Structure of the Diatom Shell, &c. By Br. J. B. Cox. 399 



and 3rd. That the film may be of such tenuity that not even its 

 broken edge refracts the light sufficiently to make it visible. The 

 first of these answers will be a satisfactory one in many cases, and 

 does not need to be enforced by much argument. The second is illus- 

 trated by the condition in which the finely dotted film of Heliopelta 

 is often found. When it is separated from the heavy sub-hexagonal 

 network of the alternate segments of the valve it shows no mark of 

 attachment to these, but is as free from any trace of their outline 

 as if it had never been in contact with them.* It is quite possible 

 that the outer film of Triceraiium may have a similar loose 

 connection with the hexagons below. 



It is, however, to the third of the suggested answers that I 

 wish to call more particular attention ; for I have myself been 

 surprised at the conclusiveness of the proof that in microscopical 

 work we have frequently to deal with laminae so thin that our 

 widest-angled glasses entirely fail to detect the outline of a broken 

 margin. 



The example which I have selected to illustrate this is Isthmia 

 nervosa Kiitz. In a former study of this species I noted the 

 fact that the alveoli sometimes show a notched margin where 

 connected with the valve, as if teeth from the thicker part ran 

 down upon the thinner film covering the circular or oval space of 

 the alveoli. I also noted the occasional occurrence in the hoops of 

 this diatom, of a very thin lamina extending beyond the thicker 

 band, as if the suture which usually divides these hoops trans- 

 versely had been forcibly broken before it was quite ripe, and 

 part of the band had been split off in a very thin wedge-shaped 

 film.t The remembrance of these phenomena has led me to a 

 further investigation in reference to the question of the visibility of 

 these films, as they offer a convenient series of varying thickness. 



The hoop of Isthmia is a smooth band of silex in which the 

 alveoli are nearly circular, and showing, usually, no evidence of a 

 film covering the apparent opening, except in the increased colour 

 which is so shght as to be of no great weight as proof The rim of 

 each circlet shows a bright white light which gives the whole an 

 appearance very similar to that of the openings in the reticulated 

 Polycystin^e. But occasionally the bright rim of light is broken 

 by four or more pink bars which connect the central portion of the 

 circle with the surrounding lamina of silex at diflerent points in the 

 periphery, and in still more rare examples this central part of the 

 circle is both thickened and mottled till the visibility of the film 

 admits of no dispute. 



The alveoli of the valve are of an irregular oval shape, and in 



* Cf. Photograph No. 27 in tlie series before referred to. 



t Amer. Journ. of Microscopy, iii. (1878) pp. 97-101, 125-30. 



