400 Transactions of the Society. 



a majority of cases the notching of tlie edge which I have 

 mentioned, indicates, ^vith the lieightened colour of the central parts 

 of the oviil, the presence of a fihn connected with the stronger part 

 of the valve by little teeth which give the rim the notched 

 appearance. But I have found specimens in the fossil deposits of 

 ^anta Monica and San Luis Ohispo, Cal., and Moron, Spain, in 

 which the vigour of growth has been so great that the film 

 covering the alveoli is strongly marked over its whole surface 

 with a well-defined arabesque pattern. Whenever these shells are 

 broken, it is only in the most robust that the edge of the fractured 

 film within the alveoli can be traced. In by far the greater 

 number the space appears to the eye quite as empty as the "eye- 

 spots " in the Cose i nodi sci, and no care in scrutiny can detect the 

 broken margin after it enters the oval or circular space.* Yet we 

 know that the film is there by the manner in which it is incon- 

 testalily presented in the stronger specimens of the same kind. 



The evidence found in the split iilm projecting beyond the 

 suture of the hoop of Isthmia is more direct, and if possible more 

 conclusive. In this, though the siliceous film is so thin that one 

 hesitates to say that it has given even the faintest tint of pink to 

 the background, the outlines of the alveoli are plainly seen near 

 the suture; but these grow more indistinct until they disappear. 

 Beyond them, no marginal line of the film can by any possibility 

 be traced in those cases in which the split film gradually 

 diminishes to a knife edge. The faint circles of the alveoli are 

 seen in their quineuncial order with nothing apparent to connect 

 them with the more solid parts of the shell, and as if they were the 

 ghosts of small circular diatoms regularly scattered in the field.f 

 In such cases I have exhausted patience in toying with the fine 

 adjustment of the Microscoj)e, trying to find some point in the 

 curved surface at which the margin could be traced, but in vain. 

 In examples in which the split film was broken near the suture 

 and where it had appreciable thickness, its edge was apparent 

 enough; but not in cases where it ran out to its greatest thinness. 

 Yet the curved or thickened margin of the alveoli produces effects 

 of refraction which demonstrate the presence of a lamina of which 

 we should have no knowledge whatever by any other means. 



The negative evidence which is found in the apparent 

 emptiness of the circles of broken alveoli is thus demonstrably 

 answered. It is shown that thin films do exist where they cannot 

 be seen of themselves and whose broken edges cannot be traced, in 

 places strictly analogous to tliose of the "eye-spots " in Coscinodiscus 

 and Triceratinm. The affirmative evidence which is found in the 



* Pliotograplis Nos. 101-M, 107, 108. 

 t Pl.ot(.grapli.s Nos. 102, lO:^. 



