MOLLUSCOUS ANBIALS. 393 



my notice, and I do not even recognise, in the case of the lower sub-class 

 Polyzoa, any reason for the particular series, which I accept as usually 

 given. In the higher sub-class the divisions are no more than family 

 gfoupes, though for our present purpose, equivalent to orders. In the 

 Polyzoa they have more extent, and include more variety of form, but 

 are less certainly established. They are as follows : 



A. TUNICATA 



SUB-CLASS SUB-CLASS 



Polyzoa. Tunicata proper, or Ascidioidea. 



1 Cheilostomata. 1 Ascididae. 



2 Cyclostomata. 2 Clavellinidae. 



3 Ctenostomata. 3 Pyrosomatidae. 



4 Pedicellinea. 4 Botryllidse. 



5 Hippocrepia. 5 Salpidse. 



I merely refer to this class to give as much completeness as possible to 

 the view of the sub-kingdom whtch I submit to you : I am myself ac- 

 quainted with it only from books and preserved specimens, most of them 

 belonging to Polyzoa. 



We proceed to class Conchifera : of its highly interesting lower sub 

 class, Palliobranchiata, I can only say at present that the eight families, 

 into which they are divided in the best recent works, seem to me to 

 require some combinations. This will, of course, depend on our views 

 as to the mode of limiting families, some founding them chiefly on a sin- 

 gle structural character, whilst others require the concurrence of several 

 characters to mark the type of the groupe, but admit of variability in 

 one or more of these, so long as the organism appears to be nearer to a 

 particular type than any other ; some admitting a family wherever they 

 observe a certain resemblance amongst a few genera, whilst others expect 

 the principal families which, in a sub-class, might almost as well be 

 called orders, to express certain tendencies of development in analogy 

 with corresponding divisions of other classes, thus occupying a definite 

 place in a general system. Perhaps we may be jiistified in treating 

 Craniadce and Discinida^. as in our sense one family with their hingeless 

 oyster-like shells, and as exhibiting the lowest Palliobranchiate type, next 

 to these would come the Lingulidce, and I should be tempted, conjectu- 

 rally, to include Orthidce, under Proditctidce, and to unite Spiriferidoi 

 and KhynconellidcB with their unpunctate shells and spirally folded 

 arms, usually more or less supported by a shelly prop, suitably bent for 



