88 Transactions British Mycological Society. 



these had lost their contents and had germinated, sending out 

 hyphal threads that penetrated the rhizoids and were growing 

 in them (Fig. 6). This would seem to afford proof of the sug- 

 gestion that the vescicles form a means of propagating the 

 fungus and infecting the plants. Very few hyphae remain in 

 the thalli of dead plants, and those that are present are for the 

 most part devoid of contents and in various stages of dis- 

 integration. 



The ventral tissue of uninfected plants of Lunularia was found 

 to contain considerable quantities of starch. When infected 

 thalli were examined, it was found that the cells immediately 

 below, and those immediately above the infected cells contained 

 starch, the infected cells themselves however contained none; 

 it would appear therefore that the presence of the fungus is 

 responsible for this disappearance of the starch. Stahl (1900) 

 in his "Sinn der Mycorhizenbildung " connected the formation 

 of starch in the thalli of Marchantiaceae with a highly developed 

 transpiratory organisation, and the complete absence or at any 

 rate meagre development of a mycorrhiza; while in the case 

 of the Jungermanniaceae he connected the formation of sugar 

 in the leaves with low transpiratory activity, and the extensive 

 occurrence of mycorrhiza. It is now certain, however, that the 

 occurrence of mycorrhiza in the Marchantiaceae is as frequent 

 as in the Jungermanniaceae. Moreover, the large quantities 

 of starch present in the former group, and also in Pellia, 

 Fossombronia and other members of the anacrogynous Junger- 

 manniales and which indicates, according to Stahl, vigorous 

 transpiratory activity, does not seem to bear out his hypothesis 

 that the amount of transpiratory activity is connected with the 

 presence or absence of a mycorrhiza in the Hepaticae. 



The presence of Sphagnol in the thallus was tested for, but 

 it was found only in the walls of the upper epidermal cells in 

 both infected and uninfected material; it is unlikely therefore 

 that it has anything to do in this case with the regulation of 

 the growth of the endophytic fungus, but more probably serves 

 to prevent bacteria and moulds from penetrating the upper 

 surface and causing disease. 



Isolation of the Fungus. 



Pieces of the thallus were teased out with a sterilised platinum 

 needle, immersed in i per cent, mercuric chloride solution, 

 washed in sterilised distilled water and placed on plates con- 

 taining agar and some nutrient substance. Thick sections of 

 the liverwort were also placed on nutrient media without 

 previous sterilisation. All cultures were kept in the dark at 



