Rev. T. Hincks on British Polyzoa. 523 



in this a natural arrangement, and should certainly sepa- 

 rate the two forms generically, though quite prepared to 

 admit that genealogically they may be not very remotely 

 connected. 



Proceeding to the application of the principle that I have 

 just laid down, the old genus Lepralia^ founded oil mere simi- 

 larity of habit, without reference to the zoooecial characters, 

 must of course be dismembered and divided into groups, based 

 on the structural peculiarities of the cell. I have suggested a 

 number of such groups in another part of this paper. 



As to the question whether forms which exhibit an ered 

 mode of growth should be combined in one genus with those 

 which are crustaceous in habit, I believe that no universal 

 rule can be laid down. One thing seems to me clear — that the 

 tendency to form free expansions, consisting of a single layer 

 of cells, ought not to be accounted a generic diagnostic. The 

 genus Hemeschara represents a very trivial variety of habit, 

 and forms a most unnatural group, including as it does very 

 distinct types of cell. 



Nor can the development of erect foliaceous expansions, 

 composed of two layers of cells placed hack to hack, by forms 

 which very commonly assume a simply crustaceous habit, 

 be taken as a distinctive character in itself, and apart from 

 structural peculiarities in the zoooecium. Lepralia Lands- 

 horovii has been transferred to the genus Eschara, be- 

 cause it takes on at times such an erect, foliaceous habit, 

 though more usually crustaceous in its mode of growth. But 

 this very trilling and occasional change of habit is a very 

 insufficient ground for severing it from the simply incrusting 

 forms to which it is closely allied in zoooecial character (e. g. 

 Lepralia reticulata, L. trispinosa, &c.). 



On the other hand, the habitual formation of erect, well- 

 compacted, more or less dendroid zoaria, which marks the 

 adult or perfect condition of the species, is a character fairly 

 included in the generic diagnosis. The old genus Eschara 

 rests on this foundation ; but, like Lepralia, it will require to 

 be subdivided should it be found to include various distinct 

 zoooecial types. I cannot see that the validity of this cha- 

 racter is affected by the fact that in many cases the dendroid 

 zoaria exhibit a tendency to be decurrent at the base, and 

 spread out into a lepralioid crust of varying size. This is no 

 doubt a significant genealogical indication ; but if we are to 

 have any distribution into groups at all, I see no reason why 

 forms which have made a great and distinctive advance in 

 colonial development should not be set apart from those which 

 have been left behind at a lower grade. At the same time we 



