154 
PACIFIC SCIENCE, Vol. I, July, 1947 
an agricultural country recognize. In an effort to 
determine how much arsenic plants might absorb 
and remain normal in appearance, Machlis (1941) 
grew bush beans and Sudan grass in culture solu¬ 
tion. He presented striking evidence that the bean 
plant may absorb arsenic far in excess of legal 
tolerance and yet show no reduction in growth. 
To discover the factors which may or may not 
make the use of arsenic-contaminated food safe 
for animal or human consumption requires further 
critical studies. On the basis of recent work it 
appears, however, that there has been more emo¬ 
tion and less knowledge about this subject than 
is needed for an understanding of it. 
Recent work: Arsenic accumulation in soil, its 
effect on crop production, and corrective measures. 
—Swingle (1923), in an effort to test the effect of 
prolonged application of arsenic on plant growth, 
applied various arsenicals to plots of ground in 
the spring of the year, and the effect on crop pro¬ 
duction was noted. After 7 years of such proce¬ 
dure, beans and cucumbers made little growth, 
while wheat and timothy grew fairly well. No 
further applications of arsenic were made for 6 
years and at the end of that time it was found 
that very little of the arsenical had been removed 
by rains or irrigation. Furthermore, it was found 
very difficult to get the land back into condition 
for cropping. 
Paden and Albert (1930), working in South 
Carolina, reported a relationship between the un¬ 
productivity of certain soil types and the accumu¬ 
lation of soluble arsenic in the soil resulting from 
heavy applications of calcium arsenate. Lime im¬ 
proved the growth of cotton in the poisoned soil. 
Soils which were relatively low in iron and other 
materials which would be expected to render ar¬ 
senic insoluble were found to be the most seriously 
affected by the arsenates. 
Albert and Arndt (1931) found the concentra¬ 
tion of soluble arsenic as measured by collodion 
bag dialyzates to be a more reliable index of ar¬ 
senic than is the total arsenic present in the soil. 
In greenhouse experiments, the addition of 1 ppm 
of arsenic definitely retarded root and top growth 
of cowpeas. It was observed that the concentra¬ 
tion of 1 ppm of soluble arsenic as measured by 
the collodion bag test was not unusual in soil 
which had been receiving doses of arsenates. Lim¬ 
ing and the use of fertilizer along with iron and 
clay compounds of the soil played an important 
role in rendering arsenates harmless to sensitive 
crops. 
Hurd-Karrer (1936) believed as a result of field 
tests and culture solutions that phosphate applica¬ 
tion will reduce or prevent arsenic injury to plants 
where the soil type is such as to retain the phos¬ 
phate in available form. 
Heggeness (1940), growing tomatoes in culture 
solution, found no evidence of toxic stimulation, 
even the most dilute solution (Y 2 ppm) reducing 
the yield by 20 per cent. Arsenic toxicity in the 
tomato was partially dependent on phosphate 
availability. 
Keaton (1938) studied the oxidation-reduction 
potentials of arsenate-arsenite systems in sand and 
soil mediums. In these soils arsenic was fixed by 
absorption and combination, but it was observed 
that a higher percentage of arsenate than arsenite 
was fixed by these soils. The addition of iron to 
the system when the original ratio of arsenate to 
arsenite was unity increased the redox potential 
independent of the medium used. In the two soils 
studied the colloidal fraction possessed a greater 
reducing capacity and a lower potential than the 
soil from which it was extracted. 
Keaton and Kardos (1940) attempted treat¬ 
ment of orchard soil to overcome toxicity of arse¬ 
nic residues. Their studies indicate a relationship 
between the oxidation-reduction potentials of the 
soils treated and the conditions of plant growth. 
The addition of ferric oxide caused an increase in 
the redox potential. Alumina produced no effect 
in oxidation or reduction. Soils with a high colloid 
content were characterized by low potential and 
small percentage oxidation. They suggested that 
poisoned soils be treated with some mild oxidizing 
agent capable of arsenic fixation, and named iron 
oxide as such an agent. 
Kardos, Vandecaveye, and Benson (1940) re¬ 
ported that severely toxic soils have been rendered 
productive by making applications of 3 to 4y 2 
tons of ferrous sulfate per acre. 
EXPERIMENTAL WORK 
In order to study certain phases of arsenic 
toxicity, the experiments reported in this 
paper were undertaken. Data will be pre¬ 
sented and discussed under the following 
headings: 
Part I. The comparative toxicity of triva- 
lent and pentavalent arsenic on 
bean (Phaseolus vulgaris L.), Su¬ 
dan grass (Sorghum vulgare Per- 
soon var. sudanense [Piper] 
Hitch.), and tomato (Lycopersicon 
esculentum Miller). 
Part II. The effect of different phosphorus 
levels on the toxicity of trivalent 
and pentavalent arsenic on bean, 
Sudan grass, and tomato. 
Part III. The influence of repeated crop¬ 
pings of bean, Sudan grass, and 
tomato on a red residual soil and 
