341 
son. Thus 102 rows were planted, 51 standing south of a 5-foot alley, 
and 51 north. Every other row in the south block was treated, begin¬ 
ning with the second row. Each treatment made on the south block 
was duplicated upon the north block, rows treated with the same sub¬ 
stances being placed as far apart as possible, in no case nearer than 
00 feet. All rows were planted with seed from the same lot and as 
nearly as possible in the same manner. They were given precisely the 
same normal nursery treatment, being thinned out, hoed, and cultivated 
as nearly as possible on the same days. The seMs germinated normally 
and produced u stands” of uniform vigor, and not until the influence 
of the soil began to make itself felt was there the slightest difference 
noticeable between any of the rows. The unevenness of the soil, 
however, soon disturbed this uniformity and proved a more potent fac¬ 
tor than the disease, but owing to the arrangement of the rows in 
duplicate it in no way disturbed the experiment. 
The twenty-five chemicals which it was designed to test were all care¬ 
fully weighed out, and concentrated solutions were prepared during the 
winter of 1891-92 in the Department and shipped to Geneva ready to 
be diluted and applied to the seedlings with a sprayer. 
The test of the preparations must be considered as wholly prelimi¬ 
nary and designed to form a basis for further investigations. Hence 
the fact that a large number of the substances failed to prevent the 
disease by no means signifies that they may not yet prove to be true 
fungicides when of a suitable strength. According to the writer’s 
notion, there is one requisite for, and two main limitations to, the prep¬ 
aration of a valuable fungicide. The requisite is that the prepara¬ 
tion be a true fungicide and prevent infection from the disease. The 
limitations are, (1) that the expense of the material and its application, 
including the element of danger, shall not be greater than the benefit 
will warrant; and (2) that the effect upon the plant to be protected 
shall not be injurious. Any substance which fulfils the above require¬ 
ments and does not overstep the limitations will prove a valuable fungi¬ 
cide. 
Before it is possible to thoroughly test preparations with regard to 
the limitations above mentioned, it is necessary to gain some idea as 
to what mixtures are likely to be available for use and to eliminate 
those which from a combination of injurious and nonfungicidal prop¬ 
erties are manifestly unworthy of further trials. In the summary the 
various mixtures have been grouped into three classes, clearly showing 
which are likely to be valuable. 
In designating the different mixtures the exact chemical name of 
the supposed active salt has been used wherever the composition of 
such is known, otherwise the less specific title has been given. 
The following is a list of the substances used, with the formulae for 
their preparation and a statement of their effect upon the seedlings. 
The term u mixture” is here used in its broadest sense, to include the 
