of Eclinhurgli, Session 1883-84. 
385 
other glands of Nepenthes (lid-glauds as well as peptic ones) 2^ro- 
jecting entirely free from the surface, covered in though they may 
be by a pocketdike flap, or sunk at the bottom of a surface depres- 
sion. In their immersed character, the marginal glands are perfectly 
comparable to the immersed glands of the secreting surface of the 
pitcher of Ceplialotus. As to the function of these marginal glands, 
I cannot as yet speak definitely. Sir Joseph Hooker says (Belfast 
Address) that the pitcher-margins of Nepenthes always secrete 
honey; but from his making no allusion to these very remarkable 
glands, I am doubtful whether he refers to them or not. Probably, 
however, they are honey-secreting, and afford to the insect the last 
drops, just as it is on the brink of destruction. 
C. The Conducting surface. — This agrees essentially with that in 
the adult forms. I have here to note that each crescentic ledge 
consists of a single semilunar cell, which overlaps a lower and 
smaller cell. Occasionally these two cells somewhat puzzlingly 
resemble deformed stomata, but I have not as yet been able to 
trace a more definite relation in this direction. 
D. The Secreting or Digestive surface. — The e23idermis cells here 
are remarkable for their wavy outlines, differing from the more 
angular form exhibited in the adult plants of the parent forms. 
The glands, moreover, are in the first-formed leaves entirely 
exposed, although in the later ones the rudiment of the protective 
pocket-like flap may be seen. In this connection Dr Macfarlane 
has pointed out to me that in N. phyllamxjhora the ^^eptic glands 
are only to a very slight extent covered by flaps, so that in this 
plant we have the persistence of an embryonic character. In the 
first pitchers of the seedling the peptic glands are few in number — 
about ten ; but in the later developed ones they become very 
numerous. 
