20 



JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY. 



operation to that of transplanting or bedding-out cuttings, 

 rhizomes, stools, bulbs, &c, in contrast to that of seed-sowing, 

 and some very valuable results have been obtained by this 

 method of culture. 



The application of De Bary's methods to the study of para- 

 sites of the grasses and cereals has yielded some very remarkable 

 results of late years, especially at the hands of Eriksson and 

 Banning. These experimenters have examined the behaviour of 

 rust on 105 species of grasses, and have found that just as we 

 have varieties and races of the higher plants, so with Puccinia 

 gra minis, the fungus of wheat-rust, there are several distinct 

 varieties which behave very curiously. For instance, the variety 

 which infects the barley will not infect oats or wheat, but will 

 grow on rye and twitch ; that on oats refuses to infect wheat, 

 and conversely ; that on Aira ccespitosa is different again, and 

 so is that on Poa. Nevertheless, all these form-species or 

 varieties will grow on the barberry, and since they are practically 

 indistinguishable under the microscope we are driven to the con- 

 clusion that it is owing to some influence of the host-plant on 

 the physiology of the fungus that it has thus become specialised 

 in its parasitism. 



If these were the only cases where the fungus is found to 

 have become so closely adapted to the conditions of its natural 

 seed-bed, I might be tempted to pass over the matter as too 

 speculative to bring before you ; but it is not so, for, in addition 

 to other species of Puccinia, we know that such specialisation 

 applies to an Accidium on the spruce, as De Bary found in 1879, 

 to a Peridermkm on the pine, as Klebahn showed in 1892, to 

 several forms of Gymnosporangium on junipers, as Mr. Plow- 

 right has shown, as well as other observers ; and to Ustilago, 

 according to recent researches by Swingle and others. That we 

 are here face to face with phenomena of the same order as those 

 where races or varieties of yeasts are formed by cultivation, and 

 of bacteria by variations in the conditions, may safely be asserted ; 

 but I would go further than this, for it seems to me that the 

 phenomena also come into the same category as the variation of 

 plants like the cabbage, Indian corn, wheat, potatoes, &C, on the 

 one hand, and the specialisation of the pollen in heterostyled and 

 other flowers to certain stigmas, on the other. 



Bat there enn be no doubt that among the most important 



