X. E. BROWN ON THE STRUCTURE OF DIATOMS. 325 



probably because the photographs were taken at a focus where 

 the membrane is not visible and where diffraction effects are at a 

 maximum, whilst at the focus of the surface of the membrane 

 they are reduced to a minimum. 



Since writing the above I have had the advantage of being able 

 to examine a realgar mount of P. balticum belonging to Mr. E. 

 M. Nelson. The realgar of this slide is not nearly so clear and 

 brilliant as that of Mr. CapelTs slide, and on some parts of it I 

 cannot see the pores in the films at all, but there are some films 

 where they can be most distinctly seen. I mention this, because 

 others possessing realgar mounts of this diatom might fail to find 

 the pores on some of the films and believe them not to be present ; 

 they may be extremely difficult to make out, or quite invisible on 

 parts of the valve where both longitudinal and transverse grating 

 or strengthening bars are present. 



Pleurosigma angulatum. Upon Mr. Oapell's realgar slide 

 are also numerous specimens of this diatom ; some are bent or 

 contorted, but otherwise, with the exception of two or three 

 specimens, seem unaffected by the heat or dissolving action. 

 One of these exceptions, however, is an exceedingly interesting 

 specimen, and clearly confirms Mr. E. M. Nelson's statement in 

 the Journ. Q. M. C, Ser. 2, Vol. XII. pp. 98-100, that the valve 

 of this diatom is composed of two gratings. It is a single valve 

 and therefore its structure is not obscured by images from the 

 opposing valve, is fractured in places, and has its outer surface 

 next the cover-glass, as can be verified by examination under a 

 binocular. Over a small area some solvent has caused a portion 

 of the outer grating to peel off, and at one place a small patch of 

 it is seen adhering to the cover-glass ; this patch is represented at 

 fig. 5, as seen when magnified 3,000 diameters. At this magnifi- 

 cation the bars of silex forming the boundaries of the meshes are 

 seen to cross one another diagonally, forming diamond-shaped 

 meshes, and are thickened at the nodes or points of intersection 

 just as in P. balticum, and, as in that diatom, it is these nodes 

 which produce the black-dot appearance. At the centre of 

 the membrane covering each mesh a very minute pore can be 

 seen when the surface of the membrane is accurately in focus. 

 These pores do not seem to be visible under direct central light 

 without the interposition of a stop in the condenser, and I find 

 that they are best seen when illuminated by means of a Leitz 



