The influence of selected soil physical properties, soil type and site characteristics, soil temperature,
and soil moisture on the response of small grains to potassium on Montana soils
by Richard Harold Veeh
A thesis submitted in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE
in Soils
Montana State University
© Copyright by Richard Harold Veeh (1981)
Abstract:
Two hundred twenty-two small grain experiments established on 127 site locations throughout
Montana were selected to study the influence of pertain soil physical properties, soil classification
parameters, soil moisture and temperature, and site and soil profile characteristics on crop response to
applied K fertilizer. From 2 to 5 rates of K, ranging from 0 to 134 kg K/ha, were applied in the
experiments studied. Rates of N and P were held constant within an experiment but varied from one
experiment to another. A total of 48 independent variables in various combinations were inserted into
multiple linear stepwise regression programs. The dependent variables were average and maximum
percent yield response (to added K) as well as actual crop yield. The variable relationships were
analyzed over the whole data set as well as over part of the data file subdivided according to: 1) crop;
2) crop and geographic location; 3) crop, geographic location, and percent response class; and 4) K rate
at which maximum percent response occurred. Over 60 regression analyses were performed. Fifty-one
of the resultant regression equations produced significant R^2 values ranging from the .10 to .005
significance level. The variables entering the regression equations most often (and the number of times
each appeared) were elevation (19), latitude (17), dry consistence of the B horizon (17), dry consistence
of the Cca horizon (15), textural class (15), moisture regime (14), dry consistence of the Ap horizon
(14), and aspect (14). The most consistently correlated (Simple R) variables to the dependent variables
were the K-treatment (rate) variables, mean annual soil temperature, dry consistence of the B horizon,
elevation, slope, and moisture regime. These latter variables were consistently positively correlated to
percent crop response to applied' K. The results of this study indicate that crops grown on the soils
associated with the warmer and drier site locations responded to a greater degree to applied K.
However, low spring soil temperatures were also associated with greater crop response to added K.
Increased crop response to applied K was also associated with the more fine textured soils. Rainfall,
although not consistently correlated (Simple R) to crop response, had a marked influence on the effect
of applied K. STATEMENT OF PERMISSION TO COPY
In presenting this thesis in partial fulfillment of the require­
ments for an advanced degree at Montana State University, I agrfee
that the Library shall make it freely available for inspection.
I.
further agree that permission for extensive copying of this thesis
for scholarly purposes may be granted by my major professor, or, in
his absence, by the Director of Libraries.
It is understood that
any copying or publication of this thesis for financial gain shall
not be allowed without my written permission.
Signature
THE INFLUENCE OF SELECTED SOIL PHYSICAL PROPERTIES, SOIL TYPE AND
SITE CHARACTERISTICS, SOIL TEMPERATURE, AND SOIL MOISTURE ON
THE RESPONSE OF SMALL GRAINS TO POTASSIUM ON MONTANA SOILS
ty
.
RICHARD HAROLD VEEH
A thesis submitted in partial fulfillment
. of the requirements for the degree
of
MASTER OF SCIENCE
Soils
■I
Approved:
ittee
Head, Major Department
Graduate Dean
MONTANA STATE UNIVERSITY •
Bozeman, Montana
December, 1981
ill
ACKNOWLEDGEMENTS
The author wishes to express his sincere gratitude.to Dr. Ear3.
Skogley for his direction, inspiration,.and understanding during the
development and preparation of this thesis.
Appreciation is also extended to the thesis committee members.
Dr. Larry Munn and Dr. Cliff Montague.
Special thanks is extended
to Bernard E-. Schaff, upon whose M.S. thesis this thesis seeks to
build and without whose direct assistance this task could not have
been completed.
Also appreciation is extended to Dr. Gerald Nielsen I
for his aid in obtaining soil mapping and soil series description
information.
Others whose assistance deserves recognition include:
Dr.
Richard Lund and Les Dover for their help in the necessary statistical
analysis and computer programming, respectively; Montana Experiment
Station personnel, who contributed to the location and sampling of
the K-fertility plots; Soil Conservation Service personnel who pro­
vided soil series information for locations not previously mapped; the
cooperators who provided, their land for the fertility plots; Evelyn
Richard, typist.
Finally, the author wishes to express sincere appreciation to his
wife, Angela, for her understanding and support during this term of
study and thesis preparation.
TABLE OF CONTENTS
Chapter
Page
Vita................... .. . . ............... ; . . '
Acknowledgements............................ .
Table of Contents . . . .........................
.
■ List of Tables...................................
.
List of F i g u r e s .................................
.
Abstract. ................... . . ...................
ii
iii
iv
vi
viii
ix
1
INTRODUCTION..........
I
2
LITERATURE REVIEW-................. . . ........... "
Soil Texture and Clay Type..................... ..
Soil Structure, Consistence, and Bulk Density . .
Soil M o i s t u r e ...............................
12
Soil Temperature........................... * . .
5
5
8
3
4
MATERIALS AND METHODS .............................
Selection and Location of Plot Sites.......... ..
Fertility Plot Sampling .........................
Dry Consistence and Bulk Density Determination. .
Determination of Soil Series.........
.........
Variables Determined from the Soil Series
Description and Taxonomic Classification.. . .
Variables Determined from the Montana Agri­
cultural Experiment Station. AnnualReports.
. .
Determination of Latitude, Elevation and
Geographical Location ...............
Determination of Percent K Response ............
Statistical M e t h o d s .......................
RESULTS AND DISCUSSION.........................
37
Regression No. I - No. 5: Yield as the
Dependent Variable. . .........................
Regressions No. 6 - No. 10: Average Percent
Yield Response (V42) as the Dependent Variable.
Regressions No. 11 - No. 14: Maximum Percent
Yield Response (V50) as the DependentVariable.
Regressions No. 15 - No. 19: V50 or V42 as the
Dependent Variable...................... .
Regressions No. 20 - No. 31: V50 or V42 as the
Dependent Variable............................. .
Regressions No. 32 - No. 43: Average Percent
Response (V42) as the Dependent Variable. . . . . .
15
18
18
19
20
22
22
27
29
30
34
40
47 .
53
59
65
74
V
Chapter
Page
Regressions No. 44 - No. 47: Maximum Percent
Response (V50) as the Dependent Variable . .
Regression No. 48: Average Percent Response
(V42) as the Dependent Variable . . . . . . .
Regressions No. 49 or No. 50: Maximum (V50)
and Average (V42) Percent Response as the
Dependent Variable. . .
.............. ; .
Regression No. 51: MAST (V34) as the Dependent
Variable...........................
91
Regression Equations - Summary......... .
5
SUMMARY AND CONCLUSIONS . . . . .
...............
APPENDICES. ................................
Appendix 1 .....................................
Appendix I I .........................
Appendix III............................... . . .
Appendix IV. . . . . .
.................
LITERATURE
CITED.
..... .....................
83
87
89.
93
100
113
114
117
119
122
141
Vi
■ LIST OF TABLES
Table
Number
Page
1
Coding
2
Coding Scheme for Temperature Regime.
3
Scheme for Moisture
Regime . .............
.
.24
24
■ Coding Scheme for Structure ............... ..
25
4
Coding
Scheme for Textural
Class. .........
5
Coding
Scheme for Textural
Family . . . . . . . . .
6
Coding
Scheme for Crop Type . . . ...........
...
27
7
Variables used in Multiple Stepwise Linear
Regression Analysis .............................
.35
Regression No. I - No. 5; Actual Crop Yield (V06 V10) as the Dependent Variable; Data File not
Subdivided; Independent Variables not Restricted.
41
8
9
10
.11
12
.. . .
