Mar. 3,1933 
Gray Mold of Castor Bean 
709 
present was, of course, promptly dissolved. Test-tube samples of these 
solutions, in duplicate, were then set up in racks and treated with potas¬ 
sium ferrocyanid. The color intensities produced were compared with 
those in a graded series of copper-sulphate solutions of known concen¬ 
tration, and readings were taken accordingly from the nearest in color 
in terms which were readily convertible to milligrams of copper per 100 
gm. of the sample. It is evident that each unit of weight of such mate¬ 
rial as castor-bean inflorescences, containing a large proportion of heavy, 
immature pods, would have much less exposed surface than the same 
weight in leaves, as given in the paper referred to (24 ); consequently the 
results obtained in this case are not comparable with theirs. They are 
comparable, however, within this experiment, one condition with another. 
In order to establish a basis of comparison to determine persistence of 
the different materials on the plants, a special application was made at 
the close of the series. The spray was allowed to become thoroughly 
dry and the dusted spikes to blow in the wind for a couple of hours; 
then samples were collected and tests made for copper, with results as 
follows: Three of the dusted samples gave 41.7, 62.6, and 50 mgm. 
copper per 100 gm. of sample, averaging practically 50 mgm. per 100 gm.; 
two of the sprayed samples gave identical results—25 mgm. copper per 
100 gm. of sample. Consequently these figures are taken as very rough 
bases upon which to calculate percentages of persistence of the copper 
applications. The results of the spraying experiment, week by week, are 
given in Table II in figures representing total number, number of moldy, 
and percentage of moldy spikes. In addition, figures are given showing 
the weight in milligrams of copper, as such, per 100 gm. of plant parts 
sprayed, at the close of each week, together with the relation between 
that and the “perfect’' persistence, expressed in percentages. Some of 
the results are more graphically presented in Figure 4. 
Several interesting facts can be ascertained by study of this table and 
figure 4. In the first place, there is a very definite correlation between 
rainfall and the development of mold in all the plats. During a week 
when less than inch of rain fell, as a rule less than 20 per cent of the 
flower clusters in the untreated plats developed the mold; when over 1 
inch fell, more than 25 per cent; and in one week over 50 per cent became 
moldy. An exception to this was the week ending August 9, when the 
rainfall was bunched and several days were dry. The difference in the 
persistence of the copper applications can also be traced, on the whole, 
to rainfall, though some inconsistencies appear to be present. These 
are due, as a rule, to the time of application of the fungicides in relation 
to some of the heavier downpours of rain. For example, during the week 
ending September 6, the single application of Bordeaux dust almost 
entirely disappeared, whereas a comparatively large proportion of the 
second application, made after most of the rains were over ( 20) remained 
(fig. 5). The copper of the Bordeaux spray was, on the whole, more 
persistent than that of the dust. The anomaly evident for the week 
ending September 6, in which the Bordeaux spray has a sticking quality 
better than 100 per cent, can be explained only by the supposition that 
an unusually heavy application was made in the middle of the week, after 
the rain was over for the week. 
The comparative efficiency of the fungicides used is best shown in the 
graph (fig. 4) in which curves are used to represent the percentages of 
moldy spikes left at the end of each week, under the different treatments. 
Only the twice-a-week applications are compared here. In so far as 
control of the mold is concerned, the sulphur applications appear to be 
