266 PROCEEDINGS OF THE 



as : (1) hexagonal structure ; (2) small dot-like structure, or 

 (3) linear markings difficult to resolve into dots, as in Pinnularia. 

 Although these variations appear at first sight to be distinct 

 types, Mr. Brown is disposed to regard them as modifications 

 of one type. Broken pieces of valves are more useful for 

 investigating structure than complete ones, because, where the 

 shell has broken across the cells, the fractured edges of the 

 membranes closing the cells can be clearly seen when properly 

 focused, thus demonstrating decisively that the cells (or so- 

 called dots) are not perforations or holes through the shell 

 as they are commonly but erroneously supposed to be. The 

 structure of Coscinodiscus oculus iridis was explained with the 

 aid of lantern slides of a broken specimen and a section of 

 Mors Cementstein, in which the diatom occurs. The valve was 

 shown to consist of hexagonal cells closed by siliceous membranes 

 at both ends. The partitions between the cells are thickened 

 and slightly projecting at the outer surface, and the " eyespots " 

 in the lower membrane are shown in the sections to be closed by 

 a drum-like structure. When the focus upon a diatom having 

 hexagonal cells is lowered below the upper surface, appearances of 

 the side walls are seen that are considered to be only difiraction 

 efiects, but Mr. Brown thinks it possible that they may be very im- 

 perfect views of the structure of the cell walls blurred by diffraction. 

 For upon examining the scar left on the inner wall of the valve 

 by the breaking ofi of the partition walls, he finds it to be formed 

 of contiguous dots, as if the walls were formed of rods of silex 

 cemented together. Photographs of C. asteromphalus were 

 shown next and the structure explained. The lecturer did not 

 consider either the " white-dot " or the " black-dot " image to 

 be the true focus. There has been much controversy concerning 

 the nature of the so-called dots seen upon diatoms. After much 

 careful examination, Mr. Brown has come to the conclusion that 

 most, if not all, dots are cells similar to those of the Coscinodisci, 

 but without an eyespot ; and that they are not holes, but are 

 covered by a membrane. Isthmia enervis and Navicula lyra were 

 taken as examples and explained with the help of photographs 

 and diagrams. The bead-like appearance which is often men- 

 tioned in books he ascribed to faulty illumination. In the case 

 of N. lyra the true focus (for which Mr. Brown suggested the term 

 " structure-focus ") was to be found between the " white- 



