62 THE STRUCTURES OF PITCHER PLANTS. 



digest small vertebrata in its native haunts. The size of the glands, their 

 number, and relative area are all less in this zone than in those below, 

 because its digestive power would be less frequently called iSto action 

 than would be that of the others. 



The complete covering of the glands in this zone may be of advantage 

 in protecting them and their secretion from accident and the depredation 

 of insects, for the glands here are much more likely to be uncovered by 

 water than those further down. I think it also very likely that these 

 hoods store up the digestive principles of the pitcher until they are 

 required or until it is washed out by the contact of water, it being 

 retained in their cavities by capillary attraction. 



In the second zone the glands gradually alter from a round shape 

 to an oval one, increasing at the same time in size as they are viewed 

 from above downwards, and they become less covered by the hoods. The 

 relative gland area is also greatly increased. The greatest amount of 

 work would necessarily fall on the third zone, so we have here the glands 

 at their maximum, and almost uncovered by the hoods, which still 

 remain in existence, however. The glands are so large and so close that 

 the bulk of the surface is occupied by them. 



The epithelial surface of the pitcher (N. Rafflesiana) is composed 

 of irregularly polygonal cells, which, in the young pitcher, are regularly 

 nucleated. Between these run continuous interspaces, forming a perfect 

 network over the whole surface. 



On the surface of the hoods the cells are elongated, and the spaces 

 run up between them, at right angles, to the lip of the hood, where a 

 canal seems to run, into which they all enter. These intercellular 

 spaces are truly walled on the upper surface, whether they may be or 

 not on their surfaces next the cells ; for, in a mature pitcher, recently 

 fed and perfectly fresh, I am quite satisfied they are slightly raised above 

 the cellular surface. In a virgin pitcher they are scarcely perceptible, 

 whilst in a mature pitcher they are large and distinct, and especially 

 large in the lowest zone. 



Further, I have on several occasions compared two pitchers as nearly 

 alike a? possible from tbe same plant, one of which I had fed, and the 

 other I had starved, and I have been satisfied that in the fed pitcher the 

 intercellular spaces were larger than those in the one which had not been 

 fed. The stream of protoplasm in these canals can be occasionally seen 

 broken, and in such a state the reality of their canalicular structure 

 can be demonstrated. In one observation which I made on a fragment 

 of a mature pitcher of N. distillatoria which had never been fed, I placed 

 over it a drop of fluid taken from the results of the digestion of albumen 

 in another pitcher. I then saw the canals present slight dilatations here 

 and there, and I was satisfied that some of them underwent general 

 increase in calibre. But in several repetitions of this observation I did 

 not obtain convincing results, probably because it was getting late in the 

 season and the pitchers were inactive. 



In old pitchers the superficial canals sometimes seem to communi- 

 cate with a deeper set. 



Silver and gold staining did not help me in the investigation of these 

 structures. From what I have seen I am inclined to regard these canals 

 as absorbent vessels. 