Regressions Nb. 6 - No. 10; Average Percent
Response (V42) as the Dependent Variable; Data
File Subdivided on Crop, Independent Variables
hot Restricted. ............. . . . . . . . . . .
26
,
26
.
48
Regressions No. 11 - No. 14; Maximum Percent
Response (V50) as the Dependent Variable; Data '
File Subdivided on Crop; Independent Variables
V01-V15 not Included.............................
54
Regression No. 15 - No. 19; Maximum (V50) and
Average (V42) Percent Response as the Dependent
Variables; Data File. Subdivided on Crop; VOl V15. and Soil Temperature, Rainfall, and Soil .
Moisture Variables Not Included as Independent
Variables
^ .
60
Regression No. 20 - No. 31; Maximum (V50) and
Average (V42) Percent Response as. the Dependent
Variables; Data File Subdivided on Crop and
Geographic Location; Independent Variables
Restricted as in Table 11 . . . . . . . . . . . . . . .
66
vii
Table
Number
13
14
15
16 ■
17
18
19
20
Page
Regression No. 32 - No. 43; Average Percent
Response (V42) as the Dependent Variable; Data
File Subdivided on Crop, Geographic Location,
and Percent Response Class; Independent
Variables as in Table 11. . . . . ...............
75
Regression No. 44 - No. 47; Maximum Percent
Response (V50) as the Dependent Variable; Data
File Subdivided on K-rate at Which Maximum
Response Occurred (V51); Independent Variables
as in Table 11....................................
84
Regression No. 48; Average Percent Response (V42)
as the Dependent Variable; Data File Restricted
to Include Only Cases With Complete Data; Inde­
pendent Variables Restricted to. Soil Temperature,
Rainfall, and Cumulative Soil Moisture Variables.
88
Regression No. 49 and No. 50; Maximum (V50) and
Average (V42) Percent Response as the Dependent
Variables; Data File Restricted to Include Only
Cases With Complete Data; Independent Variables
as in Table 11 and Table 15 Combined........ .. .
90
Regression No. 51; MAST (V34) as the. Dependent
Variable; Data File Subdivided on Crop; Inde­
pendent Variables Restricted as in Table 11 . . .
92
Overall Correlations of the Independent Variables
with the Dependent Variables - Yield (V06 - V10)
and Maximum (V50) and Average (V42) Percent Yield
Response. . ......................................
9.4
Summary of Small Grain Response to Applied K
Fertilizer, Montana Statewide Study, 1968-1980. .
97
Number of Experiments at the Various Rates of
Applied K at Which Maximum Percent K Response
Occurred..................... ....................
98
vi.ii
.LIST OF FIGURES.
Figure
Number
Page
1
Coding Scheme for Aspect.
............. .
„
2
Delineation of geographic location (V27) and the
number of experiments according to crop in,
each. W = winter wheat, S = spring wheat,
B = barley, 0 = oats. . . . . ............. .. .
20
.
31
ix
ABSTRACT ■
Two hundred twenty-two small grain experiments established on
127. site locations throughout Montana were selected to study the
influence of pertain soil physical propertiesi soil classification
parameters, soil moisture and temperature, and site and soil profile
characteristics on crop response to applied K .fertilizer. From 2 to
5 rates of K, ranging from 0 to 134 kg K /ha, were applied in the
experiments studied. Rates of N and P were held constant within
an experiment but varied from one experiment to another; A total of
48 independent variables in various combinations were inserted into
multiple linear stepwise regression programs. The dependent
variables were average and maximum percent yield response (to added
K) as well as actual crop yield. The variable relationships were
analyzed over the whole data set as well as over part of the data
file subdivided according to: I) crop; 2) crop and geographic
location; 3) crop, geographic location, and percent response class;
and 4) K rate at which maximum percent response occurred. Over 60
regression analyses were performed.
Fifty-one of the resultant
regression equations produced significant R^ values ranging from the
.10 to .00.5 significance level. The variables entering the regression
equations most often (and the number of times each appeared) were
elevation (19), latitude (17), dry consistence of the B horizon (17),
dry consistence of the Cca horizon (15), textural class (15),
moisture regime (14), dry consistence of the Ap horizon (14), and
aspect (14); The most consistently correlated (Simple R) variables
to the dependent variables were, the K-treatment (rate) variables, mean
annual soil temperature, dry consistence of the B horizon, elevation,
slope, and moisture regime. These latter variables were consistently,
positively correlated to percent crop response to applied' K. The
results of this study indicate that crops grown on the soils associated
.with the warmer and drier site locations responded to a greater
degree to applied K. However, low spring soil temperatures were
also associated with greater crop response .to added K. Increased
crop response to applied K was also associated with the more fine
textured soils. Rainfall, although not consistently correlated
(Simple R) to crop response, had a marked influence on the effect
of applied K.
Chapter I
INTRODUCTION
Accurately predicting K-fertilizer recommendations for crops
grown on Montana soils has been a recognized problem for several years.
Research during, the past decade has shown K-response by various crops
to be unpredictable, and poorly related to soil test K ratings.
Although a soil may test "high" in extractable K, a yield response to
added K frequently occurs.
Significant crop response also occurs
regardless of yield level; this suggests that K may be yield-limiting
independent of the status of other growing conditions (Skogley, 1976).
A coordinated statewide research effort was begun in 1971 to try
to determine what factors were contributing to the variability in
K-response.
That K-fixation had long been associated with the
presence of micaceous clays (particularly illite) led to a study in
which clay type and amount of clay were related to crop response
(Phillips, 1973).
No significant relationships existed upon which a
K-soil test could be based.
Wang (1975) also studied K-fixation
and release potential of selected soils incubated over various time
periods.
This research showed that after small additions of K, the
level of extractable K could be greatly increased; this suggests that
a type of K-release mechanism is operative in some soils.
Haby (1975) investigated the possibility of using a more
reliable extraction procedure than the standard NHyOAc extractable
.2
'K-soil, test;
Fifteen different extraction techniques were studied
on a wide range of soil types.
The results of this investigation
showed that soil test values could at best, only account for 40 percent
of the variation in crop response.
This strongly suggests the lack
of a close functional relationship between extractable K (regardless
of extraction procedure or method of expressing the results) and K
availability to plants grown.under the climatic and soils conditions
of Montana.
Some system which relates more closely to factors
controlling K availability over time will need to be developed as a
K soil test.
Uptake of K by plant roots has been shown to be governed
primarily by ion diffusion, accounting for a minimum of 80 percent of
the K-supply.
Mass flow and direct root contact can account for no
more than 15r20 percent of the K supply (Barber, 1962; Beckett, 1964).
Massee (1973) found a very good correlation between K-diffusibn in .
a variety of soil types and crop response to added K under controlled
conditions.
Developing a K soil test based on factors that influence K
diffusion presents a complex problem.
Not only is K diffusion related
to numerous soil physical properties (e.g.- texture and structure) but
also to various climatic and weather-related factors such as moisture
availability and temperature.
These are extremely variable from year
to year,in Montana, as well as during any one growing season.
Shrader, et al (1957) hypothesized that soils should possess
3
inherent characteristics which influence crop growth and are defined
and/or expressed by the parameters used in soil classification.
Schaff (1979) investigated this hypothesis on a number of soils which,
had been used during two years of soil fertility field research
experiments with winter wheat.
The soils were characterized as to
their p h y s i c a l c h e m i c a l , and climatic properties and then correlated
to winter wheat response to added K-fertilizer.
Variables with the
highest correlation (soil and site combined regression analysis) with
percent yeild response were:
I) mean annual soil temperature at 50
cm.; 2) moist consistence of the Ap horizon; 3) moist consistence of
the B horizon; 4) dry consistence of the B horizon; and 5) the clay
/
content of the Ca horizon.
