ERADICATING THE COMMON BAEBEEEY 11 
without much success. Rumbold (12, 13) injected various chemicals 
into chestnut trees in an attempt to kill the chestnut blight. Although 
the blight was not controlled by the treatments, the effects of the 
injections on the host were of great interest. 
In reviewing the literature it is noticeable that most of the work 
has been done with substances used as sprays and with the intention 
of killing the tops of the plants. Very little has been done with 
poisons applied to the roots. Even with perennials that sprout 
readily from underground portions, attempts at killing usually are 
by means of sprays applied to the tops, as it is too costly to treat the 
soil sufficiently to kill all vegetation. The soil of walks, tennis courts, 
and similar areas has been treated to kill off a few annuals and pre- 
vent the establishment of other plants. A relatively small dose was 
sufficient. Rudolfs (11) recently has reported experiments in clear- 
ing land of brush and stumps by heavy applications of salt or 
sulphur, and a few other attempts of this nature have been made. 
Mention should be made particularly of the recent work of Regan, 2 
of the Office of Blister-Rust Control, Bureau of Plant Industry, 
United States Department of Agriculture. In the attempt to find an 
efficient, economical method of killing Ribes plants he experimented 
with several chemicals. He made some tests with barberry plants 
also, using sodium-arsenite solution, dip oil, fuel oil, sodium-chloride 
solution, and dry sodium chloride. He found that when the chemical 
was applied to the base of the plant it was more effective than when 
applied as a spray to the aboveground portions. He states that 
" application of the chemicals to the cut-off stumps appears to be 
most effective in destroying the plants, but application to the base 
of the plants with the tops on gave practically as good results and 
involved much less labor." 
REQUIREMENTS FOR A SATISFACTORY CHEMICAL 
A chemical satisfactory for use in eradicating the common bar- 
berry must fulfill the following requirements, listed in the order of 
their importance : 
Effectiveness. — The chemical must be effective in killing barberry 
bushes under varying soil and weather conditions and at most sea- 
sons of the year. It should be 100 per cent effective under favorable 
conditions and should approach this even under adverse conditions. 
Availability. — The chemical should be readily available ; that is, 
it should be some substance which is in more or less common use, 
•so that the supply will not be limited and the price subject only to 
the law of supply and demand. 
Cost. — The cost of the chemical and of its application per bush 
should be less than the cost of equally effective digging. 
Ease of handling. — The quantity needed for a single bush should 
not be excessive, nor should the method of application be difficult 
or require special tools. 
Safety. — The substance should be such that, little or no danger 
will be experienced by those handling it or by persons or animals 
having access to treated bushes or to the chemical itself. It also 
2 Regan W. S. final report on tests with chemicals for destroying barberry 
plants. Unpublished manuscript. 
