J. DEBY ON THE STRUCTURE OF THE DIATOM VALVE. 313 



this membrane, which is generally thinner in the middle portion of 

 the areola?, does really occasionally become highly silicified, and 

 may support particles or granules of highly refractive silica placed 

 over the so-called " eye spots," in which case the cavities must be 

 hermetically sealed on both sides to all but osmotic influences. 



(e.) That the lower closing membrane of the areolae frequently 

 carries various designs, the nature of which, on account of their 

 minuteness, has not yet been well established, but which must 

 depend upon structure, as no diffraction images produced by any 

 organisation lying at a lower level can be the cause of them, as no 

 such lower organisation exists below this bottom or closing internal 

 diaphragms. 



(/.) That the thin upper membrane of the areola? forms the 

 extensions of the edges of the so-called " nail-headed " bars which 

 form the limiting walls of the areola? as figured by Otto Miiller, by 

 Dr. Flogel, and by Messrs. Prinz and Van Ermengem.* 



(g.) That the cavities in the valve are bounded by walls of solid 

 silica. That these walls often extend beyond, above, or below the 

 closing membranes of the areola?, and that they frequently run 

 into points or spines of various shapes and lengths, which project 

 beyond the valve between the areola?. 



(h.) That the median slit or fissure, which is observed to run 

 through the rachis, or thickened median line of most of the 

 Navicular, is also closed top and bottom by a very thin organic 

 slightly silicified membrane in recent normal valves. I believe, how- 

 ever, that minute apertures may exist in these narrow closing 

 membranes in the neighbourhood of the central and of the terminal 

 nodules, but this is a subject requiring further elucidation. 



(j.) That the so-called " secondary ' or internal valves — 

 " Regenerationshiille " — of some Diatoms do not exist in the 

 very young valves, a fact which gives us the reason why the 

 frustules, which are formed of an old and of a younger valve, 

 generally split up into an odd number of secondary valves, either 

 three, or five. It is my belief that the young secondary valves 

 are always perforate at first, but that as they grow older siicces- 



* In most fossil diatoms and in nearly all specimens boiled in acids, the 

 external film closing the areola has disappeared, and the valve has in con- 

 sequence become really perforate on the upper surface. In some cases the 

 lower plate has also ended by presenting orifices, so that the sections 

 examined by the above-named microscopists really showed what they have 

 figured in the plates accompanying their various papers. 



Journ. Q. M. C, Series II., No. 16. b b 



