70 
W. B. McDOUGALL 
of the cell sap pass from cell to cell through the walls, the fungus 
has an opportunity to obtain food from that source. 
We are forced to the conclusion, therefore, that the tree is not benefited 
by association with the fungus, and that the ectotrophic mycorhizas are 
not symbiotic associations, but are instances of the parasitism of fungi on 
the roots of trees. Usually this parasitism is comparatively harmless, 
although the fungus kills all rootlets which it attacks. On the average 
tree only a small percentage of the roots are transformed into myco- 
rhizas, since the mycorhizas are present only in the superficial layers 
of the soil and all the deeper roots are allowed to function normally. 
For this reason the tree can endure the mycorhizas without any serious 
inconvenience, just as it can endure a considerable number of insect 
galls on its leaves. In the case of seedlings, however, the mycorhizas 
may become quite serious. The roots of a seedling are all in the 
superficial layers of the soil and, therefore, subject to the attacks of 
fungi. If all the rootlets are transformed into mycorhizas, the supply 
of mineral salts will be cut off from the seedling and it will be killed. 
Such a case has actually been reported by Nadson (15) . The seedlings 
of Quercus studied by him were dying off, and he was unable to find 
any possible reason for their death except the ectotrophic mycorhizas 
which were present in unusual abundance. 
In the case of endotrophic mycorhizas the situation is more com- 
plex, because the mycorhizas of different species of plants differ so 
greatly. No doubt an important reason why so many hypotheses 
have been offered to explain the physiological relations of endotrophic 
mycorhizas is that the different authors worked with very different 
sorts of plants, and the same hypothesis could not apply to all. 
Bernard (i) and Burgeff (2) have shown conclusively that the seeds of 
certain orchids will not germinate except in the presence of their 
mycorhizal fungi. This fact, of course, proves the symbiotic nature of 
the orchid mycorhizas studied by these men, but it would be unsafe 
to draw any conclusions from this with respect to other families of 
plants, or even to all orchids. Magnus (12) demonstrated just as con- 
clusively that in the endotrophic mycorhizas of Neottia certain cells 
of the root digest the fungus hyphae, and so derive benefit from the 
association. Shibata (22) later found the same thing to be 'true of 
Podocarpus mycorhizas. Groom (7), who did his work on Thismia 
a few years earlier, failed to find that the hyphae were digested by 
the host cells, but he did find evidence that the root obtained food 
