398 Transactions of the Society. 



IX. — Structure of the Diatom Shdh Siliceous Films too thin to 



show a broken edge. By Jacob D. Cox, LL.D., F.R.M.S. 



(Bead 13th May, 1885.) 



In the current discussion of tlie structure of the diatom valve, an 

 interesting question has been started regarding the possibility of 

 proving the actual jtrescnce of lilms of silex, whose tenuity is so 

 great that they are not visible by ordinary transmitted light. Many 

 observers have questioned, for instance, the existence in Triceratium 

 favusoi an outer film covering the hexagons on the convex surface 

 of the valve, and others have denied the existence of a similar film 

 covering either the " eye-spots " or the hexagons in Coscinodiscus 

 ocidiis-iridis, &c. 



In former ])apers on tliis subject, in which I have maintained 

 the presence of such films, I have relied mainly upon what seemed 

 to me the very conclusive evidence obtained from the examination 

 of the diatoms as opaque objects, either with ordinary powers or 

 with high powers aided by the vertical illuiuinator.* A thin homo- 

 geneous film of silex of even thickness, so little affects the passing 

 light, that the test of colour alone is hardly sufficient to detect its 

 presence in silicified membranes so extremely delicate as those I 

 have mentioned. In valves broken across the alveoli there is no 

 great difficulty in demonstrating, by photography, the presence of 

 the outer (convex) film in the Coscinodiscus and the inner 

 (concave) one in the Triceratium. '\ These are made visible by 

 the dotted markings upon them, and they have, moreover, in the 

 stronger examples, a sufficient thickness to present a visible broken 

 edge, and to make the colour test of value in dis^criminating the 

 film. But in the case of the outer (convex) film of the Triceratium, 

 and that covering the " eye-spots " of the inner (concave) one of 

 the Coscinodiscus, whilst 1 insist that both by colour and by the 

 fractured edge they may sometimes be detected, I fully admit that 

 by the latter test (the discrimination of a fractured margin) their 

 presence cannot commonly be proved. 



If the evidence gained from the examination of the frustules as 

 opaque objects shall be recognized as conclusive (as I certainly 

 regard it), the question would be simply, How can we account for 

 the apparent emptiness of the areolae in nearly all the cases in 

 •which these thinnest films are broken across ? Three answers may 

 be given : 1st. That so fragile a film may have been destroyed by 

 the force which fractured the valve ; 2nd. That the film may have 

 been so imperfectly attached to the hexagon walls (in the case of 

 the Triceratium) as that when fractured it floats off or is removed ; 



* See this Journal, iv. (18'84) p. 941. 



t Cf. Broken Sliell Seri. s, in the Societ>'.s Library, \c9. 21-3. 26, 63, 95. 



