446 TRANSACTIONS LIVERPOOL BIOLOGICAL SOCIETY. 



Tlieir position, therefore, becomes very apparent if we 

 allow the other portions of the wall to swell up. 



The innermost of the three layers, of which the cell 

 wall is composed, does not apparently take any part in the 

 formation of the pit, except by its being interrupted at 

 these points. 



The large pits of the central conducting cells have on 

 either side a cap, which is most likely of protoplasmic 

 origin. The cap is a short cylinder, the one open end of 

 which (fig. 12, 13) overlies the pit membrane, with which 

 it is co-extensive. At its otlier end the cap is closed, a small 

 depression being noticed in the centre of the wall. It is at 

 this depression that the cytoplasm is most firmly attached to 

 the cap. This depression corresponds with the thinnest 

 portion of the pit membrane. In younger cells nearer the 

 growing point the caps on the sides of the pits in the 

 central tissue are not so marked. It is from observations 

 made in such parts that the protoplasmic origin of the 

 cylindrical caps is made likely. The sides of the cylinder 

 are seen to be continuous with cytoplasmic strands. They 

 seem, in fact, to be hardened portions of the 

 cytoplasm. 



Owing to the complete absence of any hard woody tissue 

 in the thallus of Chondrus crispus, it seems very probable 

 that these hard caps have the important function to per- 

 form of preventing the collapse or closing up of the open- 

 ing on either side of the pit. 



The cells of the collecting tissue usually have smaller 

 pits, which may or may not be devoid of any cap-like 

 structures. The pits connecting the assimilating cells are 

 usually quite unprotected, but nevertheless form clearly 

 marked thinner portions in the separating cell wall. 



The Plastids met with in Chondrus crispus occur in two 

 different forms, namely, as rhodoplastids and as leuco- 



