THOMAS: INFECTION OF APIUM GRAVEOLENS 25 
established and the size of the spots produced. On the other 
hand top dressing the soil of pots with lime decreases the infection. 
Also the infestation of the roots of celery plants by nematodes 
partially or entirely inhibits the development of the fungus. 
The retention of chlorophyll and water in the tissue adjacent 
to infected spots after these have disappeared from the remainder 
of the leaf is further evidence of a tendency toward mutualism 
between the fungus and host. McCue (22) observed that tomato 
plants treated with phosphatic fertilizers developed less leaf 
blight (presumably Septoria) than control plants, while those on 
nitrogen and potash plots were more heavily infected than the 
controls. At the same time the highest yields (showing greatest 
vigor of growth), were obtained from the plants which received 
nitrogen and potash. Norton (24) also noted a decrease in infec- 
tion by Septoria on tomatoes treated with phosphates. 
Cereals grown by Spinks (37) in nutrient solution seemed to be 
susceptible to infection by Erysiphe in proportion as vigorous 
growth of the host was maintained. Excess of phosphates and 
potash diminished susceptibility while nitrates increased it. 
Stakman (38) noted that heavy manuring of rye increased the 
number of successful inoculations with Puccinia graminis Avenae. 
Even the seed fungus, however, is surpassed in some mutualistic 
characters by the mycorrhizas and lichens. The mycorrhizas 
show grades of interrelation between host and parasite from active 
parasitism to finely adjusted mutualism. Stout (39) has shown 
that Sclerotium rhizodes may be parasitic on aerial parts of Calama- 
grostis canadensis and at the same time assume a mycorrhizal 
habit on the roots of the same host. Nieuberg (23) found that in 
the lichens mutualism persists for a long period perhaps even 
after the fungus finally penetrates the algal cells. This habit is a 
close approach to that of producing haustoria as in the more 
specialized parasites. 
With the exception of the nitrogen fixing bacteria, only the 
mycorrhizal and lichen fungi have been proven to contribute 
anything of value to the host. It is important, however, in the 
development of methods of avoiding disease in plants, as well as 
for a clearer understanding of the nature of parasitism, that it be 
recognized that the relation of host and parasite is not of necessity 
