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



sive depositions of silica generally take place, which end by 

 obliterating the orifices, and in some cases fill these quite up by 

 dense and projecting masses of silica of a higher refractive index 

 than the substance proper of the surrounding shell, so as to appear 

 as red or pink coloured granules on a greenish ground under the 

 best immersion lenses. 



(£.) That the connective zones or bands of some genera, such 

 as Isihmia, are really and truly perforate. 



(/.) That the so-called " areola*," " beads," " pores," " orifices," 

 " granular projections," "depressions," "hexagons," " moniliform 

 dots," "puncta," etc., of authors are all one and the same thing 

 under varying microscopical interpretations, idiosyncrasies, or pre- 

 conceived ideas. 



VII. 



If I may be allowed the free expression of my opinion on the 

 vexed question of the real structure of the Diatom valve, I would 

 say that I attribute to diffractive images the existence of the 

 appearances seen by many observers in the Diatom valve, and that 

 I further believe the perpetuation of erroneous views on the subject 

 resides in the nearly insurmountable difficulty which exists of 

 rightly interpreting, or, in other words, of reducing to their true 

 meaning, optical phenomena of a most bewildering and complex 

 character. 



Prof. Abbe's statement should never be lost sight of by the 

 Diatomo-microscopist when he tells us (and proves it) that what 

 we actually see through our instruments in looking at very minute 

 objects is not what the eye should see, but a something generally 

 very different indeed from it. 



VIII. 



HISTORICAL. 



As over two hundred different papers have been published on 

 the structure of the Diatom valve, I hope I may be excused if I 

 confine myself in these pages to a few words regarding what has 

 been written on the subject of my present communication, and if I 

 limit what I have to say to the history of the valve of Triceratium, 

 which, of all others, I consider the easiest of demonstration and of 

 verification, and which may, I think, be taken as typical of all 

 Diatom structure in general. 



Omitting mention of work done more than fifteen years back, we 



