DIATOMACE^E 609 



others a few of the central areolee are the largest, and the rest are 

 of nearly uniform size ; while in others, again, there are radiating 

 lines formed by areolee of a size different from the rest. Most of 

 the species are either marine or are inhabitants of brackish water ; 

 when living they are most commonly found adherent to seaweeds 

 or zoophytes ; but when dead the valves fall as a sediment to the 

 bottom of the water. In both these conditions they were found by 

 Professor J. Quekett in connection with zoophytes which had been 

 brought home from Melville Island by Sir E. Parry ; and the species 

 seem to be identical with those of the Richmond earth. The in- 

 vestigations of Mr. J. W. StepheAson l on Coscinodiscus oculus iridis 

 si LOW that the peculiar * eye-like ' appearance in the centre of each 

 of its hexagonal areolse arises from the intermingling of the mark- 

 ings of two distinct layers, differing considerably in structure, the 

 markings of the lower layer being' partially seen through those of 

 the upper. By fracturing these diatoms Mr. Stephenson succeeded 

 in separating portions of the two layers, so that each could be 

 examined singly. He also mounted them in bisulphide of carbon, 



FIG. 455. Structure of siliceous valve of Coscinodiscus oculus iridis : 1, hexagonal 

 areola of inner or ' eye-spot ' layer ; 2, areola of outer layer. 



the refractive index of which is high ; and also in a solution of 

 phosphorus in bisulphide of carbon, which has a still higher refrac- 

 tive index. If we suppose a diatom to be marked with convex 

 depressions, they would act as concave lenses in air, which is less 

 refractive than their own silex ; but when such lenses are immersed 

 in bisulphide of carbon, or in the phosphorus solution, they would 

 be converted into convex lenses of the more refractive substance, 

 and have their action in air reversed. Analogous but opposite 

 changes must take place when convex diatom-lenses are viewed first 

 in air, and then in the more refractive media. Applying these and 

 other tests to Coscinodiscus oculus iriclis, Mr. Stephenson considered 

 both layers to be composed of hexagons, represented in fig. 455 

 from drawings by Mr. Stewart. The upper layer is much stronger 

 and thicker than the low^er one, and the framework of its hexagons 

 more readily exhibits its beaded appearance. The lower layer is 

 nearly transparent, and but little conspicuous when seen in bisulphide 

 of carbon, except as shown in the figure, when the framework of 



1 Monthly Microscopical Journal, vol. x. 1873, p. 1. 



R B 



