i 7 ° M. Chevely Rayner. 
development is strictly localized and controlled by the digestive 
power of the root-cells. It may surely be said that in these plants, 
as in the animal body, immunity —in the sense of an absence of 
pathogenic symptoms following upon the presence of a foreign 
organism in the tissues—is determined by the activities of certain 
specialized cells or phagocytes. 
The details of nutrition in the non-chlorophyllous orchids are 
still so uncertainly known, that the case of Gastrodia, recently 
described by Kusano (16), is of special interest. 
The curious saprophytic orchid, Gcistrodia, data , is a native of 
Japan, where it is frequently found growing in woods, below 
Quercus serrata and Quercus glandulifera. The whole vegetative 
body consists of a colourless, rootless tuber, 10-17 centimetres long 
at maturity, associated with which are invariably a number of small 
daughter tubers, arising as offsets from the parent. It is noteworthy 
that the latter are relatively very abundant as compared with tubers 
of flowering size. The rhizomatous portion of the tuber bears scale 
leaves and the whole structure is invested with a corky covering 
like a potato. The immense inflorescence appears at the end of 
May and is produced only by tubers of full size. The plant is 
destitute of chlorophyll, has no root-system, and the corky covering 
of the tuber must render it a singularly inefficient absorbing organ. 
Before the work of Kusano, Gastrodia was assumed to be a 
humus-saprophyte with mycorhiza,—in spite of the absence of roots! 
Nothing was known definitely of the mode of nutrition. This 
author has now demonstrated a very remarkable case of symbiosis 
between the Gastrodia tuber and the rhizomorphs of Armilleria 
mellea (Rhizomorpha subterranea ), a common fungus on dead and 
living roots of Quercus in the neighbourhood. Kusano regards the 
rhizomorphs as usually saprophytic on Quercus , but they are recorded 
as troublesome parasites of the potato-tuber on farms near the woods. 
The history of the association described by Kusano is briefly 
as follows. The great majority of the small tubers are entirely free 
from fungal infection ; investigation at various stages of development 
yielded no evidence of an association with any fungus-mycelium, 
and gave no clue as to the mode of nutrition of the plant. Such 
tubers never flower. Flowering tubers, on the other hand, are 
invariably infected. The rhizomorphs of the fungus ramify on the 
surface and at certain points penetrate the flesh of the tuber by 
special haustorial branches. Below the surface, these haustoria 
give off mycelial branches which are responsible for an elaborate 
and highly-differentiated infection of the outer tissues of the tuber. 
