Tlie Structure of Diatoms, By Dr. J. H. L. Flogel 521 



perfection which distinguishes my transverse sections, although 

 it shows nearly all one can reasonably expect. Putting trans- 

 verse and longitudinal sections together, one readily sees that the 

 former must look a little ditferent when they are cut through the 

 elevations or when through depressions. Pfitzer has already drawn 

 attention to this fact, and as far as I can judge he has deduced it 

 from the optical transverse sections of the raised frustules (p. 109). 

 The real transverse sections confirm his view. It is not uncommon 

 that the section is so thick, that lying at the declining edge of a 

 wave its outlines in the upper part diSer from those in the lower. 

 Figs. 13, 15, and 16 illustrate this by finely drawn lines. These 

 are difi'erences which occur with the transverse section of the 

 wings. The section goes either through a place where the mem- 

 branes holding together the wing He closely one upon another, 

 exhibiting the image fig. 14 ; or else it goes through the inter- 

 vening space, then the wing looks like a flat-pressed smooth 

 surface having a lumen in open communication with the cell 

 (fig. 15). In the former case one observes at the highest margin of 

 the wing the transverse section of the extremely thin tube above 

 mentioned. Pfitzer (vide p. 1 1 0) has expressed the opinion that along 

 the entire wing-margin runs a fine cleft, or that there exist a large 

 number of extremely small openings standing in one line. This 

 cannot be taken for more than a mere opinion. My transverse 

 sections in no way corroborate this opinion ; on the contrary, they 

 show these fine marginal tubes closed everywhere outwards. These 

 details can be best understood by comparing it with the surface view 

 of the wing, fig. 1 8. By comparing the figured transverse sections 

 together it will be seen that the proportion of the size of the 

 wing to the surface towards the end is different from what it is in the 

 middle of the valve. We see further that there is a difference in 

 the curvature of the surface of the valve, about which more further 

 on. With regard to the finer sculpture, the transverse sections 

 exhibit the midrib as an irregular thickening; one sees there a 

 point. However, the membrane in its whole extent is so extremely 

 delicate (the measures give 0'4-0'5/i, even this is too high) that 

 it becomes very difficult to distinguish difi'erences of thickness. 

 Transverse ribs are delineated in the longitudinal section (figs. 19 

 and 20) clearly like small ovals on the crests of the waves, and these 

 ovals are mostly more pointed whilst the valleys are rounded. I 

 see the transverse striae in the delicate longitudinal sections as clear 

 pearl-like punctures, as in fig. 20. No rib-like projection can be 

 observed, however, for the shadow permeates the v/hole mass so 

 that it must necessarily be caused in a manner similar io Pleurosigma. 

 The longitudinal striae I could not perceive with the desired clear- 

 ness in the thinnest transverse sections, not even with oblique 

 light. Sometinu'S I observed a kind of glimmer, but nothing 

 yiT. 2.— Vol. IV. 2 N 



