180 BOTANICAL GAZETTE [MARCH 
he was able to establish. He did not succeed in infecting pieces 
of leaves arranged for the purpose. If, under natural conditions 
of infection, the germinating tubes have a limited growth unless 
supplied with nutrient material from the host in some way, it is 
not clear how they can obtain it so as to grow down into the 
leaves for comparatively long distances, as Phyllactinia does, 
before producing haustoria, unless the fungus appropriates inter- 
cellular material. The subject is still open for investigation. It 
is possible, of course, that the first side branch, passing into the 
interior of the leaf, places an absorbing organ near the stoma 
and thus supports the whole branching system until other side 
branches in other stomata are able to assist. 
There is some evidence to support the possibility of intercel- 
lular nutrition in this fungus, as has been established for others 
by Nordhausen, Ward, Biisgen, and Frank. This investigation, 
however, has not embraced the question of how far, if at all, the 
abundant intercellular material in Xanthoxylum leaves is available, 
though some facts were observed which suggest the possibility 
of intercellular nutrition of the parasite which have not been 
observed in other hosts infested by Phyllactinia. 
There are two points connected with my observation of 
Erysiphe on the geranium leaf which may be mentioned briefly. 
At the end of a section where the scissors with which the pieces 
of leaves were cut had destroyed the section in part, a hausto- 
rium was found in what was clearly a subepidermal cell. Its 
connection with the surface was destroyed, but there is no doubt 
that it was a normal haustorium in one of these cells. Whether 
it was due to the chance presence of Phyllactinia on this hats 
whether Erysiphe occasionally adopts the practice of Uncinula 
Salicis, or whether it was Uncinula itself on this host, could not 
be determined. With this circumstance is related a note of 
page 9 of Ellis and Everhart’s Pyrenomycetes. This note cor- 
rected the mistake of the artist (F. W. Anderson) who, it wae 
Supposed, wrongly represented (fig. 3, pl. r) a germinating tube 
from a conidium of Sphaerotheca Castagnei as entering 4 stoma. 
My observations render it probable that the artist saw what he 
MAnEeEeamN 
SIR aN ICA el OR ASS AO MNT IE IAC ct 
