SOME EFFECTS OF POTASSIUM SALTS
ON SOILS |

~ A THESIS

PRESENTED TO THE FACULTY OF THE GRADUATE SCHOOL

- OF CORNELL UNIVERSITY FOR THE DEGREE OF

DOCTOR OF PHILOSOPHY

BY

RAYMOND STRATTON SMITH

SEPTEMBER, 1918

Reprinted from Mernoir’35, June, 1920, of Cornell University Agricultural
Experiment Station, ;

SOME EPrECIS«OF POTASSIUM SALTS
ON SOILS

A THES!S

PRESENTED TO THE FACULTY OF THE GRADUATE SCHOOL

OF CORNELL UNIVERSITY FOR THE DEGREE OF

DOCTOR OF PHILOSOPHY

BY

RAYMOND STRATTON SMITH

i}

SEPTEMBER, 1918

Reprinted from Memotr 35, June, 1920, of Cornell University Agricultural
Experiment Station.

LIBRARY OF CONGRESS

seceyan

MAR 261921

a
DOCLMENTE UIVISION

CONTENTS

PENS TG Le AAU acre tes ery ent ie, Seals oe aca cs. Fak ere Sache wee wa
Effect of potassium and manganese salts on plant growth........

Potassium salts

Effect of reaction of the soil on nitrification
Interchange of bases

Experimental work
Soils used

Plan of the experimental work

Nitrification

Experimental results

Nitrification

Summary
Literature cited

567

DANS Oe AN FY oc 4 le Bac ba as Oe Oden eRe te hn 5,3 Bk es
Effect of potassium and manganese salts on nitrification in soils. .
Preece HAS ce oa on fonk soe eee ay de Saws cs Gam oAge Seid
VIM AMERC SALINAS y 24008 » tog oe eae oe te ad ae Die tae eam

Cop ales NE cike)s <5 a a oc ee ee ea en a ae eee
Preparation Of toe Bouse. 23sec peens eos 1eFhu these vede sus
Experimental methods............0..0.0 000. c ce cece ees
Pet ets, bec? i device x’ a.a a bloatellc ta Saheb a bares das wack
Soil extract cultures. ..c. can «osaaneadddewoblscaba<scece cw els

Soll acidity AeverminatilON; seis ad casas Vues Sadaces ace ec vars

Bete) Caen Oe Ws 5 La nk ge crekie d ach Baal de cae fs wha
Sieibigey date c Ulin h , = a ee ee a es a er
Piienchanve Or Waseda ny oo gale sco Vana heroedeaee ele sasdee 6 edu

08 6 el One SS. eS we eke Rw! 8! o: 0 “er el ecle, 4.8 a ce Ble ae Welw 40.89 E) Bice ew oh 4 S18

PAGE
57]
571
571
573

. O74

rn

574
576
578
579
580
581
581
581
582
584
584
584.
584
585
585
585
587
590

. 595

599

SOME EFFECTS OF POTASSIUM SALTS ON SOILS

SOME EFFECTS OF POTASSIUM SALTS ON SOILS
R. S. SMITH

The factors that determine the ability of a soil to support plant growth
are known to be very complex, and any modification of this ability
brought about by materials added to the soil is at least equally complex.
It is now generally recognized that the secondary effects of fertilizing
materials which are added to a soil may ultimately prove either beneficial
or injurious when measured in terms of crop yields. The deleterious
effects of ammonium sulfate have been particularly noted. The secondary
effects of other fertilizer salts have been less thoroly studied because
their action is thought to be less pronounced. However, attention has
been called to various effects exerted by other materials, including the
salts of potassium. The somewhat conflicting experimental data bearing
on the effects of the chloride and the sulfate of potassium on the soil as
a medium for plant growth led to the work reported in this paper.

The method of attacking the problem was, first, to determine the effect
of various applications of potassium chloride and potassium sulfate on
the growth of wheat, both in the variously treated soils and in water
extracts of the soils; and secondly, to attempt to get at the causes of the
effect of these salts on crop growth as noted in this work and as noted
by other investigators.

HISTORICAL REVIEW
EFFECT OF POTASSIUM AND MANGANESE SALTS ON PLANT GROWTH

Potassium salts

The stimulative action on the growth of the higher plants exerted by
the salts of potassium which are commonly used as fertilizing materials,
is recognized. That this action is in part secondary seems evident from
the fact that the specific effects noted vary with different soils and with
the same soil variously treated.

Ordinarily, these salts would probably not be used in sufficient quantity
to prove directly harmful to plant growth; but under certain intensive
systems of farming, in which heavy applications of fertilizers are made,

571

572 R. S. Smite

such a result might follow. Lyon, Fippin, and Buckman (1915) make
the statement that ‘it [potassium] may be present in large quantities
in the soil and yet exert no harmful effect on the crop.” Whether this
statement refers to soluble salts of potassium added to the soil, or to the
slowly soluble compounds in the soil minerals, is not stated. There is a
possibility, however, that even the ordinary applications of potassium
salts may result in an increased loss from the soil of other bases, particu-
larly calcium.

But little work has been done to determine at what concentration the
salts of potassium become toxie to plant growth in soils. Headden
(1915) found that yellow-berry in wheat is increased by the application
of 150 pounds of potassium to the acre. He ascribes this condition to
the excess of soluble potassium over soluble nitrogen. This effect of a
comparatively small application of a potassium salt in aggravating an
abnormal condition in the wheat crop is of interest in this connection in
that it indicates a significant modification of the soil as a medium for
plant growth.

Harris (1915), from an extensive investigation of the effect of alkali
salts on the germination and growth of seedlings in three different. soils,
reports the concentrations of potassium chloride and potassium sulfate
at which these salts become harmful to wheat seedlings. He found that
heavier applications of these salts were required to cause injury to the
seedlings than would ever be applied, even in the most intensive systems
of farming.

MeCool (1913) determined the effect of the chlorides of ammonium,
magnesium, potassium, and calcium on the germination of pea seeds in
soil. The salts were harmful in the order given. Potassium chloride
caused slight injury when used at the rate of about 7456 pounds in
2,000,000 pounds of soil. The character of the soil is not stated by the
writer.

Voelcker (1909), in conducting pot experiments with wheat at the
Woburn experiment station, in which the chloride, the sulfate, the car-
bonate, and the nitrate of potassium were used in such amounts as to
supply the soil in each case with 0.0075 per cent of the metal potassium —
which is equivalent to 166 pounds of the chloride, 312 pounds of the
sulfate, 248 pounds of the carbonate, and 369 pounds of the nitrate, per
2,000,000 pounds of soil — noted injury with the carbonate. It is difficult

SoME Errects oF Potassium SALTS ON SOILS 573

to understand how such small applications of any of these salts could
cause injury to wheat.

Much work has been done on the toxicity of bases in solution cultures
with various crop plants. This phase of the study is typified by the
investigations of McCool (1913) on a large number of bases, including
potassium. This type of investigation, however, has little direct sig-
nificance in soil studies, because the conclusions drawn cannot be applied
with a soil medium due to side reactions which are involved when so
complex a medium is employed. McCool found that the chlorides of
barium, manganese, ammonium, magnesium, sodium, potassium, and
calcium were toxic to pea seedlings in the order named. It is of interest
to note that manganese stands near the head of the list.

The degree of toxicity of all the salts is much less in soil than in nutrient
solution. As has been noted, McCool found that potassium chloride
caused slight injury to the germination of pea seeds in soil when applied
at the rate of about 7456 pounds to 2,000,000 pounds of soil, and Harris
reports much higher concentrations than this as being necessary to produce
a toxic condition except in the case of coarse sand.

It thus appears that injury to plant growth has been found to result
from the use of potassium salts in large quantities; that applications at ‘
the ordinary rate have been found to cause injury in but one case; and
that small applications may possibly accentuate pathological conditions
in the growing plant.

No reports of experiments on the growth of seedlings in soil extract
made from soils to which only potash salts had been added, have been
found in the literature.

Abbott, Conner, and Smalley (1913) report some soil-extract-culture
experiments with corn, using soil high in soluble aluminium salts. This
work is of interest in this connection in that it agrees with the con-
clusions of other investigators that the water extract from unproductive
field soils is toxic to the root growth of seedlings.

Manganese salts

Manganese, as is noted later in this paper, is one of the bases replaced
by potassium in some soils, and, since it has been shown by some investi-
gators to have considerable influence on plant growth, a brief review
of the literature regarding its action is here given.

574 R. S. Smita

As shown by Skinner and Sullivan (1914), manganese increases the
oxidizing power of plant roots. This, however, was not accompanied
by increased growth when the plants were grown mm fertile soil. Infertile
soils seemed to respond to manganese when it was used in small quantities,
varying from 5 to 50 parts of manganese to 1,000,000 parts of soil.

