486 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY. 



Inoculation. — It was found that a drop of water could not remain on 

 the leaf because of the waxy epidermis. The wax was therefore removed 

 by means of a dry cloth. The leaves so treated were marked in such 

 a manner that the inoculated area could be easily recognized. The 

 soil was removed at a little distance around the plant, so as to leave 

 a depression into which water could be poured without removing 

 the cloches, and by this means the atmosphere around the plants was 

 kept constantly in a state of humidity. The conditions were precisely 

 the same for the plants acting as controls. The plants were under 

 daily observation, and twenty-nine days after inoculation small 

 discoloured spots were observed on the leaves in the inoculated areas. 

 Some of the spots were cut out, and hand sections made which showed 

 an abundance of mycehum ; other pieces were imbedded in wax in the 

 usual way. No spores of the fungus were seen. The spots increased 

 in size, and a period of six weeks elapsed between inoculation and 

 production of spores from the infecting mycelium. The spores proved 

 to be those of Heterosporium gracile. The control plants remained 

 perfectly healthy, and showed no signs of disease or discolouration. 

 The experiment was repeated, and different clumps of plants were 

 used. The weather at this time was warm and sunny, and in this case 

 the conidia were produced after a period of four weeks, the shorter 

 period being due to the high temperature and dampness of the atmo- 

 sphere within the cloches— conditions favourable to the growth of 

 the fungus. 



Examination of Diseased Tissue. — The material for examination was 

 taken from diseased Iris leaves growing in the gardens, and also from 

 the artificially infected plants. Killing and fixing reagents used 

 were Flemming's medium solution, chromacetic acid, and acetic 

 alcohol. The material was cleared in xylol and imbedded in wax 

 with a melting point of 54° C. Microtome sections were cut 5-15 /x 

 thick. Material killed with Flemming's solution and stained with 

 safranin — gentian violet — orange G, gave the best results, although 

 iron haematoxyhn with a counter stain of eosin was found excellent. 

 Observations were also carried out with hand sections and macerations 

 of fresh material. It was a rather difficult matter to arrive at the 

 true nature of the mycelium within the tissues of the leaves, and 

 old diseased areas were found most unsatisfactory for cutting. The 

 mycehum was multiseptate and was present in large quantities in 

 the intercellular spaces (fig. 107). No haustoria were found, and in 

 material which had been cut through the centre of a diseased spot 

 hyphse were observed penetrating the cells of the tissue, in which 

 cases the cell walls and protoplasm became brown and partially 

 consumed (fig. 107). In the air space adjacent to the stomal opening, 

 the hyphae formed clumps of a pseudoparenchymatous nature, and 

 it was from these that the conidiophores were given off through the 

 open stomata to the exterior (fig. 107, c). 



In a few instances thick-walled, closely septate hyphae were seen 

 in the macerated material which had been taken from naturally 



