16 



The preceding table shows the red soil to have absorbed the 

 most arsenic, while the results in the lysimeters show the organic 

 soil to have the highest absorbing power. This is due to the fact 

 that the arsenic solution acts more slowly upon the physical 

 properties of the red soil and hence in the lysimeters penetrate? to 

 a greater depth before fixation is complete. Further proof of this 

 may be had by reference to the tables showing effect on capillary 

 rise and fall of water. The movement in the red soil is influenced 

 least by the arsenic. 



Summary. 



1. Arsenite of soda has been used as a spray for killing weeds 

 in Hawaii for about five years. 



2. As a means of eradicating these pests it has proved cheap 

 and effective. 



3. Two formulas may be used in preparing the spray. One 

 results in a neutral sodium arsenite, the other in an acid sodium 

 arsenite. The former is the more economical, adds less arsenic 

 to the soil, and is equally efficient as a weed destroyer. 



4. Plant growth is very sensitive to this poison, the resistance 

 varying with the different plants and the different types of soil. 

 It is assimilated by the plant and on analysis of the dead plants 

 may be found scattered throughout the tissues. 



5. The mechanical condition of the soil is materially altered 

 by its presence. ' Its action is primarily that of a defloeculating 

 agent, thus checking the movement of the water. 



6. The influence upon the soil bacteria varies considerably in 

 the different types of soil and no general rule seems to apply 

 within reasonable limits. 



7. This chemical is strongly fixed by the soil even resisting 

 the washing of excessive rains, and is shown to accumulate in the 

 top layer. A sample of soil taken from a tract of land sprayed 

 for five years, on analysis showed all the arsenic to be present in 

 the top four inches of soil. 



While there is no immediate danger from the use of arsenite 

 of soda in killing weeds it is imperative that it should not be used 

 in excessive amounts if, as these experiments indicate, the accumu- 

 lation in the top few inches of the soil is to continue indefinitely. 

 REFERENCES. 



1 Press Bui. 30, Haw. Exp. Sta. 



2 Colorado Exp. Sta. Buls. 131 and 157. 



s Bur. Chem. U. S. Dept. Agri., Bui. 39. 

 * Press Bui. 30, Haw. Exp. Sta. 

 sBiochem. Bui. 3 (1913), pp. 2-16. 