Experiments by Skinner and Reid (1916) on silty clay loam of an acid
nature at Arlington, Virginia, in which manganese sulfate was applied
annually at the rate of 50 pounds to the acre previous to planting, show
a decrease in the yield of wheat and cowpeas and inconsistent results
with rye. When the lime requirement of the soil was just satisfied, the
depression was decreased; and when an excess of lime was used, the crop
yields were increased by manganese, except in the case of potatoes.

The results of other workers agree in the main in showing that the salts
of manganese increase the yields of field crops when used in small quanti-
ties. in some instances a decrease results, and the work of Skinner and
Reid seems to indicate that the reaction of the soil is an important factor.

Little work has been done to determine at what concentrations salts
of manganese become toxic to plant growth in soils. McCool (1913),
using a sandy loam soil, found that manganese chloride in solution was
toxic to peas when added at the rate of 330 cubic centimeters of N/50
solution to 1000 grams of soil. This rate of application is equivalent to
about 181 parts of the element manganese to 1,000,000 parts of soil.
McCool found also that calcium overcame this injurious action, while
Kelley (1908) reports that lime had no such effect.

The conclusion seems justified that if a neutral salt when added to
the soil replaces even small amounts of manganese, the presence of this
replaced base may affect crop growth adversely or occasionally beneficially,
depending on other factors not well understood.

Other bases replaced by various fertilizer treatments are known to
be toxic to plant growth, particularly iron and aluminium. But since
neither of these elements was found to be present in the water extracts
of the soils used in this work, no discussion of their action seems necessary.

EFFECT OF POTASSIUM AND MANGANESE SALTS ON NITRIFICATION IN SOILS
Potassium salts

Potassium salts have been found to produce specific effects on nitrifica-
tion. Under certain conditions a stimulation has been noted, while under

; PECTS OF SALTS ON SOILS Dia
SomE EFrects oF PoTrassiuM §S

other conditions the reverse has been the case. The literature has been
searched in an attempt to discover what effects potassium salts have
on nitrification, and any specific effects that have been found to accompany
certain conditions resulting from the application of potassium salts to
soils. A close correlation between the nitrifying power of a soil and its
crop-producing power may not exist, but the two are likely to be associated.
A study of the nitrifying power of a soil should, then, furnish some
indication of its crop-producing power and help to explain any departures
from the normal in crop growth.

Dumont and Crochetelle (1893) report favorable effects of potassium
salts on nitrification in soils rich in organic matter and limestone. They
later (1894) report work with a sandy humous soil stated to be poor in
lime. This soil as reported contained 17.5 per cent of humus and 0.285
per cent of limestone. Potassium carbonate and potassium sulfate, both
with and without lime, were used in different amounts. Potassium
carbonate was used in increasing amounts from 0.1 to 6 grams, to 100
grams of soil. Marked stimulation was found to accompany its use
up to 4.5 grams, and then there was a steady decrease in the nitric nitrogen
found. Potassium sulfate without lime had no consistent effect. The
character of the results indicates that the differences found in the latter
treatments were due to factors other than those under study. When
2.5 grams of limestone were applied in addition to the potassium sulfate,
there was a constant increase in the amount of nitric nitrogen found
with an increase in the amount of potassium sulfate used, the heaviest
application being 5 grams to 100 grams of soil.

Lyon and Bizzell (1918) report the nitrogen recovered from the lysimeter
tanks at Cornell University. Apparently potassium sulfate without
lime depressed nitrification. Lime counteracted this effect, but even
with lime the sulfate did not cause any appreciable stimulation of the
process.

Greaves (1916), in laboratory experiments on the effect of potassium
salts on the bacterial activities of sedimentary soils derived from limestone
and quartzite, found that potassium chloride and potassium sulfate used
at the rate of from 6.1 to 8602 parts per million depressed nitrification
at all concentrations. Potassium nitrate and potassium carbonate, used
at the same rates, stimulated nitrification at the lower concentrations |
but became toxic at the higher, the nitrate at 48.9 parts per million and

576 R. S. SMITH

the carbonate at 3910 parts per million. Greaves concluded that the
extent of stimulation is governed largely by the cation, and that the
toxicity of potassium salts is governed by the electro-negative ion com-
bined with the potassium, since he found that the chlorides of sodium,
magnesium, manganese, and iron, and the sulfates of calcium and manga-
nese, increased bacterial activity, while the chlorides of potassium and
calcium and the sulfates of sodium and potassium failed to cause any
stimulation.

Pichard (1884) found that potassium sulfate caused strong nitrification
of the organic nitrogen in a soil high in organic matter, but that its influ-
ence was not so marked as was that of calcium sulfate or sodium sulfate.

Allen and Bonazzi (1915) studied nitrification in soil samples from
the plots at the Ohio experiment station. Ammonium sulfate in solution
was used as the nitrifiable material at the rate of 21.2 milligrams of
nitrogen per 100 grams of soil. The samples were incubated for ten days.
The results with the samples from the potassium sulfate plots — which
had received 80 pounds of the salt to the acre on corn, oats, and wheat
of the five-years rotation — failed to show any increase in nitrification
over the check; in fact, denitrification apparently took place in some cases,

Peck (1911) found that potassium sulfate used at the rate of 0.5 gram
in 500 grams of sugar-cane soil decreased the bacterial activity, as measured
by bacterial numbers and nitrogen fixation during one month of incubation.

Renault (1910) cites experiments by Dumont which show that slow
ammonification and subsequent nitrification is always accompanied by
a low percentage of potash.

It thus appears that potassium fertilizers, when applied at the usual
rate under field conditions, commonly exert a depressing influence on
nitrification. Under laboratory conditions both the chloride and the
sulfate of potassium have generally been found to exert a depressing
effect on nitrification, even when used in amounts as small as 12 pounds
to 2,000,000 pounds of soil. Lime apparently counteracts the injurious
effects of small applications of the sulfate and permits some stimulation of
the process of nitrification.

Manganese salts

Salts of manganese are known to have marked influence on nitrification,

and since manganese, as has already been stated, is one of the soil bases

Some Errects or PorasstuM SALTS ON SoILs Dit

replaced by potassium, it is of importance in this connection to note what
its effects have been found to be.

Kelley (1912), working with Hawaiian soils, found that those high in
manganese had a stronger nitrifying power than those low in this element.
However, the soils high in manganese were in a better physical condition,
and their higher nitrifying power was attributed to this fact rather than
to any difference In manganese content.

Montanari (1914) found that manganese dioxide and manganese car-
bonate apparently stimulated nitrification, while the sulfate exerted less
stimulation or even depressed the process.

Leoncini (1914) found that manganese dioxide increased nitrification
when used in amounts as high as 2.2 per cent, but that heavier applications
apparently had no influence.

Brown and Minges (1916) determined the effect of various manganese
compounds on nitrification and ammonification in Carrington clay loam.
In the ammonification tests dried blood was used, and in the nitrification
trials ammonium sulfate was used. Manganese chloride apparently had
no effect on nitrification in amounts less than 0.5 per cent; but from that
point on, increasingly heavy applications caused increased depression,
until, with 5 per cent of the salt, nitrification was inhibited. With
manganese sulfate there was decisive depression of nitrification when
0.5 per cent of the salt was used, but with increasingly heavy applications
the results did not show an increasing depression. Manganese nitrate
apparently depressed nitrification, the magnitude of depression increasing
with the amount of the salt used. Manganous oxide in most. cases
depressed nitrification, altho definite conclusions regarding this point
cannot be drawn from the data presented.

Greaves (1916) found the chloride, the sulfate, and the nitrate of
manganese toxic to ammonification in soil at concentrations of 68.6,
137.3, and 274.6 parts of added manganese, respectively, to 1,000,000
parts of soil. The carbonate of manganese was without effect even at
the highest concentration used, 6045.6 parts per million.

Olaru (1915) reports three experiments on nitrogen fixation in nutrient
solutions with varying amounts of manganese. He found that stimulation
of the process resulted from all the concentrations of manganese used,
but that the proportion of 1 part of manganese to 200,000 parts of
solution gave the greatest stimulation. Olaru suggests that increases in

578 R. S. Smit

crop yields which have been found to follow the use of fertilizing materials
are due, not only to the direct action of the materials on the plants, but
also to their modification of the bacterial activities of the soil.

There appears to be much conflict in the data cited regarding the effect
of manganese compounds on nitrification. In some cases very low con-
centrations of the various salts proved to be toxic, while in others relatively
high concentrations were stimulative. Too little information is given
regarding the nature of the soil used in the various experiments to permit
any attempt to account for the discrepancies.

EFFECT OF REACTION OF THE SOIL ON NITRIFICATION

The reaction of the soil is generally considered to be an important
factor in determining its capacity to support a vigorous nitrifying flora.
Brown (1911:55) apparently takes an extreme position when he says:
“ The effect of lime on nitrification and the necessity for the presence
of lime in the soil for the process to occur, have long been a matter of
common knowledge.”

The literature bearing on this problem is voluminous and no attempt
is made here to summarize it. The stimulating action of lime on nitrifi-
cation is generally conceded, but apparently the process may go on
in soils very deficient in lime.

