I IO 
Ernest S. Salmon. 
the infection-powers of ascosporcs are strictly similar to those of 
the conidia in the case of “ biologic forms.” 1 
The method of inoculation, &c. adopted in the present experi¬ 
ments was that used in previous work, and has already been fully 
described'-. My sincere thanks are due to Professor H. Marshall 
Ward, F.R.S. for again kindly allowing me to carry out the work in 
the Cambridge University Botanical Laboratory, where special 
facilities exist for such work. My thanks arc also due to Mr. R. 1. 
Lynch, A.L.S., Curator of the Cambridge Botanic Garden, and to 
Mr. Hosking, the Foreman, for the trouble and care taken in raising 
the plants used in the experiments. 
I.—Inoculation-Experiments with the Conidia of E. Graminis 
on Avena sterilis, Agropyron repens, Poa pratensis, and 
Dactylis glomerata. 
The Oidium on Avena sterilis (see Table 1) proved able to cause 
full infection when sown on A. sativa, the six inoculated leaves all 
becoming infected. In three experiments nine leaves of A. pratensis 
were inoculated ; eight of these were completely passed over, but on 
the ninth leaf “subinfection” 3 occurred. Inoculation of ten leaves 
of Arrhenatheruiu avcnaceum resulted only in the weak “subinfection” 
of one leaf. It may be noted that Marchal has recorded 4 that the 
Oidium on Avena is able to infect the present species. No details 
of the inoculation-experiments in which this result was obtained 
have been published, but the author has kindly informed me that 
only “subinfection’’occurred on the Arrhenatheruni. In my previous 
experiments 5 in which leaves of A. avenaceum were inoculated with 
conidia from Avena sativa, no infection resulted. No instance is 
at present known of an Oidium possessing the power of causing 
full infection of species of another genus than the one to which its 
host-plant belongs. In the present experiments inoculations of 
1 See Salmon, K. S.; in Journ. of Hot., XI,I., 159-165. 204-212 
(1903); Marchal, E.; in Comptes Rendus, CXXXVI., 
1280-1281 (1903). 
2 Beiliefte z. botau. Centralbl., XIV., p. 266 (1903). 
3 The term “ subinfection ” is used to denote cases in which 
inoculation is followed by the production merely of a few 
scattered conidiophores, which never form powdery Oidium- 
patches, and which usually disappear after a few days. 
I have already treated of the possible significance of these 
cases of “ subiufection." (See Beiliefte z. botan. Centralbl. 
XIV., p. 270 (1903) and “ A1111 ales Mycologici” II , 75 (1904) 
Comptes Rendus, C'XXXV., 211 (1902). 
1 See Beiliefte z. botan. Centralbl.. XIV., p. 289. Table XI 
