10 BULLETIN 1451, U. S. DEPARTMENT OF AGRICULTURE 
But where this method is not practicable only two methods can 
safely be recommended. One is the actual digging of all the roots. 
This has been described already, and while it is expensive and may 
need to be repeated several times, it will lead ultimately to complete 
eradication. The other method is the use of a chemical. 
CHEMICAL METHODS OF ERADICATION 
A review of the literature shows that the -use of chemicals as 
herbicides has been practiced for many years and that many dif- 
ferent chemicals have been recommended at one time or another. 
Although a great deal of work has been done with certain chemicals 
and with their influence upon certain groups of plants, most of it 
has been of an empirical nature. Very little is known of the action 
of the chemicals upon the plant or the methods by which killing is 
effected. 
Certain chemicals were reported to be very effective in killing 
certain plants when applied in a definite manner. Without knowing 
the laws governing the action of the chemical on the .plant, one could 
only guess what the result of applying any certain chemical to a 
barberry might be. It was necessary actually to try a chemical to 
find out what it would do to the barberry. Moreover, it was neces- 
sary to apply it by different methods in varying quantities and at 
different seasons under different soil and climatic conditions. 
One of the earlier reports of the use of a chemical in killing weeds 
is by Jones and Orton (?) in Vermont in 189 7. They found salt 
effective in killing the orange hawkweed and mention the fact that 
this chemical " has long been used in fighting weeds." In 1899 they 
(8) reported on the use of salt, copper sulphate, potassium sulphide, 
kerosene, sodium arsenite, and crude carbolic acid, and concluded 
that the last two were the most effective in keeping down weeds in 
tennis courts and along walks and drives. 
About this same time Dusserre (4) in France reported experiments 
with copper sulphate, iron sulphate, and sodium nitrate in killing 
mustard in grain fields. Bolley (2) in Xorth Dakota reported suc- 
cess in controlling weeds in grain fields with a copper-sulphate spray. 
In the next few years many papers reported the effectiveness of 
different chemicals in controlling weeds in grain fields. Ferrous 
sulphate and copper sulphate were used most commonly, but sul- 
phuric acid, phenol, sodium chloride, sodium nitrate and other fer- 
tilizers, sodium cyanide and cyanamide, and other chemicals were 
tried. All these materials were used as sprays or as dusts, and their 
effectiveness was due to killing the tops of the weeds. 
Experiments on the eradication of other plants also were reported 
from time to time. In 1902 Potts (10) suggested pouring waste sul- 
phuric acid from milk testing about the main roots of briers, black- 
berries, wild roses, etc. In the same year Valder (17) found that a 
solution of sodium arsenite was very effective in killing the tops of 
prickly pear, but sprouts later developed from the roots. He also 
injected copper sulphate, sulphuric acid, carbolic acid, potassium 
oxalate, potassium ferro cyanide, sodium arsenite, caustic potash, 
iron sulphate, and common salt into the lower portions of the plants 