•
.
:
2
These five variables produced an R
value
of .88.
These results were very encouraging.
Td further investigate
the hypothesis that genetic soil characteristics influence crop
response to added K 9 the present study was conceived.
The data base
would be expanded to determine if significant correlations could be
observed statewide, and for additional crops.
Soils and sites were
classified according to Soil Taxonomy (U.S.D.A., 1975) at selected
locations at previous small grain soil fertility experiments (127
sites).
The soil samples were analyzed by horizon (Ap9 B 9 and Cca)
to determine dry consistence and bulk density values; other selected
soil and site properties which could be determined from the SCS
4
soil series description for a particular, site and for which a coding
scheme could be devised were also included.
The objectives of this research project were to I) determine
if known or easily determined soil and site characteristics were
significantly correlated to crop response to K fertilizer when a
large number of sites from,past experiments throughout the state was
included, arid 2) investigate the^feasibility of developing a system
for predicting future crop responses to K-fertilizer based on
selected soil and site classification parameters.
Chapter 2
LITERATURE REVIEW
-
Crop responses to applied fertilizers and yield differences
have been associated with soil type differences (Shrader et al., 1957;
Olsen, 1977).
Dass and Shankhayan (1979) reported that the dry
matter production of wheat and response to added K differed signifi­
cantly when the crop was grown on different soil types.
Some of the
soil and site properties which have been reported to be most important
in this regard are discussed below.
Soil Texture and Clay Type
Liebhardt and Cotnoir (1979) studied 28 Delaware soil series
ranging in texture from silt loam to loamy sand..
Soils higher in sand
content did not need as much K added to raise the soil test values as
soils higher in silt and clay.
Von Braunschweig (1980) found similar
effects of clay content upon K-availability.
On coarse textured soils
with clay contents up to 12 percent, a K-saturation of the clay
minerals, from 1.7 to 2.5 mg. K/percent clay was necessary for adequate
plant uptake of K.
Soils, with clay contents from 12 to 25 percent and
greater than 25 percent required 1.2 - 1.7 mg K/percent clay and 1.11.4 mg K/percent clay, respectively, for adequate K uptake -
Calculating
these values in terms of mg K/100 g soil, it was apparent that the
more clay the soil contained, the higher the exchangeable K content
6
should be for adequate K nutrition.
The level of K available at the plant root is the critical
factor governing plant K uptake.
aspects.
.
This involves a number of soil
Potassium-fixing.soils may show extreme, variability in their
K-supplying power.
In one study it was observed that K-fixation. values
were relatively constant on the plots without or with low K-fertilization; after high K application, fixation values fluctuated much more
(Burkart and Amberger, 1978).
Also, the rise in the level of
available K by fertilization was negatively correlated to soil clay
content.
In a similar study, the K concentration of a saturated
paste extract was inversely related to soil clay content (Mengel and
Aksoy, 1971). ■ These experimental data support the concept that the K
concentration of. the soil solution influences the K-supply. for plants
and that the K concentration of the soil solution is influenced by the
degree of K saturation of the clay minerals.
In a study of four Ohio.soils with a wide range of K release
capability, it was observed that the two soils highest in clay and
total K content reacted in a significantly different manner to added
K than did soils with less clay (Munn and McLean, 1975).
Initial
cropping decreased exchangeable K in all K-treated soils eliminating
the effect.of K treatments on exchangeable K.
However, after initial
cropping where no K was added, exchangeable K varied sixfold from.the
lowest (least, percent clay) to the highest .(greatest percent clay).
7
In the two soils highest in clay content, there was a tendency for
prior cropping to increase K fixation and reduce the plant recovery
of subsequently applied K.
Sparks et a l . (1980) reported that soils with sandy surface
horizons and clayey subsoils had a pronounced accumulation of K
in the subsoil layers.
The K in the clayey subsoil, probably a result
of leaching of both applied K and that of genetic origin, was avail­
able to plants depending upon the ease of root, penetration.
The
ability, of the plant to extract subsoil K caused a lack of response
to surface applied K.
Singh et al. (1977) observed higher K uptake by dryland wheat
grown on a clay loam soil than oh a loamy sand, soil in a year of normal
rainfall and profile storage.
This difference was attributed to the
higher water storage capacity of the clay loam soil.
It is probable
that the increased water supply in the clay loam soil is related to a
greater potential, for K ion diffusion.
Particle size within the clay fraction has been observed to
influence the rate of weathering and K fixation and release of clay
minerals.
Potassium release has been shown to be more rapid as the
particle size becomes smaller.
However, at a certain point in
decomposition at which the diameter and thickness of the clay particles
approach
equality, little or no K release is observed.
This is
probably due to the stable overall charge of the clay particle
8
(Sawhney, 1972).
Particle size was also related to K-selectivity and
to ease of collapse of the frayed edges of the particle.
A greater
area of collapsed central core relative to the edges of the particle
was associated with greater ease of collapse of the edges.
Beckett
and Nafady (1967) also associated specific K sites with the edges or
peripheral interstices of stacks of clay plates and the non-specific
(Gapon) sites to their planar surfaces.
Soil Structure, Consistence, and Bulk Density
Soil structure can influence the pgrceptage of total soil volume
that can be utilized by roots.
Soil consistence as well as perme­
ability can be interrelated with many other physical properties.
A
hard dry consistence commonly implies slow permeability, low porosity,
and high bulk density (Niekerk and Lambrechts, 1977).
In a study of
some Singapore soils. Wells and Leamy (1977) observed that the
physical properties of the soils were different and could be related
to the nature of the parent rock.
For intensive market gardening, the
moist soil consistence and grade of structure were both determined to
be of major importance.
Soil consistence is a soil parameter which measures the force .
required to crush a soil ped.
Consistence is probably an important
parameter in that it is dependent upon various soil physical properties
such as porosity, bulk density, arid texture as they interrelate.
Cone
9
index (Cl) ] defined as the force required to push a penetrometer into
the soil divided by the cross-sectional area of the penetrometer cone,
is a similar type of measurement.
In a study fo determine those
factors which influence the Cl value, it was observed that Cl for
each texture increased monotonically with decreasing soil water
pressure, but no simple relationship between Cl and texture was
found (Byrd and Cassel, 1980).
Regression equations relating Cl
to water content, percentage of sand, and the volume of pores greater
than 150 pm in diameter explained 67 - 72 percent of the observed
variation (.0001 probability level.).
It was pointed out that
the importance of roots in natural systems should not be overlooked
as an influence, in determining soil physical properties.
Reddy et al. (1978) observed that the hydraulic conductivity
and water holding capacity of the soil had increased by 25 - 30
percent and 10 - 16 percent, respectively, under all of the cropped
plots (42 ikg. K/ha level) at the full flowering stage as compared to
the control.
The root weight of all crops had also increased by
25-40 percent with the increased K application.
Root growth has been shown to be one.of the most important factors
for improving soil structure; bulk density can be used as a sensitive
index of soil structure.
This suggests that the higher rates of
hydraulic conductivity (associated with the lower bulk density values)
10
observed under the 42 and 83 kg. K/ha treatments might be attributed to
better root growth.
The relationship between root growth, bulk density and soil
structure has been shown to be important.
Schuurman (1971) observed
that root weights, numbers, and rooting depth decreased with increasing
soil density.
In addition, uptake of water and minerals, particularly
P and K, decreased.
The yield of dry matter of wheat decreased in all
soil series except one as the bulk density was increased from 1.02
3
to 1.52 g/cm
(Sharma and Verma, 1971).
A similar trend was also
observed for root growth, grain yield, and uptake of N, P, and K.
