474 POPULAR SCIENCE REVIEW. 
stems nor leaves should come in contact with water containing 
zoospores. 
If a drop of water thus charged is placed on the surface of a 
living leaf of the shepherd's-purse, for instance, and left at 
rest for a few hours and examined minutely at the end of that 
period they will he found to have germinated. Let the 
epidermis he removed carefully and placed on a glass slide and 
submitted to the microscope. Many zoospores will he found 
to have produced from that point of their surface which is 
nearest to one of the stomata, or pores of the leaf, its slender 
tube, and to have thrust it through those openings, with the 
swollen extremity resting in the air-cavity situated beneath 
the pore. If many days, or even weeks, are allowed to pass, 
and the leaf is examined again, or another leaf similarly 
treated, and kept in a living and vigorous condition by 
remaining attached to the parent plant, still no further change 
or advance will be observed, the germs will appear fresh, and 
still in the same condition. Hence it is concluded that plants 
are not infected through the medium of their leaves. 
If the cotyledons, or seed-leaves, are watered with similarly 
impregnated water, a different result has been observed to 
take place. The germination of the tubes till their entrance 
at the stomata is the same, but, having entered, the swollen 
extremity elongates, becomes branched, and takes all the 
appearance of mycelium such as we at first described. If 
the infected plant endures through the winter the mycelium 
endures with it, to recommence vegetating in the spring. 
The experiments which Dr. de Bary performed were all 
upon plants of the common garden-cress. If will be unneces- 
sary to repeat the details of these, as given in the memoir 
recently published on the subject, but it will suffice to give a 
summary of results. In two series of plants cultivated at dif- 
ferent periods from good seed, one hundred and five plants 
which had not received the water impregnated with zoospores 
upon their cotyledons vegetated without any indications of 
the parasite. Amongst the eighteen plants which were 
inoculated by watering the cotyledons, four only were not 
attacked by the parasite, fourteen bore the e< white rust,” In 
six of these it did not extend beyond the cotyledons, in the 
others it also appeared on the stems and leaves. 
From these experiments it may be deduced that plants are 
not infected by spores of the parasite entering at the roots, or 
by their leaves, but that inoculation takes place through the 
medium of the cotyledons or seed-leaves ; that the agents in 
this inoculation are the zoospores produced either from 
the conidia or the oospores ; that they do not enter the stomata 
or pores themselves, but thrust out a germinating tube into 
the extremity of which the contents of the zoospores pass; 
