64 
W. B. McDOUGALL 
effect. Kauffman (9) has shown that the same fungus may cause 
mycorhizas on several hosts, and my own work shows that the converse 
of this is true, at least in the case of ectotrophic species of forest trees. 
All mycorhizal fungi, however, are not capable of forming mycor- 
hizas with all species of mycorhizal trees. Individual trees of common 
mycorhiza-forming species, growing in a soil which is known to be 
inhabited by mycorhizal fungi, are often found entirely free from 
mycorhizas. A specific instance of this may be given here. In a 
woodlot west of Ann Arbor, two trees, a Carya and a Quercus, were 
found growing a few feet apart, and their roots intermingled. On the 
Carya, brown colored mycorhizas were found in great numbers, but 
on the Quercus not a specimen could be found. Apparently the 
fungus which formed the mycorhizas with Carya was not capable 
of forming any with Quercus, though this species readily forms mycor- 
hizas. Probably the following year, or at some future time, a fungus 
capable of forming mycorhizas with Quercus would chance to grow 
here, and this same tree would then have mycorhizas. 
Not only are some trees without mycorhizas likely to be found in 
such a habitat, but on those trees which have mycorhizas, there are 
always found some roots which are uninfected, even in the humus 
layer. This was brought out by my glass plate experiments, and is 
easily verified by digging up the roots of any mycorhiza forming tree. 
The reason for this may be that in the case of the uninfected roots, 
the proper fungus did not happen to be present in the immediate 
vicinity at the time when the root was susceptible to infection ; or 
there may have been some physiological reason why the roots were 
not infected. We have no evidence either for or against the latter 
hypothesis; there is some evidence in favor of the former. In collect- 
ing mushrooms it is often noticed that a great abundance of specimens 
of a certain species may be found in a particular habitat one year, and 
the next year there may be none at all there. If the mycelium of this 
mushroom happens to be colored, so that it can be easily found in the 
soil, it can usually be demonstrated that when the fruiting bodies are 
abundant the mycelium is also abundant, while if no fruiting bodies 
are found the mycelium will also be lacking, or will be present only in 
small amounts. Now, when the mycelium of a fungus, of the proper 
kind, is in the soil around a tree in great abundance, the chances of 
the two organisms coming together and forming mycorhizas are very 
great. On the other hand, if the mycelium is present only in small 
