I.—PHYSIOLOGY. 201 
relation existing between orchids and Mycorhiza, based as it is upon 
physiological studies. All who had to do with orchids in the last 
century found the greatest difficulty in raising these plants from their 
seed ; a successful result appeared to depend largely on chance. Culti- 
vators of orchids found that success was obtained more frequently by 
placing seed in soil upon which orchids had previously lived, and much 
secrecy was observed as to the methods employed by the more successful 
cultivators. 
The seeds of orchids are exceedingly small—a million may be found 
in a single capsule of an exotic species; they possess no albumen and 
contain an embryo consisting merely of a mass of undifferentiated cells 
provided with a suspensor. ‘The essential discovery of Bernard was that 
orchid seeds do not germinate in the absence of fungi belonging to the 
genus Rhizoctonia. The fungus enters the seed through its least 
resistant and highly permeable cells, which apparently emit a secretion 
that attracts the fungus. Each species of orchid, according to the subse- 
quent researches of Burgeff (1909), possesses a special species, variety, 
or race of fungus that is particularly adapted to it—he distinguishes 
fifteen species of fungus. When mutually adapted orchid seed and 
fungus are brought together, the mycelium of the latter penetrates the 
suspensor cells by digesting their cellulose wall. The mycelium 
traverses the epidermal cells of the seed without undergoing development 
within them. As soon as the primary infestation has occurred, even 
where the mycelium has penetrated but slightly, the cells of the seed, 
situated at the posterior pole of each embryo, cease to be vulnerable. 
In other words, a local immunity appears to be established, this 
immunity lasting at any rate until new regions are attacked by the 
fungus. This, in Bernard’s experience, is the general rule. The 
mycelium, having attained the parenchyma cells, develops into charac- 
teristic filamentous masses recalling the appearance seen in bacterial 
agglutination. Nevertheless, there comes a time, this varying according 
to the associated species involved, when the development of the fungus 
is arrested by the deeper parenchyma cells of the seeds. These cells 
are altered before they are penetrated by the fungus ; they become hyper- 
trophied and acquire large lobose nuclei. They digest the mycelium 
which enters their protoplasm, but the cell continues to harbour remains 
of the fungus (‘ corps de dégénérescence ') which occur abundantly in 
the tissues of orchids. The seed now proceeds to sprout, giving rise to 
a small tubercle (‘ protocorm ’), which only at a later period produces 
leaves and roots. 
The cuitivation of Rhizoctonia of various species was carried out 
successfully by Bernard, the cultures being used to reproduce germina- 
tion in orchids. Orchid seeds alone remained unchanged for months in 
cultures on agar with salop-decoction added, but when pure cultures of 
Rhizoctonia mycelium were added to such orchid seeds, the latter were 
invaded by the fungus, germinated, and gave rise to a ‘ protocorm.’ 
Bernard gives excellent figures illustrative of the development described. 
The relation between the fungi and orchids varies in different groups 
and plants. In primitive forms like Bletilla germination occurs in the 
absence of the fungus, but the ‘ protocorm’ does not develop; the 