Fred and Graul (1916) state that ‘it seems that under laboratory
conditions, the beneficial effect of calcium carbonate on plant growth
must be accounted for by some processes other than the direct effect on
nitrification.” Temple (1914) and White (1914) report vigorous nitrifica-
tion in strongly acid soils.

In the work herein reported, the heaviest treatments with the chloride
and the sulfate of potassium caused a slight increase in the lime require-
ment of the soils, but in no case was the increase more than 300 pounds
of caleium carbonate to 2,000,000 pounds of soil. This smail difference
in reaction is not considered significant so far as nitrification is concerned,
particularly in view of the fact that nitrification has been shown to proceed
in strongly acid soils. The increasing depression in nitrification which
will be shown to have accompanied increasingly heavy applications of
the potassium salts must be accounted for on some basis other than
increased acidity.

Some Errects oF Potrasstum SALTS ON SOILS 579

INTERCHANGE OF BASES

As stated by Sullivan (1907), the fact that water is purified by filtration
thru sand was known in the time of Aristotle. That common salt ean
be removed from water by filtering the water thru sand or soil has likewise
been known for many years. Hilgard (1911 : 267) states that the latter
is a clearly physical effect. When neutral salt solutions are filtered thru
soil, the filtrate may be either acid or alkaline, depending on whether
the cation or the anion of the salt has been removed the more strongly.
This phenomenon has been attributed to selective ion adsorption. Truog
(1916) and Sullivan (1907) think that it is better accounted for by an
exchange of bases, in which the base of the soluble salt interchanges in
part with the iron or the aluminium of the soil. The salts of the latter
metals hydrolyze strongly in dilute solution and give an acid value.

The fact that soils enter into a chemical exchange with salt solutions
was recognized at an early date by Thompson (1850), who found that
an ammonium sulfate solution filtered thru soil gave up its ammonium
in exchange for calctum. Way (1850, 1852, 1854), in a number of experi-
ments, extended the observations of Thompson and found that the
nitrates, the chlorides, and the sulfates of ammonium, potassium, sodium,
and magnesium, when filtered thru soil, exchanged their bases for calcium
from the soil. Way concluded that the active constituent of the soil
entering into this interchange was a hydrated alumino-silicate of the clay
fraction. It is now thought that any silicate is capable of entering into
these reactions, according to Sullivan (1907).

Peters (1860) found that the absorption of the cation of a salt in neutral
solution was of about the same magnitude regardless of the form of com-
bination. Thus, he found that potassium was absorbed in about equal
amount from equivalent solutions of ifs chloride, its sulfate, and its
carbonate. In an extensive investigation Wiillenberg (1867) confirmed
Peters’ conclusion. He found that the base entered into the reaction
in about the same amount, whether it was combined with the sulfate,
the nitrate, or the chloride.

The bases are mutually replacable, but are not replaced with equal
facility. The stability of the silicate or the alumino-silicate is the con-
trolling factor. Lemberg (1870, a and b, 1872, 1876), in a series of studies,
found that the sodium in silicates is replaced more readily by potassium

580 ; R. 8. SMITH

than is potassium by sodium, and that magnesium is replaced less readily
from its silicate by calcium than is calcium by magnesium.

Van Bemmelen (1878) treated a soil with a solution of potassium
chloride, and found that the potassium had been exchanged for sodium,
calcium, and magnesium. Van Bemmelen states that the absorption of
the entire salt takes place very slightly if at all.

The important point brought out by Van Bemmelen in this early work
and reemphasized by him later (Van Bemmelen, 1900), is that colloidal
silica and silicates do not abstract and concentrate the salts from neutral
salt solutions when filtered thru soils. Any such apparent effect is due,
he believes, to a redistribution of the salt in the solution between the
water of the colloid and the water of the solution.

Joly (1902-04), Briggs and Lapham (1902), and Dittrich in 1903 (cited
by Sullivan, 1907:26) also have presented data tending to show that the
action of neutral salt solutions on soils consists in an equivalent exchange
of bases.

Ruprecht and Morse (1917) report the presence of soluble salts of iron,
aluminium, and manganese in soils repeatedly dressed with ammonium
sulfate without the addition of lime.

It thus appears that neutral salts of potassium when added to the soil
are strongly absorbed, thus resulting in the liberation of other bases which
may have either beneficial or harmful effects on plant growth. These
effects may be due to some direct effect of the replaced bases on the
plant’s activities, or they may be induced indirectly by the modification
of some of the soil’s properties.

CONCLUSIONS

It appears from this summary of the literature that the common
fertilizer salts of potassium have usually been found to exert harmful
effects on plant growth only when used in large quantities. These effects
may be accounted for in part by basic exchange, in which case the com-
position of the soil would be an important factor. Significant modifications
of the bacterial activities in the soil may be another factor. In the
following pages are reported the results of experimental work which was
designed to throw light on these problems.

Some Errects or Potassium SALTs on Sorts 581

EXPERIMENTAL WORK
SOILS USED

Three soils were used in the experiments here reported — Hagerstown
silt loam, Dekalb silt loam, and Volusia silt loam.

Hagerstown silt loam is a residual soil derived from limestone. It is
known as a productive soil, and has good surface drainage and good
underdrainage. The sample was collected near State College, Penn-
sylvania, from an old field which had never been fertilized in so far as
could be learned. In collecting the soil the immediate surface was scraped
off and the soil was taken to a depth of eight inches.

Dekalb silt loam is a residual soil derived from sandstone and shale.
Its productivity is considered as poor to medium. It is typically poorly
underdrained. The sample was collected from an abandoned field near
Snow Shoe, Pennsylvania, in the same manner as was the Hagerstown
silt loam.

Volusia silt loam is a glacial soil composed of a small proportion of
glacial material mixed with soil material derived from local sandstone
and shale. There is such a wide variation in this soil that it cannot
be characterized as a series. The sample was collected near Ithaca,
New York, from an unproductive, poorly underdrained field, the same
method being used as was used in collecting the other soils.

PREPARATION OF THE SOILS

The soil samples were brought to the laboratory and immediately
screened thru a 4-millimeter screen. Small samples of the screened soils
were taken for moisture and acidity determinations, and pots were filled
with a known weight of the soil calculated to the water-free basis. After
the pots were filled, the contents of each pot were emptied on an oilcloth,
the required quantity of precipitated calcium carbonate and potassium
salt was added and thoroly mixed with the soil, and the pot was refilled.
In the case of the Volusia soil, the potassium salt was added in solution
after the calcium carbonate had been mixed with the soil and the pot
had been refilled. The pots were then brought to weight with distilled
water. Sufficient calcium carbonate was added to the Hagerstown and
Dekalb soils to just satisfy their lime requirement, which amounted to
2 tons of calcium carbonate to 2,000,000 pounds of soil in each case.

582 R. S. Smite

With the Volusia soil varying amounts of the carbonate were used, as
follows: Series I—no lime, lime requirement 3393 pounds of calcium
carbonate to 2,000,000 pounds of soil; Series IT — lime requirement just
satisfied; Series III — 2 tons of calcium carbonate to 2,000,000 pounds
of soil in excess of the lime requirement.

PLAN OF THE EXPERIMENTAL WORK

On the Hagerstown and Dekalb soils no crop was grown the first year.
Composite samples of each treatment, in triplicate, were taken for nitrate
determinations, with a ?-inch brass tube, the day after bringing the pots
to weight and at regular intervals thereafter during the first year. The
moisture content was maintained at approximately 24 per cent (water-
free basis) by bringing the pots to weight weekly with distilled water.

TABLE 1. TREATMENTS OF HAGERSTOWN AND DEKALB SittT Loam Sorts
(Lime requirement of soils just satisfied; moisture content maintained at 24 per cent)

Pounds of potassium salt to
2,000,000 pounds of soil
Pot
KCl K.SO,
7
14b f 0 0
Cc)
a |
Oe te A Caer page aie ae ea he fo eee Sey ed, SAS, ee 200 200
c
a )
3 | Ey Rieti se Raha. Scsea er nyege ebm ete adet agals s eee 500 500
c
a
AS By helo ows cara taertoch ange denice aten esi dru enaia sees Mca ites, «ao a a 1,000 1,000
c J
a
Betsy apsacehceseseta ier on coer ceeres ners eS ve peice roe em te AeA 2,000 2,000
Cc
Py
G4) Pie 2 epeie sce Wiehe pasty mpetets bias ous cuaceberalcuershe ove eeneme ack ene ae 3,000 3,000
c

SoME Errects oF PorasstuM SALTs on SOILs 583

The following year the triplicates were thoroly mixed on oilcloth, duplicate
pots were filled with each treatment, and wheat was planted and grown
to maturity thru the winter and spring. At the time of making up the
duplicate pots, samples were taken, air-dried, and stored to be used later
in determining the effect of treatment on water-soluble bases and on the
growth of wheat seedlings in the water extracts of the soils.

