93 
the Mycorrhiza of the Marattiaceae . 
p. 36), can be seen in these cells. But this circumstance, whilst indicating 
that the presence of the endophyte is in some way responsible for the 
absence of starch from the host-cell, does not necessarily prove that the 
carbohydrate in question is utilized by the fungus as a food-material. 
Since no visible injury is done to the cells of infected roots, it would 
appear that this fungus shares with the great majority of mycorrhiza- 
forming fungi investigated by Gallaud ( 1 . c.) the capacity for obtaining all 
the food it requires from the non-living cell- contents (e. g. starch, sugars, 
elaborated sap, &c.) of its host. 
And the resting spores, with their oily contents, are formed at the 
expense of the higher plant ; hence the advantage of the association is 
almost entirely on the side of the fungus, the host-plant thriving in spite of 
the presence of the endophyte. 
4. Systematic Position of the Fungus. 
The systematic position of the fungus is of special interest, inasmuch as 
our knowledge of the phylogeny and relationships of the various mycorrhiza- 
forming fungi has hitherto been exceedingly meagre. 
From the mycorrhiza of the Orchidaceae Noel Bernard (8) obtained 
cultures of a number of fungal endophytes which he placed provisionally in 
the genus Rhizoctonia. In the same year (1909), Burgeff ( 11 ), working on 
the mycorrhizal fungi of Orchid roots, came to the conclusion that they 
formed a distinct morphological group, for which he proposed the descriptive 
name Orcheomyces. 
Gallaud ( 1 . c., p. 433) observes that, except in the Orchids, the repro- 
ductive bodies of the fungal partners of all mycorrhizas remain to be 
discovered. Dangeard (18 A, 18 B), however, described and figured (l.c., 
PI. XV, Fig. 1 5) not only the mycelium, but also the spores, of a fungus 
which occurs in the rhizome of Tmesipteris , and provisionally referred the 
endophyte to the genus Cladochytrium . Peklo ( 1 . c.), however, claims to 
have isolated a species of Penicillium from the mycorrhiza of Carpinus and 
of Betula , whilst Rayner ( 1 . c.) has recently isolated from Calluna vulgaris 
an endophyte for which the new sub-genus Phytlophoma has been established. 
Judging from its mycelial characters alone, our fungus undoubtedly 
belongs to the Oomycetes, and exhibits many of the essential features of 
the sub-family Peronosporae of the Peronosporaceae, where for the present 
it may conveniently be placed. 
According to Clements ( 16 , p. 17), six genera are included in this sub- 
family ; of these, the genus Phytophthora most closely resembles the fungus 
under consideration, sharing with it several important characters, the most 
noticeable of which are mentioned below. 1 
1 Treub (43) considered the fungal endophyte of Lycopodium Phlegmaria prothallia to belong 
to the Peronosporaceae, while Noel Bernard (10, p. 252 ) suggests that the fungal endophyte which 
occurs in the roots of Solatium Dulcamara has affinities with this family; Nobbe and Hiltner ( 34 ) 
claim a similar position for the endotrophic fungus of Podocarpus nodules. 
