166 M. Chevely Rayner. 
inducing germination of seeds of Cypripedium and vice versa. With 
species of Phalcenopsis, Vanda and Odontoglossum, in which the 
seeds were notoriously difficult to germinate, the relation is more 
specialized, and each plant species has its specific endophyte which 
alone is effective in causing germination. This specificity was 
shown experimentally by sowing seeds of Phalcenopsis and infecting 
some cultures with their own endophyte and others with fungi 
obtained from the roots of species of Cattleya and Odontoglossum. 
Thus, using the endophyte from Cattleya , the seed was killed, no 
germination took place and there was no digestion of the fungus by 
the cells of the embryo. With that from Odontoglossum, infection of 
the seed occurred, germination was at first normal but was followed 
by digestion of the fungus in the cells of the embryo and subsequent 
arrest of growth. Infected by the fungus found in the parent plant, 
germination proceeded in the usual way and gave a well-developed 
plant. Again, a high degree of physiological adaptation was shown 
by experimental cultures which indicated a critical stage in the 
development of the seedling—earlier or later according to the species. 
In some orchids practically no germination takes place without 
infection ; in others, presumably relatively primitive, e.g., Bletilla 
hyacintliina , development goes so far as the production of a tubercle 
with several leaves: it never reached the stage of root-formation 
in any species investigated by Bernard. Of great biological interest 
also is the experimental demonstration that the orchid plant sets a 
limit to the spread of its fungus partner, shown not only by digestion 
of the mycelium in certain cells of the root, but also by complete 
exclusion of hyphae from the chlorophyllous tissues of the shoot. 
Bernard was struck, as all biologists must be, by the parallelism 
between certain aspects of ‘ symbiosis ’ in the orchids and the 
resistance to disease described as ‘ immunity ’ in animal pathology, 
and the machinery of phagocytosis which is one of the means by 
which such immunity is secured. His subsequent researches were 
based upon this point of view. It is much to be regretted that 
they have been brought to an end by the untimely death of the 
author. 
In 1909, Burgeff (13) published a monograph on the root-fungi 
of the orchids in which he reviewed the work of Bernard, confirmed 
his experimental results and gave an account of his own original 
researches on the mycorhizal fungi of many orchid species. 
Bernard had referred these fungi to the genus Rhizoctonia, after 
provisionally placing them in the genus Oospora when first isolated. 
Burgeff proposes the new generic name Orcheomyces for the group 