TABLE 2. TREATMENTS oF VouusiaA Sitt Loam

(Moisture content maintained at 30 per cent) ©

Series 1 — No CaCO; lime requirement 3393 pounds CaCO; to 2,000,000
pounds of soil

Pounds of KC] Number of pots
Pot to 2,000,000

pounds of Cropped to No crop

soil wheat grown
Lee ade Seas ann ee 0 4 1|-gallon | 1 3-gallon
2 5) SLO ea ae 200 4 l-gallon | 1 3-gallon
A ates er eee oer swal ss 500 4 1-gallon 1 3-gallon
erie ee tk Was Sia esa d 1,000 4 1-gallon 1 3-gallon
SC A ee er ee 2.000 4 1-gallon 1 3-gallon

Series I[—Same plan as Series I, with lime requirement just satisfied
S

Series If —Same plan as Series I, with 4000 pounds of CaCO to 2,000,000
pounds of soil in excess of lime requirement

In the case of the Volusia soil, quadruplicate I-gallon pots were made
up of each treatment, to be cropped to wheat, and one 3-gallon pot in
each treatment was filled to be used later in the other studies. All of
these pots were brought to weight weekly (30 per cent moisture content,
water-free basis) with distilled water. All of the laboratory determina-
tions, including soil extract cultures with wheat seedlings, were made on
samples from the 3-gallon pots, which had been kept in the greenhouse
under the same conditions as the cropped pots and maintained at an
approximately constant moisture content for about seven months.

The outline given in tables 1 and 2 makes the plan of the work clear.

584 R. S. Smrru

EXPERIMENTAL METHODS

Pot cultures

Two-gallon pots were used with the Hagerstown and Dekalb soils,
each pot containing the equivalent of 9} pounds of water-free soil. One-
gallon pots were used with the Volusia soil, each pot containing the
equivalent of 6 pounds of water-free soil. Duplicate pots were used
with the Hagerstown and Dekalb soils, and quadruplicate pots with the
Volusia soil. The Hagerstown and Dekalb soils were maintained at a
moisture content of 24 per cent, water-free basis, and the Volusia soil
at 30 per cent. These moisture contents gave approximately two-thirds
saturation.

Soil extract cultures

Soil extract.— The soil extract for the solution cultures and the analyses
for water-soluble bases was prepared by adding five parts of water to
one part of soil (after correcting for the water already in the soil), shaking
for three hours, and immediately filtering thru Pasteur-Chamberland
filters.

Analysis of extract.— The official methods of analysis of the United
States Bureau of Chemistry! were used, with the following exceptions:

Manganese was determined by the ammonium persulfate method as
deseribed by Hillebrand (1910). Calcium was precipitated according to
the official method, and titrated with potassium permanganate after
solution in dilute sulfuric acid.

Soil extract cultures— The soil extract cultures were run in duplicate.
Erlenmeyer flasks of 500 cubic centimeters capacity were used for culture
vessels. Jour wheat plants were grown in each flask by using a paraf-
fined paper cover thru which four holes were punched to receive the
rootiets. The plants were allowed to grow for four weeks. They were
then removed, photographs were taken of the roots, and the dry weights
were determined.

Nitrification

In the nitrification trials, tumblers were used for containers and 100
grams of soil was placed in each tumbler. Three nitrifiable materials —
ammonium sulfate, ammonium hydroxide, and dried blood — were used.

: 1Official and provisional methods of analysis, Association of Official Agricultural Chemists. U.S. Bur.
Chem., Bul. 107. 1912.

Some Errects or Potassrum SATs on Sos 585

The ammonium sulfate and the ammonium hydroxide were applied in dilute
solution, and the dried blood was well mixed with the soil on a piece
of oilcloth. The cultures were incubated at room temperature and were
brought to weight every six days with distilled water. Excessive evapora-
tion was prevented by covering the tumblers with a layer of cotton placed
between pieces of cheesecloth. The period of incubation, percentage
of moisture maintained, and nitrifiable material used, are shown with
each table in which the results are given.

Nitrates were determined by the phenoldisulphonic-acid method as
described by Schreiner and Failyer (1906).

Soil acidity determination

The lime requirement of the soils was determined by a modified Veitch
method (White, 1914).

EXPERIMENTAL RESULTS

Pot cultures

The crop on the Hagerstown and Dekalb soils was attacked by sparrows
on the afternoon of the day before it had been intended to harvest the
pots, and as a result only the yield of straw is given. In the case of the
Volusia soil, the probable error of the average for the quadruplicate pots
is so high as to render most of the poss:ble comparisons of questionable
value. The probable errors were computed by means of Peter’s formula
as given by Mellor (1909),

in which Y (+ v) denotes the sum of the deviations of each observation
from the mean, disregarding ther sign, and n denotes the number of
observations made.

The results of the pot experiments are given in tables 3 and 4.

Potassium sulfate increased the yield of straw over the check in both
the Hagerstown and the Dekalb soil. In the Hagerstown soil there was
a continued increase in yield with an increase in the rate of application
after 500 pounds was reached. In the case of the Dekalb soil the data
are not conclusive, except that, as with the Hagerstown soil, there is no
evidence of a toxic condition with any of the treatments.

586 R. 8. Smit

TABLE 3. Yreup or Wueat Straw 1N Por Cuttrures with HAGERSTOWN AND DEKALB
Sitt Loams
we y is

(Lime requirement of soils just satisfied)

Pounds of Yield of straw Pounds of Yield of straw
Pot IeSO, to in grams | KCl to in grams
2,000,000 | 2,000,000
pounds a aaa) 0.81 0G <M Be a ee
oreon Duplicates | Average ot soit pee Average
|

Hagerstown silt loam
4.36

4.12)

er ee 0 | re 4.31 | 0 a at 4.15
5.56 445

ee TA 200 | {2:38 5.41 cae 4.41

9

ae 500 oe 5.05 ree 4.47

ee ee 1,000 | - ee 5.77 z 1,000 a 3.93
, 6.50 3 86

ee Bi | a eee | 6.08 2000 | ae 3.94

6.69 3.11
BE dicts: aordercambcr 3,000 | se | 6.43 | 3,000 | (3-1) 3.26
Dekalb silt loam

1.85 2.02

1 OO) O66) 0'.8' OC 0 e108 (ere ae oF 0 ‘ 187 1.86 0 187. 1.95
1.82 1.98

We sats Peseta au 200 | {1-82 1.87 200 | {1-23 1.91
| £1.83 1.73

Be eas nee ee 500 | 4 7's8, 1.86 500 1185] 1.79
1.59 1.98

18a ean taunts 1,000 if 1.65 1,000 1/90} 1.94
: . 2 08 1.71

ae ee | 2,000 188 1.98 2,000 165 } 1.68

Br. aa kbatecaes | 3,000 sink - 3,000 | - ae 1.68

Potassium chloride apparently became toxic at the 1000-pound treat-
ment with the Hagerstown soil and at the 2000-pound treatment with
the Dekalb soil. The data, however, are not conclusive, and warrant
only tentative conclusions regarding the rate of application necessary
to bring about a toxic condition in these soils.

Some Errects oF PorassiuM SALTS ON SOILS 587

TABLE 4. Yreip or WHEAT STRAW AND GRAIN IN Port CuLturES witH VoLusIA SILT

Loam
J
ee of Straw Grain
Oo
Series 2,000,000 (average of (average of
pounds quadruplicates, quadruplicé ates,
Af Boil in grams) | in grams)

I S ig 6.3+0.41 3.03 + 0.10
200 6.5 + 0.45 3.14+0.18
No lime; lime requirement 3393 pounds 500 6.6 + 0.40 3.17 + 0.13
CaCO; to 2,000,000 pounds of soil 1,000 6.7+0.14 3.22 + 0.13
2/000 8.6+0.44 3.85 + 0.19
0 8.8 + 0.85 4.09 + 0.62
II 200 7.24 0.37 3:46 + 0.30
Lime requirement just satisfied 500 6.6 + 0.23 2.79 + 0.18
1,000 7.4+0.13 3.12+ 0.12
2,000 7.9+0.13 3.59 + 0.05
0 12.1 + 0.76 6.06 + 0.68
III 200 13.4 + 0.53 5.13 + 0.66
4000 pounds CaCO; to 2,000,000 pounds 500 10.7 + 0.22 2.80 + 0.44
of soil in excess of lime requirement 1,000 15.5 + 2.45 5.18 + 0.64
2,000 11.9+0.33 4.14+0.12

Soil extract cultures

Both the root and the top growth of the wheat seedlings were very
uniform in the duplicate water extract cultures. The dry weights, how-
ever, while uniform between duplicates, did not give a good measure
of the comparative root growth betweeu cultures, and consequently’ are
not reported.

The presence of some toxic substance or substances in certain of the
cultures is indicated in figures 161 to 163. The sensitiveness of the roots
of seedlings to toxie substances has been adequately demonstrated by
Schreiner and his associates in the United States Bureau of Soils, and by
Breazeale and LeClerc, of the Laboratory of Plant Physiology of the
United States Department of Agriculture.