In a related study, the compaction of soil to create a bulk density
3
greater than 1.40 g/cm
proved harmful for plant growth in all soil
series observed (Verma and Sharma, 1972).
The harmful effect of
compaction was again directly related to decreased nutrient uptake.
It is probable that decreased nutrient uptake is related to decreased
ion diffusion and increased soil matric potential associated with the
higher bulk densities.
Soil physical parameters may also be important in the way that
they can modify the biological-chemical nature of the soil environment.
Samra and Goswami (1978) found that grain yield increased up to a
bulk density of 1.6 and 1.75 at the 0 - 1 5
cm depth and 15 - 30 cm
depth, respectively, after which yield dropped off.
The interaction
of bulk density and moisture content was observed to significantly
11
affect the oxygen diffusion rate.
Oxygen diffusion rate increased as
yield increased to a point after which yield decreased.
The same kind
of parabolic relationship was observed between bulk density and grain
yield and between soil resistance to penetration and grain yield.
In another study soil compaction was observed to decrease total
porosity and the amount of macropores (> 50 n).
However, the amount
of water holding pores (.20 -.10 y) and fine capillary pores (< .20 p)
increased (Talha et a l ., 1979).
Total dry matter production of maize
and barley as well as. K concentration and uptake was higher at all
levels of fertilization in the compact soil and increased as K
fertilization level increased.
By compacting the soil an equilibrium
was reached among water and air-filled pores.
The potential existed
for a negative effect on plant.growth during moist years because the
plants would suffer from an excessive COg/Og ratio
(0^ deficiency).
Tillage practices have been found to significantly affect K
' concentration and uptake of K.
These effects have variously been
associated with increased compaction of high moisture content (Fisher
et al., 1975),. differences in concentration gradients of K (Drew and
Saker, 1978; Hodgson, et al., 1977), differences in moisture content
and depth and abundance of rooting (Cannell et al., 1980; Drew and
Saker, 1980), and differences in physical and chemical properties
and activity of earthworms (Lai, 1976),
12
Soil Moisture
Sharda and Gupta (1975) point out that soil moisture affects
the growth of plants by modifying soil aeration, mechanical impedance
of the soil, the concentration of readily soluble nutrients and the
heat conductivity of the soil.
In their investigation an increase
in oxygen diffusion rate with increasing soil moisture tension was
indicated.
Uptake of nutrients was found to decrease with increased
moisture stress.
The interaction between soil moisture regime and nitrogen
level has been found to be significant in respect to N, P, and K
uptake (Varma et a l ., 1976; Varma, 1976).
As soil moisture increased,
the higher rates of nitrogen increased general growth and dry matter
production.
Better root growth, in particular, allowed the plants
access to more soil which resulted in better water use efficiency.
■.
'
The influence of increasing moisture levels on N, P , and K uptake
has been similarly observed in two other studies (Bajpai and Mertia,
1977; Fedak and Mack, 1977).
Increasing moisture levels increased
ash content of the grain and straw, increased grain and protein yield,
and decreased grain protein content.
Hoyt and Rice (1977) found that efficiency of moisture use was
generally more than doubled by the fertilizer and fertilizer plus
manure treatments, versus the control (no applied fertilizer or manure)
Petinov et al. (1977) also observed that the higher the level of N, P,
13
and K, the more pronounced the positive response of the plants to soil
moisture.
Soil moisture, as it fluctuates, may also affect exchange
relations and, therefore, soil solution concentrations of. K.
Raney
and Hoover. (1946) reported that alternate wetting and drying of some
soils caused a rapid fixation of K in a non-replaceable form and that
very little fixation of this kind took place under the soils kept
continually moist.
However, Scott and Smith (1957) found that the
exchangeable K in the surface and subsoil was doubled upon drying
and that K uptake by plants was always less on soils kept continually
moist than from soils that had been dried and rewetted.
uptake, in particular, from the non-exchangeable
Potassium
K fraction, was
impaired by a dry soil medium (Mengel and Wiechens, 1979).
The release
rate of non-exchangeable K was more important for crop production
than the level of exchangeable soil K.
Beckett and Nafady (1967) also
observed that the rate of release of, K from a non-labile pool decreased
as the non-labile pool became exhausted.
Soil moisture will influence the concentration of ions in the
soil solution both by the effect of dilution and by the relative
effect on ions of different charge.
Karlen et al. (1978) found that
as soil moisture was increased, Ca and Mg concentrations in leaf tissues
Were depressed while K concentrations increased or remained unchanged.
It was felt that this differential change in cation composition of
14
wheat grown under wet soil conditions could be explained by changes
in ion activities in accord with Donnan
Equilibrium Theory. .This
theory shows that t h e 'availability of monovalent ions increases while
the availability of divalent ions decreases as soils become more
nearly saturated with water.
There is much evidence that K concentrations in soil solutions
are decreased and that K absorption on exchange sites is increased
by. liming.
These Ca:K interactions explained
the decreased yield
of potatoes in a study conducted at Rothamsted (Bolton, 1977).
Jankovic
and Nemeth (1978) observed that owing to K and P fertilization, the Ca
concentration of the soil solution increased as a result of exchange
processes in the soil.
Simple ion concentrations were determined
to be more suitable than the potentials for defining nutrient dynamics
changed by fertilizer application.
The concentration of cations in solution and pH have also been
shown to influence K release from micaceous clays.
Marked complementary
ion effects on K and Ca displacement occurred when the exchange was
from pH-dependent charges.
At 30 percent K saturation (70 percent Ca),
complementary ion effects were small.
However, at 10 percent K
saturation (90 percent Ca), the different complementary cations caused
more than a fourfold difference in the K displaced from illite ■
(McLean and Bittencourt, 1974).
Mattson (1973) found that the uptake .
of P and K was lowest, and the.uptake of Ga highest toward the dry end
15
of the treated soils.
Again, the implication here is that moisture
is important in that it modifies both the relative ionic concentrations
and ion diffusion.
Soil Temperature
At any one moment, temperature varies from soil horizon to soil
horizon.
It fluctuates with the hour of the day and with the season
of the year, and the fluctuations may be large or small according to
the environment.
Seasonal fluctuations in soil temperature are
affected by latitude, soil moisture, groundwater, air movement near
.
■
i.
the ground, clouds, rain, and ground cover.
The influence of latitude
is dominant over most of the United States.
Daily fluctuations are
affected by all of these except latitude, arid the influence of moisture
is dominant (Smith and Newhall, 1964).
The importance of slope and ■■
aspect can also be very pronounced on adjacent soils, especially in ..
the higher latitudes.
Daily changes in air temperature have a significant effect on the
temperature of surface-soil horizons to a depth of about 50 cm.
This
is particularly so in soils of dry climates where moisture can be
exceedingly important in reducing fluctuations in soil temperature.
Oliveira et al. (1979), in attempting to estimate soil temperature at
2 cm from air temperatures, found that on rainy days the measured
soil temperature was always lower than the estimated soil
temperature
because the rainwater cooled the soil surface more rapidly than the air
16
Willis and Power (1975) point out that water viscosity and sur­
face tension are inversely related to temperature, and relative
hydraulic conductivity increases as temperature increases.
A dry
soil will freeze more quickly and to a greater depth than a wet soil,
and thawing of the dry soil occurs much more rapidly in the spring.
As the soil profile cools or warms, the water table drops or
rises in response to the known fact that a cold soil holds more water
than a warm soil and loses water more slowly.
Frost may remain in
the soil profile for a significant time after the surface has thawed,
thus forcing plant roots to grow into soil colder than the surface.
Soil temperature has been shown to significantly affect K-uptake
and the K-requirement.
Boatwright et al. (1976) found that fertiliza­
tion of spring wheat with N-, P , and K appeared to partially alleviate
the detrimental effect of low surface soil temperatures.
