366 Dr. O. Schmidt on Coccoliths and Rhahdoliths. 



regarded as such is the actual margin of the medullar space. 

 The granular zone is in course of production, although it does 

 not appear in the whole circumference, but advances round the 

 margin as an excrescence starting from a particular point. 



When the development of the coccolith has taken a direction 

 as in fig. 12, PI. XVI., it appears to be concluded with the 

 granular layer. In all other cases the coccolith usually becomes 

 completed by the formation of a concavo-convex disk, which 

 generally appears homogeneous, rarely irregularly notched and 

 granular, and covers and more or less overarches the parts 

 previously formed. As already mentioned, I have most de- 

 cidedly ascertained that Hackel's so-called outer ring is never 

 in connexion with the outer margin of the granular zone, but 

 is only the margin of a regular shield projecting beyond 

 the granular zone. In every object that I succeeded in turning 

 and placing on its edge, I was able to trace and ascertain this 

 dorsal shield. 



In PL XVI. figs. 16, 17, and in PL XVII. most of the 

 figures show the character and position of the dorsal shield. 

 In figs. 1 and 2 the case is represented in which the central 

 granule has attained a perfectly excessive development. This 

 may furnish the explanation of a case occurring at Brindisi 

 (PL* XVII. fig. 18) in which not only is the central granule 

 elevated, but its base and the part representing the medullar 

 space appears granulated. This coccolith is completed by a 

 strong granular ring. In figs. 3 and 4 (PL XVII.) the central 

 granule is wanting. All these, and the other figures of disco- 

 liths with a dorsal shield (figs. 5, 6, 7, 10, 11, 12, 13, 14, 15), 

 show individually and when compared with each other that in 

 the centre of the convex side of the coccolith the dorsal shield 

 is connected either with the medullar space itself, or with a 

 peg-like prolongation of the part corresponding to the central 

 granule. In the fresh state their union seems to be so firm, 

 and perhaps elastic, that the dorsal shield is scarcely ever 

 broken away. From the deposits at Brindisi, however, I 

 have repeatedly obtained coccoliths like PL XVII. fig. 19, 

 with a regular aperture in the centre of the medullar space, 

 which I imagine to have been produced by the breaking 

 away of the dorsal shield together with its pin. 



From all these observations, therefore, the dorsal shield 

 cannot be otherwise produced than by its gradually over- 

 growing the other parts from the dorsal pole of the coccolith. 

 The observation of such intermediate stages is extremely 

 difficult, because, up to the period when it projects beyond the 

 circumference of the granular ring, it appears to adhere closely 

 to the dorsal surface in the form of a most delicate lamella. 