In the extract from the Hagerstown soil (fig. 161) the chloride is seen
to have stimulated root growth thruout the series, the greatest degree
of stimulation resulting from the 500-pound treatment. In the sulfate
series there is seen a progressive stimulation of root growth up to the

588 R. S. Smire

Fic. 161. RooT GROWTH OF WHEAT SEEDLINGS IN WATER EXTRACTS FROM HAGERSTOWN SILT
LOAM WHICH HAD RECEIVED VARYING AMOUNTS OF THE CHLORIDE AND THE SULFATE OF
POTASSIUM

Fic. 162. RooT GROWTH OF WHEAT SEEDLINGS IN WATER EXTRACTS FROM DEKALB SILT LOAM

WHICH HAD RECEIVED VARYING AMOUNTS OF THE CHLORIDE AND THE SULFATE OF POTAS=-
SIUM

SomE EFrects oF PoTASSIUM SALTS ON SOILS 589

2000-pound treatment, and a marked toxicity with the 3000-pound treat-
ment. These latter results agree in the main with the yield of straw in
the pot cultures except in the case of the heaviest sulfate treatment.
In this case the extract cultures showed strong toxicity, while no such
condition was present in the pot cultures.

In the extract from the Dekalb soil (fig. 162) no distinct toxicity was
shown in any of the cultures when compared to the checks. The checks,
however, were apparently toxic. With the chloride the 200-pound treat-

Ked Serie

oom Ca COx

Al a: -- (eaanee

Fia. 163. ROOT GROWTH OF WHEAT SEEDLINGS IN WATER EXTRACTS FROM VOLUSIA SILT LOAM
WHICH HAD RECEIVED VARYING AMOUNTS OF POTASSIUM CHLORIDE, BOTH WITH AND WITH-
OUT LIME

ment caused the greatest stimulation of root growth, while with the

sulfate there was little difference between the degree of stimulation in

the 200- and the 500-pound treatments. These results are not reflected
in the yields from the pot cultures. The yields from the pots were very
small and the final weights are probably not a good index of the relative
vigor of growth.

In the Volusia extract cultures (fig. 163) the important point brought
out.is the neutralization of the toxic condition by the calcium carbonate.

590 R. S. Smira

A distinetly toxie condition was evident in the no-lime series with the
heavier chloride treatments. This condition was less pronounced in the
series receiving enough lime to just meet the lime requirement of the soil,
and almost entirely disappeared when 4000 pounds of lime to 2,000,000
pounds of soil in excess of the lime requirement of the soil was used.

Nitrification

As a measure of the activity of the nitrifying organisms in the variously
treated soils, determinations were made of the nitrates accumulated over
long periods of time and of the nitrification of added materials. The
accumulation of nitrates in the three soils used is shown in tables 5, 6,
and 7, respectively. The figures represent the milligrams of nitrate
nitrogen in 100 grams of soil as determined when the pots were set up
and at stated intervals thereafter. The difference between the initial

nitrate content and that after a given interval represents the actual

TABLE 5. AccumMuLATION oF NITRATES IN Hacerstown Sriitt Loam Aas DETERMINED
AT INTERVALS AFTER THE EXPERIMENT WAS SET UP

(Moisture content, 24 per cent)

Milligrams of nitrogen as nitrates in 100
grams of soil

Pounds of potassium salt |

2 i i
to 2,000,000 pounds of soil ee aa After Npiee Afton
up experi- an} days 61 days 86 days
ment
(KCI)

0 re a ee gee 1.90 2200 33 5a83 3.10
DOOR ae eats sete ies ohn eae QohZ, 1.86 S07 2.86
15100 ene Ne yale ene ea 1.66 1.80 2.91 3.05

1 Ve 600 ener eer to ar ete ae aun ey Sn 1.50 1.94 Zoe 2.94
el 0,0 0 ean Gea ote eur enone anne taPy Trace 2.09 3.02 2.82
55 CMO reragestes ee daeeee ea ere 1.41 2.04 3220 2.83

(IK2S04)

EAs reser a ish a vstrs abaenchea di dat ect cronet teenies ete 1253 2.06 3.88 may toy

DOD Beene aiden seca ceed ae ape eae ae 1.96 2eod ouce 3.60
BOC etree teh ey Sembee tte A a 1.68 2.83 200 teat,
1000 eae eee tie eee es ee ee 1.69 2.61 3.89 4.06
PAU) eeaneey eon APNG} Dee 3.81 4.40
SOOO a asa eee eee 1.46 2.54 3.97 4.48

Some Errects or Porasstum Satts on SorLs 591

accumulation, or, in some cases, loss. In the case of the Volusia soil,
nitrate accumulation was determined but once, after a seven-months
period. :

TABLE 6. AccumMuLaTION oF Nitrates iN DexKAtB Sitr Loam As DETERMINED AT
INTERVALS AFTER THE EXPERIMENT Was SET Up

(Moisture content, 24 per cent)

——

Milligrams of nitrogen as nitrates in 100
grams of soil
Pounds of potassium salt
to 2,000,000 pounds of soil At time
of setting After After After
up experi- | 33 days 61 days 86 days
ment
(IxCl)

se es Si arsi key adh a brs Trace 0.65 1.14 1.69

PAULO) Sere ois oe ee ae ee Trace 0.40 0.92 1.36
(In eee ck. 2s ee eetarnee Trace 0.49 1.04 - 1.56
CO) emer a0 Sy. e hats Been Pan fid Gas Trace 0.68 0.91 1.44
72400 Saar liege ae ere Trace 0.36 0.60 1.27
ml Re ee rn event 2 5 a ciscaue he de ss Trace Trace 0.49 0.69

(K2SO4)

RR PN cS anes, ia bens, ¢ Bice as Trace Rr Al 1.44 125

20 Re i ea eee a ae aeaee Trace 0.70 1.31 1.97
OME RR Me Bie nue nate one Trace |. 0.67 1.65 Beet

11 NOTOO ESN: aye Sar eee Trace 0.89 1.86 PALE
PO ee Bee tS 3h Nera cc Slsavn acd eo Trace 0.81 2.18 2.16
33 O1GO)hs cee ee cr RAC ea nee Trace 0.63 1.87 2.08

It will be noted that in every case the potassium chloride decreased
the accumulation of nitrates, and that the depression increased regularly
with an increase in the amount of chloride applied except for one or
two minor exceptions. In the Volusia soil the degree of depression with
the heavy chloride treatments was less in the lime series than in the
no-lime series, indicating the tendency of the lime to overcome the harm-
ful effects of the potassium chloride.

Potassium sulfate seems to have exerted a stimulating effect on nitrate
accumulation. In the Dekalb soil the greatest degree of stimulation
occurs with the 1000-pound treatment, and then there is a gradual decline
with the two heavier treatments.

592 R. S. SMITH

TABLE 7. AccuMULATION OF NriTrRATES IN VoLusiA Str? LOAM AFTER SEVEN Montrus

(Moisture content, 30 per cent)

Milligrams of nitrogen
as nitrates in 100

Pounds of grams of soil
KCl to
Series 2.000000. |= =
pounds of | Aut Tater
ss up ex ae peaders
Pee months
ment
T 0 2.00 6.06
») ~
No lime; lime requirement 3393 pounds CaCO; Fi ae rie
to 2,000,000 pounds of soil 1,000 2 00 3°77
2,000 2.00 Meay
0 2.00 7.50
I 200) 2.00 7.14
tT 500 2.00 5.97
Lime requirement just satisfied 1,000 2 00 4 81
2,000 2.00 4.08
0 2.00 9.76
Ill 200 2.00 9.09
4000 pounds CaCO; to 2,000,000 pounds of soil in 500 a 8.69
excess of lime requirement 1,000 2.00 7.01
2,000 2.00 6.45

With the twenty-one-days incubation period (tables 8 to 10), all of
the soils in which the hme requirement was just satisfied show the
initial depression of nitrification with the 1000-pound treatment of
the chloride. When ammonium hydroxide was the nitrifiable material
added (table 10), altho the initial depression occurred at this point the
nitrates found in the heaviest chloride treatment exceeded those in the
check, indicating perhaps some action due to the basic nature of the
hydroxide.

In the Hagerstown soil treated with potassium sulfate (table 8),
nitrification was depressed slightly below that in the check with the
heaviest sulfate treatment. This was not the ease in the Dekalb soil
(table 9), altho in the latter soil the 3000-pound treatment caused less
stimulation of the process than did the 2000-pound treatment.

