192 



dactylifera in which the wall has been greatly swollen and the plasmodes- 

 men stained to make them more plainly visible. In this case I treated 

 the specimens according to Gardiner's method first by allowing them to lie 

 for a while in iodine and potassium iodide and then adding chloriodide of 



Fig.l. 



zinc and allowing it to act for twelve hours. The sections were then 

 carefully washed in water. The walls were found to be strongly swollen 

 to at least twenty-five times their original thickness where the pits oc- 

 curred. In only a few instances were any of the aggregate plasmodesmen 



Fig. 2. 



found broken where they entered the pit of the cell. Sometimes this 

 occurred, especially in the solitary plasmodesmen, and a small filament 

 could be seen as in Fig. 2. Attempts were made to bring these plas- 

 modesmen more plainly to view bj' the use of Hoffman's blue, but this did 

 not succeed very well. I found after considerable experimentation that 



