BY R. GREIG SMITH. 457 



were made through a black area. The examination of these 

 sections showed that the perithecia contained spores lying free 

 in the cavity and surrounded by a single layer of black-coloured 

 goblet cells, the apertures of which were directed into the cavity 

 (PI. XV., tig. 14). No other structures such as paraphyses or coni- 

 diophores were present. The section also shoM^ed smaller peri- 

 thecia in the act of fusing, the dividing cellular walls being 

 absorbed and the contents of the smaller perithecia joining to 

 form a larger pycnidium (PL xv., fig. 15). The ready absorption 

 of the cell walls indicates that the goblet cells were spherical or 

 palisade cells, tlie inner ends of which had been dissolved by a 

 cytase. 



By this time it had become evident that the presence of sugar 

 in the medium caused a proliferation of hyph^e, while the absence 

 of soluble carbohydrates induced the formation of the spherical, 

 dark-coloured cells. It also occurred to me that if small quanti- 

 ties of an almost sugarless but slightly nitrogenous medium were 

 inserted into parts of a stiff starch-paste, we should have a con- 

 dition of affairs more suited to the growth of the organism than 

 could be obtained in hanging drops or on gelatine media. A 

 trial showed that the method was valuable, inasmuch as it 

 enabled phases of the development to be observed which were 

 not visible by the usual methods. A 20 % starch-paste was pre- 

 pared and poured into small Esmarch-dishes to a height of 1 cm. 

 A drop of yeast-water, infected with spores from a perithecium, 

 was introduced beneath the tough surface skin of the starch by 

 means of a sterile capillary glass tube. In two days at 22° the 

 surface of the starch at the point of infection was scantily covered 

 with long, white, pinkish or greenish, branching septate aerial 

 hyphse, below which were many black points radiating in lines to 

 the margin of the medium. These black points were the peri- 

 thecia that were found in the potato and sugar-cane cultures. 

 Imbedded in the starch-paste were numerous hyphse which showed 

 all conditions between the simple cell and the perithecium, thus 

 enabling the various stages in the development of the perithecia 

 to be observed (Pl.'xv., fig. 13). When the cylindrical cell becomes 

 30 