Some Errects or PorassituM SALTS ON SOILS 593

TABLE 8. Nrrrirrcation in Hacerstown Sitrt Loam WHEN AMMONIUM SULFATE Is
UsEpb

(Moisture content, 24 per cent; incubation period, 21 days)

Nitrogen as nitrates in 100 Nitrogen in (NH4)2SO,

grams of soil nitrified
Pounds of potassium salt Wi) eke eae
to 2,000,000 pounds 21.2 milli-
of soil Check grams N Milligrams | Per cent
(milligrams) adde H
(milligrams)
(xCl)

(ee ee 3.11 13.98 10.87 a ee
200) Aer enon ee eee 2.86 | 14.42 11.56 54.52
9100) a eae ener i era 3.05 14.93 11.88 56.03

TES CLOGS = Aenea ee ihe eee 2.94 12.95 10.01 47.21

74,1010 0 eat rae re eee 2.82 10.94 8.12 38 30

3) 01002 ene ae teenie 2.83 8.28 5.45 25.70
(X280s)

(Sock eek ae ree 3.87 17.18 13.31 62.31
UU en 5 eu eee ee 3.60 16.80 13:20 62.26
10 ORR ene cee Byo0 Liebz 14.17 66.83

i OO Berwretre waetentr se cis irs 4.06 19.08 14.97 70.61
DOO apie e oc.edure oc + pene 4.40 21.60 17.20 81.13
PS mCD Oe ir dhe, Seely ee ee 4.48 17.64 13.16 62.07

TABLE 9. Nirrirication 1N DexKats Sturt Loam WHEN AmMonriUM SuuFate Is Usep

(Moisture content, 24 per cent; incubation period, 21 days)

Nitrogen as nitrates in 100 Nitrogen in (NHy,):SOq4

grams of soil nitrified
Pounds of potassium salt
to 2,000,000 pounds | 21.2 milli-
of soil Check grams N Milligrams Per cent
(milligrams) adde
(milligrams)
(KC))

OS AR a 1.69 sa 1.76 8.30
PAN) Seatac dso i 1.36 3.49 2.138 10.04
DOOM rime a: oF. 2% bh 1.56 4.28 eT te 12.83

MUR sia 1.44 3.61 S17 10.23

24.4(0)0, 0 Renee ncaa Teer eZ! 0.94 4.43

(33;\010]0) Sa eae ene eee 0.67 IEPA 0.52 2.45
(K2SOx.)

[Oh Se Renee aes eo 3,62 23e i Lea ep
(Mire eeterece eo eee, te 1.97 3.58 1.61 7.59
EDM) ennai, ot mtaenstcrtiscs, « Bee 4.76 2.49 11.74

1180.00), Se ea ee a AT | ‘et 1) 2.84 13.39
PRUNE! gee sate cove chat: 2.16 5.74 3.58 16.88
23) 0, US eee oe 2.08 4.82 2.74 12.92

594 R. S. Smite

The beneficial action of lime is again brought out in table 10. Here
it is shown that in the no-lime series depression in the nitrification of
ammonium hydroxide accompanied the application of potassium chloride.
When the lime requirement of the soil was just satisfied, the initial
depression occurred with the 1000-pound treatment, and when lime was
used in excess of the lime requirement the initial depression occurred with
the 2000-pound treatment.

TABLE 10. Nuirriricatron IN Vouusta Sitt Loam Wuen Ammonium HyproxipE Is
UsEp

(Moisture content, 30 per cent; incubation period, 21 days)

Nitrogen as nitrates in) Nitrogen in NH,OH

100 grams of soil nitrified
Pounds a
of KCl to a
Series 2,000,000 21.2 milli-
pounds Check grams N Milli- Per
of soil (milli- added grams cent
grams) (milli-
grams)
I 0 6.65 8.19 1.94 4.73
No lime: lime requirement 3393 200 5.88 7.41 1.53 3.73
pounds of CaCO; to 2,000,000 500 4.88 6.25 1.37 3.34
pounds of soil 1,000 4.54 6.00 1.46 3.56
2,000 2.70 3.50 0.80 1.95
0 7.69 1h ao 3.42 8.34
Tl 200 5.58 10.96 5.38 13.12
= roe OF /

Lime requirement just satisfied 1,000 ee ia cn ie
2,000 3.33 7.41 4.08 9.95
ul 0 ATi IN 102 5.91 14.41
200 11.11 20.00 8.89 21.68
4000 pounds of CaCQ; to 500 9 20 2000 10.80 296.34
2,000,000 pounds of soil in 1.000 8.00 19.52 11.52 28.10
excess of lime requirement 2/000 7.41 16.02 8.61 21.00

The results from the use of dried blood as the nitrifiable material are
given in table 11. Here again the beneficial action of lime in counteracting
the ill effects of potassium chloride is shown very strongly. It is possible
that a longer incubation period would have allowed more nitrification
in Series I. An acid condition is apparently very unfavorable to the
nitrification of dried blood.

Some Errecrs or PorasstuM SALTS ON SOILS

505

TABLE 11. Nirrirrcatron tn Vouusra Siut Loam WHEN Drisep Bioop Is Usgep
(Moisture content, 30 per cent; incubation period, 14 days)
: fs eee
Nitrogen as nitrates in, Nitrogen in blood
Barnde 100 grams of soil nitrified
of KCl to
Series 2,006,000 21.2 milli-
pounds Check grams N oa
of soil | (milli- | added nell sh
grams) (milli- aoe oe
grams)
I 0 7.69 7.84 0.15 0.71
No lime; lime requirement 3393 200 7.16 Y o 0.00 0.00
pounds of CaCO; to 2,000,000 500 6.25 5.88 0:00 0.00
pounds of soil 1,000 4.54 4.66 0.12 (Oaays
2,000 aay 3.17 0.00 0.00
- 0 8.60 10.96 2 36 11.23
II 200 8.00 10.00 2.00 9.52
. ; . . 500 7.41 8.69 1.28 6.09
Lime requirement just satisfied 1,000 5 63 7 A7 1.84 8 76
2,000 | 4.60 5.06 0.46 2.19
- a 0 11.59 2154 9.95 47.38
200 ibe ba 21.87 10.76 51.28
4000 pounds of CaCO; to 500 9.63} 20.58) — 10.95 52.14
2,000,900 pounds of soil in 1,000 8.42 18.64 10.22 48 66
excess of lime requirement 2,000 Vote 16.08 8.32 | 39.51

Another series was run with Volusia soil, using ammonium hydroxide
and incubating for fourteen days. The resuits of this series are not
included herein, for they simply confirm the results of the twenty-one-
days incubation period.

In the foregoing discussion of the nitrifying power of the variously
treated soils, it has been assumed that the increase in nitrates during
the incubation period was due entirely to the oxidation of the added
materials. This assumption is clearly not entirely justified, and yet any
other method of obtaining the desired information would probably be
open to equally serious criticism.

Interchange of bases

The marked influence of potassium chloride on the nitrate bacteria
raised the question as to whether the toxic effect might be due to replaced

596 R. S. Smiru

bases. With this possibility in view, the water extracts of the variously
treated soils were tested for calcium, iron, aluminium, magnesium, and
manganese, and when found to be present each of these elements was
determined quantitatively.

No iron nor aluminium was found in any of the extracts, and no
manganese nor magnesium was found in certain of them, as appears in
tables 12 to 15. All determinations weré made in duplicate, and
checked very closely, so that only the averages are given.

TABLE 12. Amounts or Caucrum in VARtouSLY TREATED SoILs

Parts of calcium per million parts of soil extract

Pounds of potassium Hagerstown Dekalb Volusia

salt to 2,000,000

pounds of soil . : ; ; =
Lime Lime No lime; Lime Lime 4000

requirement | requirement | lime require- | requirement Ibs. in
just just ment 3393 just excess of
satisfied satisfied Ibs. CaCO; satisfied requirement
(KCI)

estas oeuee eee 21.6 2.9 Ton Bhat 33.6
730] ae a eee 30.5 5.7 15.9 26.7 |® 40.7
O00 peer 34.8 6.1 23.2 28.5 46.2

1000 2. as aes 44.2 6.5 26.5 44.8 54.3

7330) 00 eye a ae 60.3 11.6 38.4 51.4 53.7

5 QUO Ba cxiae anced se 2 TSO bee es ce |e ea ee ee
(KySO,)

Oreja ar PA DO13| Pave sechcn pet laaereeren se cl ca eee
740, 0 a Bren ae 24.1 ial Bek cere ocala ee |
S110 ny eee ee 25.2 Cl ee ee Coe Ars aro c:

1001010 eee ear eA 29.5 Beli | tars oh Be ci ee. utes shed 5a Sa | Oh ee
D500 0 ene eee ee 36.7 sx || etek cae 2 Nl) aol Gece sera | eee
S000 vee ce cee oe 45.1 GEO Parke cit a, Pe eae || La

As shown in table 12, with equal (but not equivalent) weights of the
chloride and the sulfate of potassium the chloride replaced more calcium

than did the sulfate.

of the two salts do not carry equal weights of the base.
As has been noted, Peters (1860) and Kiillenberg (1867) both found
that the base entered into the reaction independently of its form of com-

bination.

