67 
individual parts had reached full size. The fungus germs can follow- 
this stretching of the tissue of the host plant only at their extremities^ 
not in the remoter, older parts, which are incapable of intercalary 
growth, and which, consequently, being subject to strain, must be ob¬ 
literated by being drawn out into threads. 
Even in the next series of experiments in which older plants were 
infected ; that is, somewhat older seedlings in which the sheath leaf was 
over a half inch long, but not yet broken through, the places of pene¬ 
tration occurred more rarely, and where they were to be seen many of 
the penetrated germ tubes had ceased to grow in the outer cell layers. 
They then exhibited an entirely different appearance, viz, a strong 
swelling of the membranes, which was often associated with a yellowish 
color. These objects had an unmistakable likeness to Wolff’s drawings 
of the penetration spots, which the author has described as forming a 
cellulose sheath around the penetrating germ tubes. I have never seen 
such a sheath in normal cases of penetration and I consider it quite 
probable that Wolff only saw imperfect spots of penetration, with swollen 
germ tubes which he mistook for cellulose sheaths, because he confined 
his inflections solely to the sheath leaf in which, in somewhat older 
stages, the penetrated germ tube can not push in any further. (Wolff', 
Brand des Getreidvs , Halle, 1874.) 
In order to follow up thes.‘, observations I made repeated infection 
experiments with seedlings in which the sheath leaf was nearly full 
grown and was already broken through for half an inch by the fol¬ 
lowing leaves. Here from the root node to the uppermost point I found 
no longer any normal spots of penetration. Very rarely a thread was 
found which had pushed through the two outer cell layers, then ceased 
its penetration and slowly perished with swelling of its membrane. At 
the same time there lay upon the surface hundreds of germinated 
conidia which could no longer penetrate, because the epidermis, fully 
formed in the meantime, was no longer permeable. The seedliugs, there¬ 
fore, in this stage of development already behaved toward the fungous 
germs exactly as do all parts of fully developed plants, into which, as is 
well known, the threads can not penetrate and in which they can not 
grow further. 
Up to this point of the investigation, therefore, my observations 
confirmed, with some additions and amplifications, the earlier results 
of Kiihn and some statements of Wolff. Nevertheless, it would ap¬ 
pear to me that they only partially exhaust the question as to the place 
of penetration of the smut germ into the host plant, and that even the 
new proof material which supported the old idea hitherto generally 
accepted, that the smut germs must penetrate into the young seedlings 
in order later to produce smut in the full grown plants, is still insuffi¬ 
cient and can not well be regarded as definitely concluding the investi¬ 
gation. For why should the penetration occur only in the young seed¬ 
ling which possesses no other disposition for it except the immaturity 
