78 
line of the dorsal surface, and these upper edges are united 
by the middle portion of the primitive shell, which becomes 
converted into the hinge ligament. 
In Anodonta, and apparently in Unio, the process is some- 
what similar, but each calcareous valve is formed by the union 
of a number of patches which are deposited on the inner sur- 
face of the embryonic shell at several centres of calcification. 
In the oyster the primitive shell appears to be wanting, and 
the two valves of the shell correspond to the two centres of 
calcification which are present in Cyclas. 
J The fact that the primitive shell of Cyclas or of Anodonta 
closely resembles the embryonic shell of a Gasteropod, a Ptero- 
pod or a Cephalopod in appearance and shape, as well as in 
position and method of formation, would seem to indicate 
that the manner in which the shell of Cyclas is formed is the 
primitive method, and the process, in the oyster, a secondary 
modification. 
THE MANTLE. 
The manner in which the lobes and cavity of the mantle 
are formed in the oyster appears to differ slightly from the 
process as it is exhibited during the development of the fresh- 
water Lamellibranchs. 
In Anodonta the mantle cavity is formed by an invagina- 
tion of the body wall between the free ventral edges of the 
valves, and in Cyclas the mantle folds off from the sides of 
the body in advance of the growing shell. The process in the 
oyster is about half-way between that in Cyclas and that in 
Anodonta. The mantle is formed as a ridge or fold of the in- 
tegument on each side of the body, but this ridge is situated 
at the ventral edge of the shell, not % as in Cyclas, in advance 
of it. 
LIST OF PAPERS QUOTED. 
The following is a list of the papers to which I have re- 
ferred, or from which I have quoted, in the preceding paper : 
History of the Oyster and Oyster Fisheries. By T. C. 
Eyton. London : 1858. 
