794 ECOLOGY 



berization of orchid roots are due to some stimulus instituted by the fungal 

 symbiont, so that the resulting structures are to be regarded as galls. As in other 

 galls, the cells are much hypertrophied and the nuclei often assume enlarged or 

 otherwise unusual forms. The identity of the mycorhiza fungi is in considerable 

 doubt, owing to the usual absence of reproductive organs and the difficulty in mak- 

 ing artificial cultures. Originally they were thought to be truffle mycelia, and it is 

 not unlikely that this is a correct identification for some forms, while others may 

 belong to the agarics, the molds, or to various other fungal groups. In Fagus and 

 Carplnus there is evidence that the mycorhiza fungi are molds, such as Penicillium 

 and Citromyces. In the orchids it is probable that there are specific forms com- 

 parable to the legume bacillus. Recent studies seem to indicate that the orchid 

 fungi include at least three species which appear to belong to the genus Rhizoctonia; 

 in Phalaenopsis there is a peculiar fungus (perhaps a Rhizoctonia), whose hyphae 

 anastomose and form sclerotia. 



The prevalence of mycosymbiosis . Until a few years ago mycosym- 

 biosis was believed to be a somewhat rare phenomenon, characterizing 

 only a few families, such as the conifers, ericads, orchids, and oaks ; 

 but now it is believed that a majority of ordinary green plants are myco- 

 phytes (i.e. fungus plants), as green plants with root fungi may be called. 

 In Germany seventy out of a hundred and five species taken at random 

 had root fungi, and in Java sixty-nine out of seventy-five. Doubtless 

 the ectotrophic fungi are more abundant than the endotrophic, though 

 the latter occur in numerous forms, as in aroids, lilies, and many trees, 

 as well as in orchids and ericads; it is probable also that many of 

 the former are mere contact forms without particular significance. 

 Endotrophic fungi are now well known in some mosses (as Buxbaumia 

 and Tetraplodori) and in many liverworts (notably in the Jungerman- 

 niales, but also in Fegatella and Marchantia). Mycorhizas with char- 

 acteristic coralloid rootlets and hyphal clumps have been detected in 

 certain Carboniferous plants, as in Cordaites, thus attesting to the an- 

 tiquity of mycosymbiosis.' For the. most part root fungi are absent in 

 sedges, pinks, crucifers, most ferns (i t e. Polypodiaceae) and legumes, 

 though the last are characterized by bacterial symbiosis. As might be 

 expected, mycorhizas are associated abundantly with plants rooted in 

 forest mold. They are rare in water and in wet soils (except in bogs) 

 and are almost universal in bulbous and tuberous plants. 



The role of root fungi. While root fungi have long been known, 

 it originally was supposed that their contact with roots is merely casual, 

 or that they represent ordinary parasites. Some years since it was 

 suggested that Monotropa is likely to have a nutritive relation with its 

 fungus, since the latter completely invests the root system. Careful 