This result is to be expected, since equal weights

It does not follow from this, however, that equivalent weights

Some Errrects or Porassium SALts ON SOILS > 597

of the bases in a soil would appear in the extracts from the same soil
treated with various acids of the same base, because of the different
solubilities of the products of the reactions. Calcium sulfate is less
soluble than calcium chloride, and consequently less calcium would
probably be found in the extract of a soil treated with the sulfate than
in one treated with the chloride of potassium. This is apparently the
condition that existed in these soils, for there is less caleium present in
the extracts from the sulfate treatments than should be present theoretically
if the relative solubilities are discarded and only the replacing power of
the potassium actually added is considered.

No magnesium was found in any of the extracts from the Volusia
soil. This series is reported by Robinson (1914) to be low in magne-
sium. In the Hagerstown and Dekalb soils (table 13), less magnesium
than calcium was replaced by potassium. This result is in accord with
results from previous work.

TABLE 13. Amounts or MaGNEsIUM IN VARIOUSLY TREATED SOILS

Parts of magnesium per million
parts of soil extract

: Hagerstown Dekalb
Pounds of potassium salt to 2,000,000 pounds of soil
Lime Lime
requirement requirement
just just
satisfied satisfied
(Cl)
(0) a Gc een eg 15-6 2.9
00 = GE OE Se eee ee ae eee a ae 18.0 3.5
100) (8 Acer en ee Dee ee ee 19.4 3.7
NOM ND Meerereere EUR aN te! Qc a" hiayti fen «4 3.0.4 o> ace Saremecundemte eine 212, 3.7
DP. A000. se ie tegen a 24.8 5.0
PoP HO() Meneses sot at Pays. a: a sach dybvel ad afb e Haed & Sgeesdciue One 5.7
(KeSO4)

(OY cos ek re ae re ae a 12.8 aco
i) uni et MN De BS SY ore cad anne ak ye LAM es 15.0 Zao
Po Oe ncn se aeeace eye cette 2a Ree Saiee fy ype ye. cee es 15.8 2.5
UO Css cate 4 « ERROR ee ee ere eae ae 16.8 Die
BOAR (C0) eee ere Me Pos VARESE Pinta i te Ms cites wee ee 18.0 Pe
eh) ee eee er Sete Ae a tace eg Meas see 19.6 2.8

598 R. S. Suir

That appreciable amounts of manganese went into solution in the
Hagerstown and Dekalb soils is indicated in table 14. As previously
noted, manganese has been found to be strongly toxic both to plant growth
and to nitrification. Skinner and Reid (1916), as already stated, found

TABLE 14. Amounts of MANGANESE IN VARIOUSLY TREATED SOILs

Parts of manganese per million parts of soil extract
Pounds of potassium Hagerstown Dekalb Volusia
salt to 2,000,000
pounds of soil ; ; r-=2 | ; ,
Lime Lime No lime; Lime Lime 4000
requirement | requirement | lime require- | requirement Ibs. in
just just ment 3393 just excess of
satisfied satisfied Ibs. CaCO; satisfied requirement
(IxCl)
OF eee 0.24 0.78 0.00 0.00 ; 0.00
74.010 ena eee aes re ree 0.47 [eal 0.00 0.00 0.00
D00ger ee Ceres 0.57 1.92 Trace 0.00 0.00
ROOD Sea ee 0.71 3.12 0.30 0.00 0.00
UDO pee eee 1.15 4.17 0.65 Trace 0.00
000) 0 Saree eae 1.64 Ge 25e | 56 oes Sete Meee eee mn |e
(IxxSO,4)
Oe eee 0.98 OuDON | ante cree tosene 0 eae eae en | tne ee
2) ce Sees ete 0.99 OMDB segk epvtess von Poorer mene |e
DOOR eee ee 1.44 OSB), 2s 86k || eee an eee
(O00 fer eeeee 2.03 LOS |) dacs =e een | Meee eee ani ter ee eee
PeOQOO 2 vc see eam ae vA) Ved Sal 4s oa eos Se, een eee eee eae ee
SO O008 2s aac cca weds 2.80 2DO);|, Prova acetate: all gitarce irene | eRe ee
|

that manganese chloride was distinctly harmful to crop growth in an
acid soil when used at the rate of 50 pounds to the acre. This application
would be equivalent to about 14 parts of manganese to 1,000,000 parts
of soil if it is assumed that the salt became mixed with the surface soil
only. The Dekalb soil showed approximately this concentration of
manganese when its extract became toxic to wheat seedlings. To bring
this out more clearly, the parts per million of manganese in dry soil are
calculated in table 15.

The presence of manganese, however, cannot be considered as a com-
plete explanation for the toxic condition found in the extract cultures

Some Errects or Potasstum SALTS on SoILs 599

and indicated in the pot cultures, and for the depression of nitrification
particularly with the chloride treatments. The extract from the Volusia
soil was toxic to wheat seedlings in certain treatments and no manganese
was found in solution. Toxicity in solution cultures may arise from a

TABLE 15. Amount or MANGANESE IN Dry Sorn

Parts per million of water-soluble
manganese in dry soil

> . ; Hagerstown Dekalb
Pounds of potassium salt to 2,000,000 pounds of soil
Lime Lime
requirement requirement
just just
satisfied satisfied
(KCI)
SUMO Pete), te Wien Sie ops hy sFears Lair a ak 1.20 3.90
PAU IO) fia oS ee ae ee aT Le eee ae a 2.35 O00
iN) RE eee ee a eraeh ces) ies iid sis Gaye Min edee a aon ea 2.85 9.60
TL SLONOG) 5 <sc R ee n ar O ere S200 15.60
Sern HOR ea ur ean he cl a Yaan: «<0 k Eo yea wes Sah cRuene 5.75 20.85
SR Cece ee Tee ng 2 a ah @ einen AIA cei 8.20 BlLoe
(K2SOy)

OS 6 ea, Rn er are eee ee 0.93 2.50
20) Re ee ee ea Tm ei, 4.07! iin hh one 0.99 2.65
BUC). 3 hc Pe ae ae eee ee ar oo 1.44 4.65
TL, \OS010),. Se. 8 a ee ee a ee a 2.08 5.42
2D SOCIO) or Soo See oh, tan nD ay rer err 2.35 8.92
3100), Cen ee Oe eer 2.80 12.50

number of conditions, one of which is a lack of balance of nutrients. It
is of interest, nevertheless, tho perhaps not of significance, to note that
the soil having the highest content of water-soluble manganese showed
the weakest nitrifying power and the smallest accumulation of nitrates,
as well as the smallest growth of wheat in pot cultures and of wheat roots
in extract cultures.
SUMMARY

Three silt loam soils were used in the experiments reported herein,
each soil being representative of a large area in the United States. The
productivity of the soils ranged from high to very low.

600 R. S. SmitrH

The soils were screened and the pots were filled with the treated soils
as described. The official methods of analysis of the United States
Bureau of Chemistry were used, with the exceptions noted.

Potassium sulfate increased the yield of straw in Hagerstown soil and
showed no toxic effect in Dekalb soil. Potassium chloride apparently
became toxic to wheat in Hagerstown soil with the 1000-pound application;
in the Dekalb soil there was a slight decrease in yield with the 2000-pound
treatment.

In the extracts from the Hagerstown soil, potassium chloride stimulated
the root growth of wheat seedlings at all concentrations, the greatest
stimulation occurring with the 500-pound treatment. With the sulfate
there was a progressive stimulation to the 2000-pound treatment, and
a marked toxicity with the 3000-pound treatment. In the extracts from
the Dekalb soi! the checks were toxic to the root growth of wheat seedlings.
With the chloride the 200-pound treatment caused the greatest stimulation,
and there was a decrease in stimulation and apparent toxicity with the
heavier treatments. With the sulfate the 500-pound treatment caused
the greatest stimulation, and there was a decrease in stimulation and
apparent toxicity with the heavier treatments. In the extracts from
the no-lime series of the Volusia soil, toxicity to root growth became
evident with the 500-pound treatment. Lime overcame the toxicity even
with the heaviest chloride treatment.

Potassium chloride decreased the accumulation of nitrates in all cases.
Lime overcame this effect in part. Potassium sulfate apparently stimu-
lated the accumulation of nitrates in Hagerstown and Dekalb soils.

The heavier potassium chloride treatments depressed nitrification of
added materials. Potassium sulfate stimulated the process in all three
soils with the exception of the heaviest treatment with Hagerstown soil.
Lime had a tendency to correct the depression of the chloride in the
Volusia soil, but did not entirely overcome it.

No iron nor aluminium was found in any of the water extracts, and no
manganese was found in the extracts from the Volusia soil; hence the
harmful action of the potassium salts cannot be attributed to replaced
iron or aluminium, or to manganese in the case of Volusia soil. Both
the chloride and the sulfate of potassium replaced calcium strongly. Less
valcium appeared in the extract from the sulfate-treated series than
would be expected, possibly because of the relative insolubility of calctum

Some Errects or PorasstuM SALTS ON SoILs 601

sulfate. Magnesium was replaced less strongly than was calcium.
Manganese was replaced in very appreciable amounts in Hagerstown and
Dekalb soil, particularly in the latter. The soil highest in water-soluble
manganese showed the least nitrifying efficiency, the smallest growth
of wheat in pot cultures, and the poorest growth of wheat rootlets in
extract cultures.

