most it has only a few structural 

 leanings. Its partially disjunct cells and its avicularium, 

 simulating the bird's-head form, are interesting as genealogical 

 hints, but they are nothing more. 



Diachoris (or Beania) is a true Bicellarian, with the 

 chitino-membranous, boat-shaped cell of a Bugula, and the 

 highly organized capitate and articulated avicularium so 

 characteristic of that family. The reticulate structure of its 

 zoarium is after all its least significant character ; its place is 

 determined by the essential structure of its zooecium, and, so 

 far as this is concerned, it has little in common with the 

 present species. 



3. Family ]y[enil)raniporid0e {continued). 

 Notes on the Genera. 



There is no more natural group amongst the Polyzoa than 

 the Memhraniporce^ in which the primitive membranous 

 covering of the cell endures, either wholly or in great part, as 

 a permanent character. It embraces an immense number of 

 species, and, as a matter of convenience, it would be desirable 

 to break it up into subgroups, if any natural basis could be 

 found for them. But important modifications of the typical 

 characters are rare, and of the subdivisions that have been pro- 

 posed a considerable proportion are purely artificial. In his 



