14 WEISS, The Parichnos in the Lepidodendracecg. 



open out more and more widely as they approach the 

 leaf-scar. There can be little doubt from their position, 

 that these pit - like depressions correspond with the 

 inferior scars noticed on the impressions of many Lepido- 

 dendron leaf-bases. Now, what is the relation of these 

 pits to the parichnos strands? In the sections in which 

 these pits first make their appearance, they are separated 

 from the parichnos strands by some six or eight layers of 

 undifferentiated cells ; but as soon as they communicate 

 laterally with the exterior, one can see a patch of 

 specialised cells on the wall of each pit (aer., Fig. 3 

 (iv.) of Plate) close to the underlying parichnos strand, 

 and subsequently in contact with it. The glandular 

 appearance which these patches have, is I believe due 

 to defective preservation, though undoubtedly a different 

 explanation might be put upon it. They are covered in 

 by a more or less distinct epidermal layer, though in 

 some cases it is defective. I have not been able to 

 observe any stellate cells such as occur in the transverse 

 sections ; but this may be due to two causes. In the first 

 place we have no certainty that this Lepidodendron is the 

 same species as that of which the transverse sections 

 above are figured and described. Secondly, the transverse 

 sections described above must have passed somewhat 

 obliquely through the aerenchyma, and the stellate cells 

 may on the tangential section of the leaf-base be much 

 less conspicuous. What is, however, obvious on these 

 tangential sections is that the parichnos is separated from 

 the glandular-looking tissue by a layer of small cells 

 which correspond in position with the stellate cells 

 described above. Beyond these small cells the tissue, 

 which, as stated, is very defective, consists of larger 

 elements, or of cells with larger air spaces, but they 

 have undergone so much degeneration that it is impos- 



