ON THE MICROSCOPIC STRUCTURE OF SHELLS. 21 



comparison of the general characters of the animal, with that of Avicula on 

 the one hand, and Mytilus on the other. In Mytilus there are always two ad- 

 ductor muscles, the anterior very small, the posterior much larger ; the lobes 

 of the mantle are united posteriorly at one point, so that there is a single anal 

 siphon ; the aperture of the mouth is not furnished with papillae ; and the liga- 

 ment is altogether external. In Pinna there are still two unequal adductor 

 muscles ; the lobes of the mantle have no posterior commissure (though partly 

 united along the back), and consequently there is no anal siphon ; the mouth 

 as well as the lips are covered with membranous papillee ; the ligament is 

 very narrow and elongated, often covered by a thin testaceous lamina, and 

 loses almost all the characters of the external ligaments. In Avicula there 

 is no longer any anterior adductor muscle ; there is no posterior [commissure 

 of the mantle ; the mouth is furnished with papilte, and the ligament has no 

 longer any of the characters of external ligaments, entirely resembling those 

 of the other Monomyaria. The animal of Pema, so far as it is known, ap- 

 pears to be very closely allied to that of Avicula. Hence the only impor- 

 tant character by which Pinna is connected with Mytilus, is the presence of 

 an anterior adductor muscle ; but against this are to be set the want of the 

 posterior commissure of the mantle, the difference in the position of the 

 ligament, and the presence of papillae on the inner surface of the mouth and 

 lips, — in all which points there is a much closer approximation to Avicula. 

 Thus we see how correct is the determination which would have been formed 

 from the sole consideration of the structure of the shell ; and even if we con- 

 sider this but as a single character, to be taken into account with others in 

 the determination of the position of the genus, I think it difficult to resist 

 the preponderance of evidence for detaching Pinna from the family Myti- 

 lacece, and for uniting it with the Margaritacece. 



XIV. Nay odea. 



54. Although this family is usually separated widely from the Margaritacece 

 by systematists, there appear to me many points of resemblance between 

 them. Contrary to Lamarck's statement, the lobes of the mantle in both 

 Unio and Anodon are entirely open along their whole extent, and the chan- 

 nel which forms the anal passage is made up of the two branchial laminae, 

 which are there adherent together. Now it is extremely interesting to find 

 that in this group, which conducts us so remarkably from the Lamellibran- 

 chiata with the lobes of the mantle entirely open, to those in which it is closed, 

 the prismatic cellular structure so characteristic of the former division is 

 still found, but in small quantity. The principal part of the shell is nacreous ; 

 and the prismatic cellular structure forms but a very thin layer beneath the 

 periostracum (fig. 51). It is to this that the dead- white aspect of the shell 

 is due, when the epidermis has been frayed off (as it often is during the life 

 of the animal, especially near the umbo) Avithout the nacre being brought into 

 view. I can discover no difference between Unio and Anodon in the micro- 

 scopic characters of the shell ; and consequently can offer no objection on 

 this score to the reunion of these two genera, as proposed by M. Deshayes. 



55. In connection with these last families, I may allude to the structure of 

 the cui-ious genus Etheria ; in regard to the place of which, there is not yet an 

 agreement amongst systematists. By many Conchologists it has been arranged 

 among the Chamacem, chiefly on account of its tendency to attach its lower 

 valve to solid bodies. Its removal from these, however, has been proved to 

 be required by additional knowledge regarding the structure of the animal. 

 M. Deshayes seems inclined to rank it among the Nayadece ; M. de Blain- 

 ville thinks it should be associated with the Margaritacece. The lobes of its 

 mantle are entirely open, but there is an anal passage formed by the adhesion 