In. a
similar study, N, P, and K concentrations were observed to be higher
in grain.from plants grown under reduced soil aeration and 25° C soil
temperature than in grain produced at lower soil temperatures
(Labanauskas et al., 1975).
Wicke (1973) also observed that the K requirement was greatest
at the lower soil temperature and that the response to added K was
smaller at high root-zone temperatures than at low root-zone
temperatures.
Kabli and Toop (1970) found that high soil temperature
(23.9° C and higher) could be contributing to the problem of K-induced
17
Mg deficiency by increasing the potassium uptake.
Mack (1971) confirms this in his report op yields of bromegfass..
An increase in concentration of the major nutrients (N, P, K) in the
plants coincided with the. greater herbage growth on the warm soil.
The changes in uptake for N,. P, and K per I 0 C change of seasonal
temperature were 8.7, 10.4, and 7.1 percent, respectively, and
the associated
values were 1.5, 1.6, and 1.4 at 9.2° C.
• Temperature influences may also be related to different exchange
relations which exist among potassium-bearing micas.
The heating of
various micas caused marked differences in K-exchangeability. Although
the maximum degree of K exchange was generally unaltered by heating,
major changes in the rate of exchange occurred (Scott, et a l ., 1972).
Heat treated muscovite showed a marked increase in the rate of
exchange, whereas biotite and lepidomelane showed a decrease in the
rate of K release.
Chapter 3
MATERIALS AND METHODS
Selection and Location of Pldt Sites
Between 1968 and 1980, numerous field soil fertility research
studies were conducted by Montana Agricultural Experiment Station
researchers, Extension Service Soil Scientists, and USDA-ARS
scientists in Montana.
.
Results of recent research.suggested that
these sites could provide valuable information for developing a system
for predicting crop response to K fertilizer.
Because crop yield
and response data were already available from previously conducted
research, only the site and soil characteristics (which are basically
constant over the years) needed to be obtained.
To do this, only
-
those research sites, which included small grains (winter wheat, spring
wheat, barley, and oats) as the experimental crop and upon which
various rates of K fertilizer were applied as a variable in the
experiment were selected.
The total number of sites sampled based
on these criteria was 127.
Because many of the sites had several
experimental plots established on them oyer the years, the site data
\
•
1
'
could be applied to more than one set of experimental data (i.e. yield
data, etc.).
For this reason, the total number of experiments (cases)
included in this study was 222,
In most cases site location was determined from the legal
descriptions as reported by the various researchers in the Montana
19
Agricultural Experiment Station Annual Report.
Actual location in
the field of the old plot sites was accomplished with the aid of
someone who was directly involved in the research when, the plot was
established and/or with the, help, of the cooperatofs themselves.
In
all cases, care was taken to insure that the sample taken was in close,
proximity to the old fertility plot site to insure the same soil
type (series, level).
See Appendix I for site numbers, cooperators,
legal descriptions, and soil series names;
Fertility Plot Sampling
Soil samples were taken as near the center of the old fertility
plot as could be determined.
A Giddings probe,was used to take soil
samples to a depth of approximately four feet.
The core sample was
then divided into plow layer (Ap) horizon, B horizon, based on
structural and textural differences induced by clay accumulation,
and a Cca horizon, as determined by reaction with dilute hydrochloric
acid.
Samples from each horizon at each site were then placed in a
sampling bag labelled as to cooperator, years in which the plot was.,
used, and horizon name and depth.
In several cases, because of the method of sampling, soil type
differences, or previous erosion, it was not possible to distinguish
three distinct horizohs.
In a few cases, the presence of a Cca
horizon was not detected to the maximum depth of sampling.
20
Percent slope and aspect of the plot area were also recorded.
Aspect was introduced into the analysis by the following coding
scheme in which a number from I to 8 was assigned to correspond with
the general site aspect (Figure I).
315-360
45-90
270-315
225-270
180-225
\
90-135
Coding Scheme
for Aspect
135-180
Dry Consistence and Bulk Density Determination
Traditionally, dry consistence in the field is determined by
placing a soil ped between the thumb and forefinger (and/or between
the hands) and exerting a force sufficient to crush the ped.
On the
basis of the force applied, one of six semi-quantitative values is
assigned, ranging from loose to extremely hard.
21
To obtain a more quantitative .(less subjective) measure of dry
consistence, a pocket penetrometer (CL-700; Soiltest1 Inc.) with
the foot attached was used to determine the point of failure or force
required to.crush the ped.
Five peds from each horizon at each site
‘
2
were sampled and the dry consistence measurements recorded in kg/cm .
In addition to the latter measurements, the thickness of each ped
was measured and recorded.
This was done to subsequently determine
if a significant relationship.existed between ped thickness and the
consistence measurement, itself.
If necessary, corrections could then
be made later for those horizons with either small or large average
ped thicknesses.
However, care was taken to try to eliminate this
potential problem by selecting five peds of similar range in size for
each horizon at each site.
Ped thicknesses ranged from 10 - 25 mm;
2
If the force required to crush the ped exceeded 4.5 kg/cm
(maximum
value on the penetrometer scale), a value of 5.0 was recorded. .
Bulk density determination was accomplished by using the clod
method (Black., 1965) .
medium.
Saran was used as the water seal or coating
This method was chosen because in many cases an intact core
was not preserved through handling.
Three large-sized peds were
sampled from each horizon for each site.
The average bulk density
of the three peds sampled was entered as the value used in the
statistical analysis. ,
22
Determination of Soil,Series
The soil series present at a site was determined in one of two"
ways.
The soil series name was obtained in many cases from the
researcher as reported in the Montana Agricultural Experiment Station
Annual Report.
If not given, then the legal description of the
fertility plot was used in conjunction with the appropriate Soil
Conservation Service county soil survey to determine the soil.series
present at a site.
Not all counties in Montana have been mapped,
so the soil series present at some site's is not known.
The inclusion
■
of the variables derived from the soil series description was not
possible for these sites (See Appendix II). "
Variables Determined from the Soil Series Description and Taxonomic
Classification
'' '
The soil classification and profile description system (H.S.D.A.,
.
.
■
,
.
,
1951; U.S.D.A., 1975) used in the United States makes use of soil
physical criteria as well as climatic parameters to distinguish one
soil type from another. .In the present study, some of the variables
used in the statistical analysis were derived from these physical
and climatic criteria.
These include:
I).mean annual soil tempera­
ture (MAST); 2) soil structure; 3) textural class; 4) textural family;
5) temperature regime; and 6) moisture regime.
Mean annual soil temperature was introduced as a variable in the
,
23
analysis by calculating the average of the range of values presented
in the soil series description for a given site.
If a specific tempera
ture was obtainable as the MAST in the soil series description, then
that value was used in the analysis.
Soil structure, textural class, textural family, temperature
regime, and moisture regime were all introduced as variables in the
regression analysis according to various coding schemes.
The coding
schemes were devised so that the number sequence would reflect a
logical order.
For example, textural family determination was coded
from coarse-textured to fine-textured; structural grade was coded
from weak to strong.
Moisture regime determination for a given soil series was
derived from the classification of that series at the subgroup level.
A soil with a cryic temperature regime was assumed to possess a
udic moisture regime in that the ustic moisture regime could not
technically apply at the subgroup level of classification (U.S.D.A.,
1975).
A soil with a frigid temperature was considered to have
an ustic moisture regime unless the subgroup modifier indicated
some other moisture regime (intergrade).
Any soil series in this
study classified as an aridisol always possessed an ustic intergrade
in the subgroup modifier.
See Table I and Table 2, respectively, for
moisture regime and temperature regime coding schemes.
