X. E. BROWN OX THE STRUCTURE OF DIATOMS. 323 



covering the meshes of the grating. This membrane is extremely- 

 thin, probably not thicker than the film of a soap-bubble, and 

 is raised into a slight dome or convexity over each mesh of the 

 grating. When the apex of these convexities is accurately in 

 focus, and the headings of the bars seen as black dots forming 

 squares, the light being central and with a magnification of 

 2,000 to 3,000 diameters, a very minute dot is seen at the 

 centre of each square (PI. 2, fig. 1). This central spot I conceive 

 to be a true pore through the membrane ; it is very minute, at 

 the most not more than one-third of the diameter of the black 

 dots themselves, and is probably not more than 1/200, 000th of an 

 inch in diameter. It is not quite easy to see, but can be made 

 clearer by the use of a small central stop in the substage con- 

 denser. I doubt if it can be seen at all at a less magnification 

 than 1,000 diameters; and with a dry Zeiss y-th, at a magnification 

 of 3,000 diameters, I do not feel quite sure that I see the pores. 

 There seems a suggestion of their presence, but I do not think 

 any dry lens will show them very clearly. Under dark-ground 

 illumination, with a Leitz dark-ground illuminator, the bars 

 are white and the headings on them appear much larger than 

 when seen by direct light, whilst the membrane is not seen at 

 all, the spaces between the bars being black. But if the funnel- 

 stop which cuts down the aperture of the lens is removed, the 

 illumination remaining as before, then the bars appear to be 

 very slender and black and the membrane whitish, with the 

 minute pores clear and distinct. 



Upon entire specimens of the diatom the pores are difficult to 

 see, apparently owing to the convex curvature of the shell, but 

 with a little trouble I have been able to see them in places upon 

 every specimen examined. Under certain conditions of illumina- 

 tion a small dark spot, which might easily be mistaken for the 

 pore, is seen at the centre of each of the beads of the membrane ; 

 this spot, however, is very much larger than the true pore, 

 and appears to be some diffraction image, possibly that of the stop 

 in the condenser, as can easily be demonstrated by moving the 

 mirror slightly, when the spot is seen to shift its position. 



Although all to whom I have shown these pores agree with me 

 that they are very minute, yet they appear to have a different 

 size to different observers. To my eye they appear to have about 

 the proportion to the black dots I have represented in my drawing, 



Journ. Q. M. C, Series II. No. 74. 23 



