1886.] 



HEPATICAE NATALENSES. 



13 



its habit, with its arcuate branches, thick textiire to the 

 subgenus Lembidium. Dr. Spruce who has given particular 

 attention to the genus Ceplialozia states that the J bracts 

 are monandrous in all Cephalonae; in this species they are 

 uniformly diandrous upon which fact the Dr. remarks ,,there 

 is no rule withont an exception" I do not go so far as to 

 say ,.It is the exception proves the rule" but certainly we 

 ought to note all exceptions, if only as a check to arrogant 

 positivism." 



There are slight differences between Lehmann^s de- 

 scription of his Jung. heteromorpha and mine but as Dr. 

 Spruce states „there is little doubt the plant is that species." 



Description of Tab. VII. Fig. 1. Natural size. 2. Plant 

 magnified. 3. Portion of stem with leaves X 31. 4. Ditto 

 X 24. 5. Cross-section of stem X 85. 6, 7, 8. Leaves X 

 31. 9, 10. Cells of leaves X 85. 11, 12, 13. Underleaves 

 X 85. 14. Bracts and bracteole X 31. 15. First bract 

 X 31. 16. Second bract X 31. 17. Third bract X 31. 

 18. J bract X 31. 19. Ditto X 64. 20. Antheridia X 85. 

 21. Pistillidia X 85. 22. Perianth X 31. 23. Cross-section 

 from the lower part of the perianth X 24. 24. Cross-section 

 from upper portion of perianth X 24. 25. Mouth of 

 perianth flattened out X 85. 26. Capsule X 24. 27. 

 Spores X 290. 28. Flater X 290. 



10. Kantia argiita (N.) 



Hab. On trees by a rivulet near Umpumulo. 



Obs. I refer the specimens to this species, as agreeing 

 nearest in shape of leaves and underleaves, leaves being 

 alway acutely bidentate, not sometimes entire, as in Kantia 

 fissa, underleaves deeply divided, segments divergent, outer 

 margins unidentate, cells equal in size to those of K. 

 arguta, diifering from what European forms I have seen 

 in sometimes having thicker walls. but in this feature dif- 

 ering on the same plant, often thin as in fig. 5 Kantia 

 hidentula (Web.) recorded from South Africa has according 



