280 Ward . — On Relations between Host and Parasite 
But the most irritating error of all, and one which caused 
much trouble in the immediate future, was the discovery- 
only made after the plants had come into flower- — that the 
seedlings labelled B. asper were not B. asper at all, but 
B. arvensis — that is to say an error had been made by the 
seedsman who supplied the seed. In consequence of this, 
I had to overhaul the whole stock of seeds and seedlings 
myself, and in the experiments which follow believe there 
need be no doubt as to the authenticity of the species, except 
in one or two special cases to which attention will be drawn 
where necessary, and in regard to which further investigations 
are proceeding. 
It becomes clear, in fact, that we have to keep our eyes 
open to the danger of impure cultures of the hosts, as well as 
of the parasite . As will be seen in the sequel, we have also 
to take account of variations and race-peculiarities in both. 
Considering these and other possibilities of error, I was 
led to devise some better methods of preparation and infec- 
tion of the culture-plants and spores. 
In the following experiment (see Table XII) sowings in 
pots were made of eleven carefully selected species or varieties 
of Bromus , viz. : B. mollis , L., B. secalinus , L., B. velutinus , 
Schrad., B. maximus , Desf., B. arvensis , L., B. racemosus , L., 
B. comnmtatMS , Schrad., B. interruptus , Hack., B. sterilis , L., 
B. madritensis, L., B. erectus , Huds. Good garden soil was 
used, and germination, under a hand-glass in the garden, 
resulted normally during the warm weather, so that in a week 
it was possible to start the infections. It may be noted, 
however, that whereas B. mollis showed above ground on the 
fourth day, and B. arvensis and B. maximus were up by the 
fifth day after sowing, it required six days for B. sterilis , 
B. madritensis , and B. secalinus , and seven days for B. race- 
mosus and B. interruptus to be sufficiently high, while B. 
commutatus and B. erectus were not well up 1 until the 
eighth day. 
The infections were made by placing Uredospores from 
1 I regarded the seedlings as i well up * when 1-3 mm. showed above ground. 
