586 
Journal of Agricultural Research 
Vol. XI, No. ii 
the fungus off from connection with other cells of the host. The fungus 
then dries up and dies or remains dormant in the dead host cells. The 
resistant flax plant behaves differently toward F. Uni than does wheat 
toward P. glumarum. The rust fungus is an obligate parasite and is not 
capable of growing in dead tissues; F . Uni may grow either as a parasite 
or as a saprophyte, and for this reason its development would not be 
checked by the death of the host cells. Furthermore, the breaking 
down of the cells in the flax plant is not so complete as that stated for 
wheat by Marryat and Ward, and would doubtless play very little part 
in decreasing further invasion by the fungus. 
(2) In some cases the protoplasm of the host cells immediately sur- . 
rounding the point of invasion becomes granular in appearance and 
stains green with the Pianeze stain, whereas the protoplasm of the normal 
cell fails to take the green at all. The writer is unable to offer any defi- 
ite explanation for this change other than to say that it is possibly a 
chemical change in the host protoplasm excited by the presence of the 
fungus. There is doubtless a certain amount of injury to the host pro¬ 
toplasm, which may account in part for the coarse granular condition. 
Furthermore, as will be mentioned later, there may be some substance 
produced which is injurious to the fungus. 
(3) Surrounding the area in which the cells show the granular appear¬ 
ance, there is a stimulation to cell division. This cell division is more 
abundant toward the vascular system from the point of invasion. In 
some cases the dividing walls are formed more or less irregularly, while 
in other cases a typical cork cambium seems to be formed. The newly- 
formed cells are to all appearance cork cells. 
(4) Accompanying this cell division and other phenomena is a thick¬ 
ening of cell walls, which is much more noticeable toward the vascular 
system from the invaded point. This thickening of walls may extend 
three or four cell layers beyond the point of invasion and is more pro¬ 
nounced with newly formed cells. However, the walls of cells which 
were formed previous to invasion may become thickened. The process 
of thickening seems to be a laying down of additional material which is 
produced by the protoplasm of the affected cells. The modification of 
old walls is more noticeable below and above the point of invasion and 
toward the epidermis, where cell division is not abundant. These 
thickened walls stain green with Pianeze stain, which fact indicates that 
they are either lignified, cutinized, or suberized. In parenchyma tissues 
of this kind we should hardly expect to find lignin or cutin. When 
treated with concentrated potassium hydroxid, these walls gave the 
typical yellow reaction for suberin, which confirms the conclusion that 
they are of a corky nature. 
Taking into consideration the above-mentioned phenomena with other 
possibilities, it seems that a combined explanation might be offered for 
the resistance of flax to extensive invasion by F. Uni. In the first place 
