79 
the Mycorrhiza of the Marattiaceae. 
A few plants were examined in the roots of which no trace of the fungus 
could be detected ; this condition also obtains in a small proportion of the 
roots of all plants which have passed their juvenile stage. This may explain 
Kuhn’s ( 28 , p. 494 ; 29 , p. 149) failure to find infected roots in his specimens 
of Marattia fraxinea , Sm., and of Danaea alata r Sm ., and the statements of 
Campbell ( 14 , p. 215) and of Charles ( 15 , p. 84) concerning the distribution 
of the endophyte in older roots. 
The presence of the mycorrhizal fungus, which can be detected only 
by an examination of sections of the host-root, does not influence in the 
slightest degree the external appearance of the host, infected and uninfected 
plants appearing to be equally healthy and vigorous. Neither does the 
nature of the soil bear any relation to the presence or absence of the endo- 
phyte, which is found in the roots of plants growing in soils ranging from 
light rich humus to red clay from disintegrated gneissic rocks. 1 
Thus it would appear that in the above-named genera of Marattiaceae 
symbiosis is not obligate, but facultative. 
1. Distribution , Organization , and probable Life -history of the Fungus 
in the Tissues of the Host. 
i. Mycelium. 
Microscopical examination of transverse or longitudinal sections of 
infected roots shows that fungal hyphae from the surrounding soil may enter 
the root through any of the epidermal cells in spite of the cuticularization of 
the outer wall of these cells (Text-fig. 1 ; cf. Noel Bernard, 10, p. 243). 
The mycelium is both inter- and intracellular within the tissues of the host, 
exhibits very few septa, and is copiously branched (Text-fig. 2, a and b). 
1 The author is indebted to Professor J. B. Farmer, F.R.S., for information regarding the habitat 
of a number of plants of Angiopteris evecta, Hoffm., gathered by him in Ceylon. 