The effects of potassium salts on plant growth are due to a complex
interaction of factors, involving perhaps the direct action of the salts
on plant growth and on bacterial activities, and also the action of bases
replaced by the potassium, particularly manganese.

602 R. S. Smita

LITERATURE CITED

AssotTt, J. B., Conner, S. D., AND SMatury, H. R. The reclamation
of an unproductive soil of the Kankakee marsh region. Soil acidity,
nitrification, and the toxicity of soluble salts of aluminum. Purdue
Univ. Agr. Exp. Sta. Bul. 170:327-374. 1913.

ALLEN, E. R., ANp Bonazz1, A. On nitrification. I. Preliminary obser-
vations. Ohio Agr. Exp. Sta. Technical bul. 7:1-42. 1915.

BeMMELEN, J. M. van. Das Absorptionsvermégen der Ackererde.
Landw. Vers. Stat. 21:135-191. 1878.

Die Absorption von Stoffen aus Lésungen. Ztschr. anorgan.
Chem. 23:321-372. 1900.

Bricas, Lyman J., AND LapHAM, Macy H. Capillary studies and filtra-
tion of clay from soil solutions. U. 8. Bur. Soils. Bul. 19:1-40.
1902.

Brown, P. E. Some bacteriological effects of liming. Iowa Agr. Exp.
Sta. Research bul. 2:47-107. 1911.

Brown, P. E., anp Minass, G. A. Effeet of some manganese salts on
ammonification and nitrification. Iowa Agr. Exp. Sta. Research bul.
35:1-22. 1916.

Dumont, J., AND CROCHETELLE, J. Sur la nitrification des terres de
prairie. Acad. Sci. [Paris]. Compt. rend. 117:670-673. 1893.

Influence des sels de potassium sur la nitrification. Acad. Sei.
[Paris]. Compt. rend. 118:604-606. 1894.

Frep, E. B., anp Grau, E. J. Some factors that influence nitrate
formation in acid soils. Soil sei. 1:317-338. 1916.

Greaves, J. E. The influence of salts on the bacterial activities of the
soil. Soil sci. 2:443-480. 1916.

Harris, Frank 8. Effect of alkali salts in soils on the germination and
growth of crops. Journ. agr. research 5:1-53. 1915.

Heappen, Wm. P. Yellow-berry in wheat — its cause and prevention.
Colorado Agr. Exp. Sta. Bul. 205:1-38. 1915.

Hinearp, E. W. Soils. Their formation, properties, composition, and
relations to climate and plant growth in the humid and arid regions,
p. 1-593. 1911.

SoME Errects or PorasstumM, SALTS ON SOILS 603

Hittespranp, W. F. Colorimetric determination of manganese. In
The analysis of silicate and carbonate rocks. U. S. Geol. Survey.
Bul. 422:116-118. 1910.

Jory, J. Some experiments on denudation by solution in fresh and
salt water. Rov. Irish Acad. Proc. 24:21-33. 1902-04.

Keutey, W. P. The influence of manganese on the growth of pineapples.
Hawaii Agr. Exp. Sta. Press bul. 23:1-14. 1908.

The function and distribution of manganese in plants and soils.
Hawaii Agr. Exp. Sta. Bul. 26:1-56. 1912.

KuLLenBeRG, O. Chemische und physische Eigenschaften des Bodens.
Jahresb. Fortschr. Agr. Chem. 8 (1865):15-63. 1867.

LemsBerec, J. Ueber einige Umwandlungen finlindischer Feldspathe.
Deut. geol. Gesell. Ztschr. 22:335-372. 1870 a.

Chemisch-geologische Untersuchung einiger Kalklager der finni-
schen Schareninsel Kimito. Deut. geol. Gesell. Ztschr. 22:803-840.
1870 b.

Ueber die Contactbildungen bei Predazzo. Deut. geol. Gesell.
Ztschr. 24:187-264. 1872.

Ueber Silicatumwandlungen. Deut. geol. Gesell. Ztschr. 28:519-
621. 1876.

Leoncint, G. Influenza di alcuni composti ossigenati di manganese
nella nitrificazione. Staz. Sper. Agr. Ital. 47:777-801. 1914.

Lyon, T. Lyrriteron, anp Brizzetu, James A. Relation of nitrogen
removal in drainage water to total flow of drainage. Jn Lysimeter
experiments. Cornell Univ. Agr. Exp. Sta. Memoir 12:37-40. 1918.

Lyon, T. Lytrurton, Fiepin, Etmer O., anno Buckman, Harry O.
Effects of potassium on plant growth. Jn Soils: their properties and
management, p. 551. 1915.

McCoot, M. M. The action of certain nutrient and non-nutrient bases
on plant growth. Cornell Univ. Agr. Exp. Sta. Memoir 2:113-214.
1913.

Mettor, J. W. Higher mathematics for students of chemistry and
physics, p. 1-641. 1909.

Montanari, C. Action of oligodynamic elements on the nitrifying
bacteria. Staz. Sper. Agr. Ital. 47:441-448. (Cited from Exp. sta.
rec. 33:422. 1916.) 1914.

604 R. S. SMITH

Ouaru, D. Action favorable du manganése sur la bactérie des légumi-
neuses. Acad. Sci. [Paris]. Compt. rend. 160:280-283. 1915.

Peck, 8. 8. Some biochemical investigations of Hawaiian soils with
) —
special reference to fertilizing with molasses. Hawaiian Sugar Planters’
Assoc. Exp. Sta., Div. Agr. and Chem. Bul. 34:32. 1911.

Peters, I. Studien tiber den Boden, aus dem Laboratorium zu Tharandad.
Ueber die Absorption von Kali durch Ackererde. Landw. Vers. Stat.
2:113-151. 1860.

Prcuarp, —. Action nitrifiante comparée de quelques sels contenus
naturellement ou ajoutés dans les terres végétales. Acad. Sci. [Paris].
Compt. rend. 98:1289-1290. 1884.

Renautt, P. Potash and the mobilization of organic nitrogen in humus
soils. Engrais 25°:132-135. (Cited from Exp. sta. rec. 22:714-715.
1910.) 1910.

Rosinson, W. O. The inorganic composition of some important Ameri-
ean soils. U.S. Dept. Agr. Bul. 122:1-27. 1914.

Ruprecat, R. W., anp Morss, F. W. The cause of the injurious effect
of sulfate of ammonia when used as a fertilizer. Massachusetts Agr.
Exp. Sta. Bul. 176:119-134. 1917.

SCHREINER, OSWALD, AND FAarLtyEeR, GEORGE H. Colorimetric, turbidity,
and titration methods used in soil investigations. U. S. Bur. Soils.
Bul. 31:1-60. 1906.

SKINNER, J. J.. AND Rem, F. R. The action of manganese under acid
and neutral soil conditions. U.S. Dept. Agr. Bul. 441:1-12. 1916.

SKINNER, J. J., AND Sutuivan, M. X. The action of manganese in soils.
U.S. Dept. Agr. Bul. 42:1-382. 1914.

SuLLivaN, Eucene C. The interaction between minerals and water
solutions. U.S. Geol. Survey. Bul. 312:1-64. 1907.

Tempe, J. C. Nitrification in acid or non-basic soils. Georgia Exp.
Sta. Bul, 103:1-15. 1914.

THompson, H. 8. On the absorbent power of soils. Roy. Agr. Soe.
Mngland. Journ. 11:68-74. 1850.

Truoc, E. The cause and nature of soil acidity with special regard to
colloids and adsorption. Journ. physical chem. 20:457-484. | (Refer-
ence on p. 462.) 1916.

Some Errects oF PoTAsstuM SALTS ON SOILS 605

VoELcKER, J. Aucustus. Hills’ experiments — the influence of lithium
and potassium salts on wheat. In The Woburn pot-culture experiments,
1908. Roy. Agr. Soc. England. Journ. 70:388-389. 1909.

Way, J. THomas. On the power of soils to absorb manure. Roy.
Agr. Soc. Engiand. Journ. 11:313-379. 1850.

On the power of soils to absorb manure. (Second paper.) Roy.
Agr. Soc. England. Journ. 13:123-1438. 1852.

On the influence of lime on the “ absorptive properties ’’ of soils.
Roy. Agr. Soc. England. Journ. 15:491-514. 1854.

Wuitsr, J. W. The results of long continued use of ammonium sulphate
upon a residual limestone soil of the Hagerstown series. Pennsylvania
State Coll. Agr. Exp. Sta. Ann. rept. 1912-13:55-104. (References
on p. 99, 102-104.) 1914.

_ Memoir 32, The Carbon Dioxide of the Soil Air, the third preceding number in this series of publica-
tions, was mailed on August 19, 1920.

iii