Soil structure was introduced as a three-digit variable in the
24
Table I
Coding Scheme for Moisture Regime
Moisture Regime
Coded Value
Udic
I
2
. Ustic
Ustic - aridic
Aridic - ustic
3
.
4
.
Table 2
Coding Scheme for Temperature Regime
Temperature Regime
Coded Value
Cryic
I
Frigid
2
Mesic
-
.3
'' '
'
.25-
analysis-composed of three coded 1-digit numbers for grade, size,,
and type, respectively (Table 3).
.
Table 3
Coding Scheme for Structure
Coded
Value
Grade
Size
Coded
Value
Coded
Value
Type
Weak
I ''
Fine
I .
Platy
I
Moderate
2
Medium
2
Prismatic
2
Strong
3 .
Coarse
3
Columnar
3
Angular
blocky
4
Sub-angular
blocky
5
Granular
6
Massive
7
Single grain
8
'-
■
;
Textural class always refers to that texture dominant at the sur­
face of a given soil series and is that texture usually associated with
the soil series name. Textural family is that parameter associated
with the control section of a particular soil series at the family
level of classification (Hopkins, 1979; U.-S.D.A., 1975).
Table 4 and
Table 5 contain the coding schemes for textural class and textural
family, respectively.
Soil series names and their respective moisture
26
Table 4
Coding Scheme for Textural Class
'
Textural Class.
Coded Value
Fine-sandy loam
I
Loam
2
Silt loam
3
Silty clay loam
4
Clay loam
5
Clay
6
Table 5
Coding Scheme for Textural Family
Textural Family
Coded Value
Coarse-loamy
I
Fine-loamy
2
Fine-silty
3
Fine-montmofillonitic .
4
-
27
and temperature regimes, as well as textural classifications can
be found in Appendix II.
Variables Determined, from the Montana Agricultural Experiment Station
Annual Reports
Variables obtained from the annual reports that were used in the
statistical analysis included:
I) type of crop; 3) year of the
experiment; 3) yield data; 4) K-rates applied; 5) rainfall; 6) soil ■
temperature data; and .7) soil moisture data.
The coding scheme for crop type is given in Table 6.
Table 6
Coding Scheme for Crop Type
Crop Type
Coded V a l u e .
Winter Wheat
I
Spring Wheat
2
Barley
3
Oats
4
"Year" was introduced as a variable in the analysis by recording
the last two digits of the year the experiment was conducted.
Yield
data ('kg/ha) were obtained from the annual reports only for those
28
treatments in which nitrogen and phosphorus were considered to be
adequate and in which the N and P rates were constant across the
various rates of K,
Thus, N and P rates were constant within an
experiment for the yield data recorded but were variable from one
experiment to another.
Rates of applied K varied from .experiment to experiment.
In some
cases, only two rates of K were applied (e.g. 0 and 56 kg K/ha).
a number of cases, five rates of K were applied.
In
In all cases, the
number of yield variables corresponded to the number of rate variables.
The control was always the same at 0 kg K/ha.
Rainfall (cm.) was introduced as a single variable in the analysis.
The value recorded, was that of total rainfall reported during the
growing season with no regard being given to the distribution.
Temperature data taken throughout the growing season were used
in calculating an average monthly temperature for April, May, June,
and July.
Four variables were thus introduced into the analysis.
Spring soil moisture was introduced to the analysis in the form
of seven variables.
content to 183 cm.
These variables represented the total moisture(V43) as well as the moisture content of the
soil profile taken in 30 cm. increments from the surface to 183 cm.
(6 variables - V44 through V49).
At a later date, sixe more variables
were introduced from these data as a modification of the six variables
just mentioned.
Whereas the first six variables (V44 through V49)
29
measured soil moisture within a given 30 cm. increment, the latter six
variables (V52 through V57) measured the cumulative soil moisture from
the surface to a specified depth (e.g. 0-30cm., 0-60cm., 9-90cm,etc.).
Temperature, moisture, and rainfall data were missing from the
annual reports in many cases, complicating the statistical analysis.
Because of the missing data for these variables, care must be taken
in interpreting the significance of some of the regression results,
particularly those in which the dependent variable was a yield
variable (V6 through V10).
More will be said about this problem
in the section on statistical methods.
Table 7, which presents all
of the variables used in the analysis, also details the severity of.
missing data for each variable.
Determination of Latitude, Elevation, and Geographical Location
Latitude and elevation of each site were both determined by
obtaining the appropriate topographical map which could be used in
conjunction with the legal description of a particular site.
Latitude
was introduced into the analysis as a four-digit number recorded to
the nearest minute.
Elevation, in most cases, was determined ± 7.5
meters.
Geographical location was an arbitrarily derived variable used
in the analysis.
Its derivation originated from the desire to
separate the plot sites according to the area of the state of Montana
30
in which they were established.
These areas were defined on the
basis of the proximity of the sites to one another and according to
known growing season differences, such as usual time of harvest and
general climatic conditions.
Four areas were thus delineated and a
number (1-4) used as the assigned coded value for the variable
(Figure 2).
Determination of Percent K Response
For those experiments with five treatments of applied K fertilizer
including the control (0 kg K/ha), a quadratic multiple regression
analysis was used to obtain the calculated maximum percent K response.
This method of percent response determination was described by Schaff
(1979).
The equation obtained from the regression analysis is as
follows:
Y = a + b 1 X + b 2 X2
where;
Y = calculated maximum yield
a = intercept of the Y axis; X = O
b^ = linear regression coefficient
bg = quadratic regression coefficient
X = rate of applied K fertilizer
If the derivative of the original equation is calculated, the
rate of K fertilizer needed for maximum yield can be determined.
The
W-4 4
W-U
W - 14
Figure 2.
VV- 64
Delineation of geographic location (V27) and the number
of experiments according to crop in each. W = winter wheat,
S = spring wheat, B = barley, 0 = oats.
32
equation becomes:
Y » b1 + 2 b2 X
By setting.Y = 0 and solving for X:
2 b 2 X + b1 = 0
where; X = rate of K fertilizer to obtain maximum yield response..
By substituting this value of X into the original equation, one
can then solve for maximum yield (Y).
After the maximum yield is
determined, then a simple ratio (Y/a) provides Y, or the percent
change in yield (response).
For those sites with only 2, 3, or 4 treatments of K fertilizer
including the control (0 kg/ha K ) , a second method was used to
calculate percent K response.
This method was used as an alternative
because quadratic multiple regression analysis becomes more meaningless
as the number of points to which a curve is adjusted decreases.
In
other words, a perfect-fit quadratic curve will always result in
instances of the presence of only three points.
This second method of determining percent K response is less
involved.
The yield at each rate of K application (other than the
control) was divided by the yield at the control rate (0 kg K/ha).
These percent response values were then added together and divided by
the number of values so determined.
percent K response.
This value conveys the average
33
The following equation describes the procedure.
Assume an
experiment in which 0, 22, and 48 kg/ha added K were the treatments
used, then:
where:
= yield at 22 kg/ha K
Y^g = yield at 48 kg/ha K
Yg = yield of the control
%K
= average percent K response
res
These two methods of percent K response determination were
assumed to provide similar values and were used as the same dependent
variable (V42) in the subsequent multiple stepwise linear regression
analysis.
A third determination of percent K response utilized a method
similar to the second.
In this case, the greatest yield value for a
given experiment (other than the control yield) was divided by the
yield value of the control.
K response."
This value was termed "maximum percent
This method was used over all of the experiments (222)
and the results introduced into the analysis as a second dependent
variable (V50).
Appendix III presents site number,
experiment number,
average percent K response, maximum percent K response, and the K rate
corresponding to maximum percent K response.
