98 THE NAUTILUS. 



come from tropical or southern countries. There is nothing dis- 

 tinctive in the radula or soft parts of Gundlachia, as far as yet ob- 

 served, to separate it from tlie analogous Ayicylus. 



A paper which, for the first time, brings to bear on this hypo- 

 thesis facts which seem to render it sufficiently acceptable to publish, 

 has been contributed by Erland Nordenskioid to the Zoblogische 

 Anzeiger, XXVI, pp. 590-593, July, 1903, with seventeen figures. 

 In this paper to which the reader may profitably refer, a process 

 such as my hypothesis assumes is fully illustrated in Ancylus mori- 

 candi d'Orbigny, from the Chaco region of Brazil, up to the point of 

 the completion of the epiphragm and the determination of the 

 identity of the forms bearing it with the typical first year Ancylus. 

 The formation of the second-year shell or Gundlachia by these indi- 

 viduals, alone remains to be demonstrated to establish the hypothesis 

 as a fact. 



NOTES ON THE STRUCTURE OF THE SHELLS OF UNIO. 



BY L. S. FRIERSON. 



The shells of Unio are stated by most authors to be composed of 

 three layers, known as the " epidermis," the '' columnar " or " pris- 

 matic " layer, and the "nacreous," or simply called the "nacre." 

 As a matter of fact, however, these shells are composed of four 

 layers, the nacre being composed of two distinct layers. These may 

 be readily noted in a polished section of some thick-shelled species, 

 and especially if a species be chosen, such as Obovaria retusa 

 Lamarck, showing the two layers in different colors. A clearer idea 

 of the two layers may be obtained if the secreting " mantle " be 

 studied. This part of the animal, though called by a single name 

 " mantle " really is composed of two distinct portions, and should 

 have two names. That portion extending from the beaks to the 

 pallial line is thin, and one is tempted to say structureless, while 

 from the pallial line to the margin, it is thickened, and plentifully 

 supplied with nerves and muscles. Tiie extreme edge of this is 

 thickened, and secretes both the epidermal and columnar layers. 

 From this edge to the pallial line is secreted a layer of nacreous 

 material which may be called the extra-pallial layer. If a section 

 of any thick-shelled species be made, it can easily be seen that the 



