666 Transactions of the Society. 



hexagonal cells observed in the surface view is therefore a chamber 

 in the shape of an hexagonal prism ; only the base of the same, 

 the outer surface, is not complete, but has the well-known and 

 often-described circular opening. On the vertical walls — namely, 

 the side faces of the prisms — in the corners of every third chamber 

 is a spine. If the thicknesses of these fine membranes are to be 

 measured, very choice sections must be taken. The basal mem- 

 brane has a very fine sculpture as illustrated by Weiss, but much 

 better and more accurately given by Miiller. It consists of a 

 system of fine dots, radiating in lines from the central space of the 

 entire valve, and of which 60-80 are contained in an hexagonal 

 cell. Only on the finest sections with the highest power one 

 catches a glimpse of these points. I cannot say that the line would 

 thus be seen distinctly beaded ; and equally invisible are the points 

 adhering at the inner side only of the basal membrane, as Muller 

 represents it in transverse section (his fig. 11). Possibly these are 

 also chamber-like cavities, as with Pleurosigma. The definition 

 will be very difficult, but possibly casts will help us hereafter. 



We have still to consider the margin of the valve and the three 

 horn-like protuberances. Muller has spoken of these explicitly, but 

 I find some variations. With regard to configuration of the margin, 

 Muller gives correctly (his fig. 9) the view of the surface of the 

 marginal line. The transverse section (his fig. 11 d) represents 

 this line too much bent inwards. I have therefore given a diagram, 

 plate IX. fig. 2 1, of one of my transverse sections. The line is nearly 

 vertical to the membrane surface and is slightly broader at the top, 

 not pointed. The three protuberances at the three corners of the 

 triangle have been correctly rendered by Muller ; but everybody who 

 has not seen them in sections will find it difficult to realize them 

 in his figs. 6 and 7, because they exhibit too much of other detail 

 and shadow. For this reason I give the diagram of a true vertical 

 section of such a protuberance, fig. 22. According to Mtiller, 

 these protuberances might possibly be open at the point, for which 

 I have not found the slightest justification. Nor is his cited ex- 

 ample of Ewpodiscus a proof, in my estimation, because I see the 

 points in this group also closed. 



The method of collodion casts confirms in general the results 

 obtained by the section method. The collodion enters through the 

 outer larger opening into the chamber, and leaves on the surface of 

 the dry cast a somewhat round mass which has very little likeness 

 to the beautiful regular prism of the chamber. On closer examina- 

 tion one distinguishes these spines in the cast as distinct depres- 

 sions in the corners between the masses. The cast of the inner 

 side of a valve is perfectly smooth. I could not succeed in obtaining 

 a cast of the delicate porous structure, either because it does not 

 exist or because the collodion used showed reticulation in hardening. 