34
Statistical Methods
The variables used in the analysis (Table 7) were inserted into
an SPSS multiple stepwise linear regression program (Nie et a l ., 1975).
This regression program was chosen because it contained an option which
could manage cases with missing data for a given variable.
Therefore,
a case (experiment) with missing data for certain variables was not
completely eliminated from the analysis; and the data base was pre­
served for those variables for which data were present.
I
Because of the large amount of missing spring soil moisture,
soil temperature, and rainfall data, for example, the data base would
have been reduced drastically had not this approach been taken.
However, because this approach was chosen, care must be taken in
interpreting the results with regard to variables for which only
minimum data existed.
The analysis of R
2
values and the interpretation of significant
F-values should be based on the number of sites with complete data.
Results including variables with missing values will tend to inflate
the R
2
value for the equation; the degrees of freedom used in deter­
mining the significance of the F-value will also tend to be inflated.
For example, if data are missing for a given variable at "n" number
of sites, then the R
of sites minus "n."
2
for that variable is based on the total number
The overall regression equation R
is based on the total number of sites analyzed.
2
value, however,
More restrictive
35
Table 7.
Variable
Number
I
2
3
4
5i/
6T/
7I/
4/
4/
10ii
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41i/
4243
44
45
46
47
48
49i/
SOi'
51
52
53
54
55
56
57
~
y
Variables used in Multiple Stepwise Linear
Regression Analysis
Variable Name
Site
Site-Experiment
Separation Number
Crop
Year
Card Number
Yield I
Yield 2
Yield 3
Yield 4
Yield 5
K-Trt. I
K-Trt. 2
K-Trt. 3
K-Trt. 4
K-Trt. 5
Dry Consistence A
Dry Consistence B
Dry consistence Cca
Bulk Density A
Bulk Density B
Bulk Density Cca
Structure A
Structure B
Structure Cca
Textural class
Textural family
Geographic Location
Thickness of B
Depth to Ca
Elevation
Slope
Aspect
Latltude
Mean annual soil
Temperature (MAST)
Temperature Regime
Temperature (April)
Temperature (May)
Temperature (June)
Temperature (July)
Rainfall
Water Regime
Average Z K response
Total Spring Soil H O
Spring Soil H 1O (O-M)
Spring Soil H M (30-60)
Spring Soil H M (60-90)
Spring Soil H,0 (90-122)
Spring Soil H1O (122-152)
Spring Soil H M (152-183)
Maximum Z K response
K-rate for max. response
Spring Soil H 1O (0-30)
Spring Soil H M (0-60)
Spring Soil H M (0-90)
Spring Soil H M (0-122)
Spring Soil H M (0-152
Spring Soil HjO (0-183)
Units
Card //
a number (1-127)
1,2,3
a number (1-4)
Coded Value-a number (68-80)
a number (1-3)
Kg/ha
Kg/ha
Kg/ha
Kg/ha
Kg/ha
Kg/ha
Kg/ha
Kg/ha
KgZha
Kg/haKg/cm
KgZ cm.
KgZ
gZ ClDj
g/cm3
g/cm
.
Coded value^y
Coded valuery
Coded value%y
Coded valuery
Coded valuery
Coded value-'
1,2.
1,2,
1.2,
1,2.
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
2
2
2
cm
meters
percent
_y
Coded value— '
degrees (°)
and minutes (’)
°C
2/
Coded value— ^
•c
eC
eC
eC
cm.
2/
Coded valuepercent
cm.
cm.
cm.
cm.
cm.
percent
KgZha
cm.
cm.
cm.
cm.
Format
F 3.0
&
i
6
&
3
3
3
3
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
I .0
1.0
2.0
I .0
4.0
4.0
4.0
4.0
4.0
1.0
2.0
3.0
2.0
3.0
2.1
2.1
2.1
3.2
3.2
3.2
3.0
3.0
3.0
2.0
2.0
1.0
2.0
2.0
4.0
2.1
2.0
2
F 4.2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
3.1
1.0
2.0
2.0
2.0
2.0
3.1
2.0
3.0
4.1
3.1
3.1
3.1
3.1
3.1
3.1
3.0
3.0
4.0
4.0
4.0
4.0
4.0
4.0
Cases with
Missing Data
0
0
0
0
0
0
0
16
135
119
0
0
16
135
119
0
20
4
0
20
4
42
56
42
36
37
0
20
4
0
0
10
0
36
0
194
179
176
177
150
37
0
94
99
99
100
107
145
172
0
0
99
99
99
99
99
99
Used as a dependent variable in the regression analysis.
Coded value; See Tables 1-7 for coding schemes used to Introduce the variables into the
analysis.
36
options of analysis were used later in the investigation to try to
remedy this situation and to insure statistical validity in the
results.
These are described in the Results and Discussion section.
Due to the fact that over 60 separate regressions were conducted
and because of the large number of independent variables, the number
of variables allowed to enter the equation, the F-value, and tolerance
level were set at 5, 1.0, and .10, respectively.
This was done in
order to restrict the length of the resulting regression equation
and to give some assurance of the significance of the results.
level of statistical significance was determined from standard
F-tables (Steel and Torrie, 1960).
The
Chapter 4
RESULTS AND DISCUSSION
Two preliminary investigations were performed prior to the
multiple regression analyses which involved yield and percent response
as the dependent variables.
First, an analysis of variance (ANOV)
was conducted, with bulk density as the dependent variable.
The
bulk density values were grouped both on site and horizons within
a site.
This analysis was done to determine if statistically signifi­
cant differences between plot locations existed for one of the
variables (bulk density) that would be used later as an independent
variable in the regression analyses.
The ANOV produced an F-value of 9.15 when comparing one site to
another.
At I and 126 degrees of freedom, these results are
significant at the .005 significance level.
An F-value of .20
resulted when comparing horizon bulk densities at a given site.
At
■■
I and 2 degrees of freedom, this was not statistically significant.
These results establish that horizon differences at a site are not
apparent, but that differences in bulk density between sites are
pronounced.
The second preliminary investigation was conducted to establish
if ped thickness was an important, factor influencing the dry
consistence measurements.
This was done to establish whether or not
corrections in the dry consistence values would have to be made for
38
horizons
with either low or high mean values for ped thickness.
A regression analysis was performed using consistence values as
the dependent variable and ped thickness.as the independent variable.
The F-value produced from this regression was 130.8, significant at
the .005 level.
However, the R
equation was only .07.
value resulting from the regression
These results indicate that ped thickness
does influence the consistence measurement but that it does not ■
account for a very large amount of the observed variability.
If the
pocket, penetrometer was to be used in the field to measure consistence,
uniform ped thickness would be desirable.
Because care was taken to
,select ped sizes of a comparable range of thicknesses for each
horizon and because other factors obviously influence the consistence
values, it was determined that correction of the dry consistence
data because of ped thickness was not
necessary.
The discussion of results which follows in this chapter is
organized by the order in which the regression analyses were run.
In
each regression analysis, data were organized differently to reveal
independent variable relationships with different dependent variables.
Divisions of the data file by crop, geographical location (see Figure
2), percent response class, and by K-rate at which maximum response
occurred are. explained at the beginning of each discussion section
dealing with a. specific group of regression equations.
Each group of
regression equations was organized and separated on the. basis .of how
39
the data file divisions were made.
Each regression table presented in the following discussion
sections was organized in a similar mariner.
In the column located
ori the extreme left of each table (refer to Table 8, pg 41), the
regression number and the dependent variable and .the independent
variables used in the analysis were given,.
Also, when it applied,
the part,of the data file included in the regression analysis was
identified in this column.
