220 Transactions British Mycological Society. 



arise (fig. 3). When infection is due to a spore the stromata are 

 at first isolated ; otherwise they are connected to other stromata 

 by strands of hyphae, which have made their way underneath 

 the epidermis of the leaf, usually as far as a stoma (fig. 2) . The 

 hyphal strands are of different lengths, sometimes very short 

 indeed, so that the new stromata coalesce with the parent. From 

 each cell of the young, plate-hke stroma an erect hypha is 

 produced, wavy in outline, which ultimately becomes a conidio- 

 phore (fig. i). 



Groups of characteristically waved conidiophores (fig. i) 

 burst through the cuticle, having destroyed and replaced the 

 epidermal cells. Each conidiophore produces a large obovate 

 uni-septate spore, which is so large (20-22 x ii-15/x) that it 

 appears to be slightly top-heavy. Many spores drop off and lie 

 around the bases of the conidiophores. The conidiophores show 

 two parts, a waved upper portion and a straight basal portion, 

 and generally a septum separates the two parts. The upper 

 waved portion is always a darker colour than the basal part. 

 The stroma is not buried deeply in the mesophyll of the leaf, 

 but only extends into the first two or three layers of cells. The 

 thickness of the stromata varies, some stromata being several 

 cells deep, others (usually young) forming a plate one cell deep, 

 but even this may bear conidiophores (fig. i). The hyphae 

 composing the stromata are olive-brown, broad, and many- 

 septate, giving it a pseudoparenchymatous appearance. Hyphae 

 do not ramify extensively through the leaf, but merely extend 

 peripherally into the neighbouring cells and produce fresh 

 stromata. At first the hyphae progress intercellularly, the 

 middle lamella being dissolved; afterwards the mesophyll and 

 epidermal cells with which they are in contact are consumed and 

 the hyphae occupy the space. 



The specimens of Polythrincium Trifolii which have been 

 examined have always differed somewhat from those described 

 by Kunze(i2), who considered the conidiophores to be "septate 

 structures, whose individual cells were four-cornered with 

 rounded ends." He also mentions the occurrence of "double 

 grains" at the base of the stroma — evidently the two-celled 

 conidia which have fallen off the conidiophores and are clinging 

 about their bases. Most authors follow the original description 

 and refer to septate conidiophores ; we do not find septa in the 

 undulating upper portion. It is strange that the error should 

 have been persisted in as Corda pointed it out so long ago as 

 1839, in the work in which the drawing of the fungus occurs 

 which has been copied so widely (8). 



Hanging drop cultures of conidia were made in tap water 

 and weak cell sap extract at ordinary room temperature in 



