W. M. BALE ON SOME OF THE DISCOID DIATOMS. 27 



extreme edge, and the areolae are of an appreciable size throughout, 

 while the border is not thinned away, so that on focusing the 

 margin there is visible a distinct double contour, with the walls of 

 the last row of cellules showing as coarse transverse striae. 



Several species exhibit a tendency for the border to become 

 wider in proportion as the valves are smaller. C. obscurus and 

 V. apiculatus are instances of this. In both these species I have 

 traced a series down to forms with wide borders, which are 

 only to be distinguished with difficulty from C. marginatus. 

 In Nottingham and other American deposits such forms of 

 C apiculatus are common, and one of them figured by Schmidt 

 (PI. 62, f. 11, 12) has been referred by Rattray to C. marginatus. 



In several species of the Radiati the angles of the areolae often 

 tend to become thickened, so that in a certain focus there appears 

 to be a bead at each angle. This feature has no specific im- 

 portance, and I agree with Rattray that the presence at each 

 a,ngle of a distinct spine, as occasionally found, is of no greater 

 consequence. 



I have already referred to the close affinity which exists 

 between the Excentrici and the Fasciculati, e.g. between C. ex- 

 centricus and C. subtilis. Grunow mentioned this affinity, but 

 Rattray says that it is remote. Grunow's view is undoubtedly 

 correct. In a typical G. excentricus there is a central cellule, 

 and surrounding it a circle, generally of seven. Each of these 

 seven is the centre of an arcuate line of cellules, extending to 

 the margin on either side, behind which is a succession of similar 

 arcuate series, so that the whole of the cellules may be regarded 

 as forming seven fascicles, crossing each other symmetrically, so 

 that no division-lines exist, and for the most part each cellule 

 will form part of three different fascicles. In C. subtilis and 

 C. symbolophorus the number of fascicles is greater, and the 

 divisions between them more abrupt, especially in the central 

 part of the valve, so that the fasciculation is more manifest, but 

 even in these forms the fascicles blend towards the margin in the 

 same way as those of C. excentricus. I have seen a frustule of 

 the latter species in which one valve was normal, while the 

 other was far more finely marked, and was as distinctly 

 fasciculate as C. subtilis. 



I should mention that the C. subtilis referred to is Grunow' 

 typical form, which is quite different from Rattray's, though 



