88 
characters as thickness of cuticle, the number of stomata per 
unit area, the hairiness of the leaves, etc. I may also add that 
some of the most susceptible wheats to yellow rust are charac. 
terized by their early habit of ripening. The factors which de. 
termine whether a plant will be immune or susceptible to any 
particular fungus are at present beyond the reach of our powers 
of observation. They are intra-protoplasmic and infection ex. 
periments alone can decide the matter. Where such infection 
trials are made with the uredospores of a rust they are found to 
push out a mycelium into a stoma, which develops rapidly in 
the intercellular spaces and drives its haustoria into the cells of 
the susceptible host plant. In the immune plant the mycelium 
fails to develop to any extent in the intercellular spaces and 
finally it dies away altogether. At the outset the possibilities 
of infection are the same for each, but the fungus fail to over. 
come the resistance of the cell walls and protoplasm of the 
immune plant and consequently dies for lack of food before it 
can obtain a footing. Why it fails we cannot say and can only 
ascribe it again to the constitution of the host plant, leaving 
what are probably complex problems in the metabolism both of 
the host and parasite for the future to decide. It may be, as 
has been suggested, that antitoxins in the host and toxins in 
the parasite are the deciding factors. In this connection we 
may note in passing that claims have been made, which require 
supporting, that certain plants have been rendered immune to 
the attacks of their usual parasite by being watered with solu- 
tions of copper sulphate. 
The next portion of the subject which requires investigation 
is the external conditions responsible for epidemics, for rust- 
years, for potato-disease years, etc., for knowing what to expect 
it may be that precautions can be taken for checking their 
ravages. In certain cases we know something of these external 
conditions. Thus it is now universally agreed that epidemics of 
potato-disease (Phytophthora infestans) occur when moist, dull 
weather, and consequently a water-saturated atmosphere, follow 
a forcing period of growth in July and August. Such condi 
tions one can readily realize are peculiarly favourable for the 
rapid development and distribution of the zoospores of the 
fungus, whilst it is possible that they, to a certain extent, break 
down the host-plant’s capacity for resistance by disturbing 1ts 
normal economy. One need only remember that under such 
conditions the plant would tend to become water-logged, owing 
to imperfect transpiration, to see how such disturbances would 
be possible. That the same external conditions are favourable 
to the spread of all parasitic fungi is highly improbable. 4 
rust-epidemic and a mildew-epidemic are not necessarily found 
in the same season on a barley crop, though these fungi make 
