part 2] PREGLACIAL FLORAS FROM CASTLE EDEN\ 133 



UMBELLIFERiE. 



Sanicul/g (?) genus (?) sp. 1. (PL IX, fig. 24.) 



Mericarp semieylindrical, narrowed at the apex, stylopod conical, 

 no apparent vittse ; ventral face flat, not grooved ; dorsal face 

 with three ridges scarcely apparent above, but becoming somewhat 

 winged and crested or scaly below ; surface rugose, with very fine 

 longitudinal striations ; in one of the furrows two long hairs arise 

 from the same base, and the broken bases of a few other hairs are 

 to be seen ; it appears, therefore, as if the primary ridges had been 

 scaly and the furrows hairy. 



Length=l"3 mm. ; breadth = "5 mm. 



The characters imply that the specimen belongs to the section 

 Saniculse of the order ITmbelliferae ; but 1 have been unable to 

 identify any genus to which it can be assigned. 



Ammineje (?) genus (?) sp. 2. (PL IX, fig. 25.) 



Mericarp narrow, elongate, inflated (flattened by pressure), 

 contracted to a neck above ; five ribs nearly parallel, dorsal ribs 

 thread-lite, lateral broader, vitta? doubtful. 



Length = 2 - 2 mm. ; breadth— "7 mm. 



The mericarp, though nearly perfect, is distorted and rather 

 difficult of interpretation. I have been unable to identify it. 



Umbelliferae genus (?) sp. 3. (PL IX, figs. 26 & 27.) 



Mericarp narrow, elongate, beaked ; five ribs, three dorsal nearly 

 parallel, two lateral thick (broken), partly covering the ventral face. 



Length =2*2 mm. ; breadth (broken ) = "8 mm. (?). 



Two mericarps, both much broken. I have been unable to 

 identify them. 



ERICACEAE. 



Erica cf. vagans Linnaeus. (PL IX, figs. 28-30.) 



Seed globular, stipitate ; surface of large flat-bottomed or cup- 

 shaped hexagonal cells, the raised margins of which form a strong 

 reticulation over the surface, or give rise to a honeycomb appearance. 

 Diameter =-5 mm. 



In 1910, among the unknown plants from Tegelen, there was 

 figured (fig. 58) a small, globular, strongly-reticulate seed. In 

 ' The Pliocene Floras of the Dutch-Prussian Border ' (1915, p. 132) 

 it was suggested that this seed might belong to Gentianacese, as 

 certain tropical orders have similar seeds. I have now found an 

 abundance of these seeds at Castle Eden. Some show the reticulate 

 appearance characteristic of the Tegelen specimen, but some are 

 not so much inflated, and appear honeycombed ; also some are 

 stipitate. In view of. the light shed on the species by this abundant 

 new material, I think that it should be referred to JEJrica, and is near 