The coding schemes used for crop (Table
6, p g . 27) and geographical location (Figure 2, pg. 31) were used
to identify the .part of the data file in which the regression
equation applied.
The division of the data into percent response
classes and according to K-rate at which maximum response occurred
(V51), when applicable, were also designated in the extreme left
hand column of the. table.
Also included in this column in some
instances are the number of cases (experiments) to which the regression
equation applies.
A complete list of.the variables used in the analyses can be
found in Table 7 (pg. 35).
It will need to be referred to in
interpreting the regression analysis results, presented in Table 8
through Table 17 and the summary tables (Table 18 through 20) .
Because of the large number of variables used in this study,
some: limitations were applied to each regression analysis.
One of
these limitations was to allow only a set number of variables to
40
enter each regression equation. ; In regression No. I through No. 14,
ten variables were allowed to enter the equation.
In regression No.
15 through No. 5.1, a maximum of five variables were allowed to enter
the equation.
It was felt that this limitation would still allow for
identification of all important relationships. ■
,Table No. 11, Regression No. 15 (pg. 60), can be used as an
example of how the actual regression equation would be constructed
from the table using step number, constant, and beta values.
This
equation would read:
. Y - 78.82 - .186
+ .145X2 + .164X
+..i47X^
where:
Y = maximum percent yield response (V50)
X1 ,= value of V28
1
X
\
’
= value of V17
2
'
X^ = value of V30
X. = value of V19
4
Regressions.N o . I - No. 5:
Yield as the Dependent Variable
The following regression analyses were conducted using V06 .
through VlO (actual crop yield at the various K rates) as the depend­
ent variables.
In this group of analyses,. ten variables were allowed
to enter each of the regression equations.
The independent variables
for each analysis were the same and included those variables which
were listed in the table of the regression results (Table 8).
The
Table &.
R e g r e s s i o n No. I - No. 5; A c t u a l Crop Y i e l d (V06-V10) as the
Depend e n t Variable; D a t a fi l e n o t Subdivided; Independent
Variables not R e s t r i c t e d . .
Regression
No..Dependent
Var. VO6 .
Independent
Vars. V O l , '
V03, V04,
V H , V16,
to V41, V43
to V49: ■
.2
Dependent •
Var. V07
Independent
Vars. VOlj
V03» V04,
Vi2, Vl6 to
V 4 l , V43 to
V49
Step Variable
into eq.
Name
I
2
3
4
5
6
7
8
. 9
10
Variable
Number
Constant
V23
Strue. B
.V34
,'MAST
V03
Crop
■
Vl 6 '
Dry Con. A
2450.19
V35
Temp.cls.
Temp.(May)
V3.7
V18
Dry Con. C
■
Bulk Den.C ■-V21- Rainfall
V40
Soil H 2O
(90-122)
.V47 '
Struc. B
V23 ■
I
Vl 6
Dry Coru. A
2
V03
3 ' Crop
.
V34
MAST
4 .
Vl
2
2306.73
K-trt.
2
5
6
. Thickness ,
V28
of B
Soil H 2O
'7.,
>44
(0-30)
8 . Temp.(May)
V37
9
Temp. cls. - V35
10
Soil H 2O
(152-183) V49
Beta
Simple
R
R
F-sig.
level
.161 '
.41
.005
.17 '
,21
.353
• .39
■ .005
-.217.
-.27
.24
' .005
.100
.16
' .26
.005
-.01 ■ .28
-.248
. .005
-;288
' -.04 .
.30
.005
, ..13 ■
.235
.31
.005
-.160
.03
.33
. .005
.121
, .21
.34
.005
.133
.129
.201
-.172
• .288
-.181
'
'
.06
.35
.005
..43
■ .18
.22
.20
-.29 .
.26
.39
.29
-.22
.30
.005
.005
.005
.005
.005
.095
.13
.32 .
-005
.303
-.300
: -.167
.16
-.09
.04
.34.36
.37
.005
.005
.005
■-.145
-.04
.38
.005
Table 8, Continued.
R e g r e s s i o n Np. I - No. 5; A c t u a l Crop Y i e l d (V06-V10) as
the Depend e n t V a r i a b l e ; D a t a Fi l e not Subdivided; I n d ependent Variables
not Restricted.
Regression
No.
V49
Dependent
Var. V09
Independent
Vars. VOl,
V03, V04,
V14, V16 to
V41, V43 to
V49
■ I
2
3
■4:
5
67
8
9
10
Bulk Den.A
Struc. B ■
Crop
Rainfall
Dry Con.A
Temp. (June)
MAST '
Tex. cls. •
Year
Site
I. • K-trt-4
2
■ Bulk Den.A
Water Reg.
3
Dry Con.A
4
Temp.(June)
5
Site
6
Geog. Ioc.
"7
Struc.C
8
Dry Con.B
9
Year
10
Constant
Vl 9
V23 .
V03 ■
■
V40 '
Vl6 ■ -1386.72 .
V38
V34
V25
V04
VOl
V14
.V19
V41 .
V16
V38
VOl
V27
V24 ..
Vl 7.
V04
Beta
-.611 . -. 44
.30
.172
-.251
-.26
.22
.238
.210
-.10
.03
-.191 .29
.234
-.181
.13
.231
.14
- .165 ' . — .03
.844
-.251
.089
.103
4285.93
-.080
.150
■ -.093.
.062
.061
. -.074
2
R .
Simple
R
.84
— .18
. .16
.01
-.32.
.04
.. -.15
• .02
■ -.01
.06
■
.19;
.30
.36
.40
.42.
.44
.46
.47
.48
.49
F-sig.
level
.005
.005
,005
.005
.005
.005
.005
.005
.005
. .005
.70
.005
.76
.005
.77 ■
.005
.78 ■
.005
.79
.005
.79
.005
.80
.005
.80
.005
' .005
.81
.005
. .81
42
3
Dependent
Var. VO8
Independent
Vars. VOl,
V03, V04,
V13, Vl6 to
V41, V43 to
Step . Variable . Variable
into eq.
Name
Number
Table 8, Continued.
R e g r e s s i o n No I. - No. 5; A c t u a l Crop Y i e l d (V06-V10) as
the Depend e n t Variable; D a t a Fi l e no t Subdivided; Independent Var i a b l e
not Restricted.
Regression
No.
5
Dependent
Var. VlO
.Independent
Vars. VOl,
V03, V04,
V15, Vl6 to
V41, V43 to
V49
Step Variable
into eq.
Name
I
2
3
4
5
■6
7
8
9
10
Variable
Number
K-t'rt.5
Bulk Den. A
Water Reg,
Aspect
. Rainfall ■
Temp.(May)
Dry Con.A.
Soil H 9O(152-183)
Soil H O
(90-122)
Soil H O
(0-30)
Constant
Beta
R^
F^sig.
level
.664
— .468
.180
-,.080
•.135
-.278
'.124
.69
-,42
•.29'
.01
.03.
-.05 .
-.09
-.184
.07
V47
.267
.19
.76
.005
V44
-.111
-.10
.76
. .005
V15
V19
V41
V32
V40
V37
Vl 6
V49
.
' Simple
R
3932.32
'' .
•48
.68
.70 . .'
.72
.73
.73
.74
.005
.005
.005
.005
.005
.005
.005
.005
44
analyses were conducted over the entire data set so that the results
were based on 222 cases, the total number of experiments included in
this study.
Regression No'. 4 and No. 5 (actual crop yield as the dependent
variable) produced the highest R
2
values, .81 and .76, respectively.
The F-signif.icance level of the results was .005 at each step in
all five of the regression equations.
and probably misleading.
Methods section that R
2
This was somewhat surprising
It was pointed out in the Materials and
values and F-signif icanc.e levels tend to. be
inflated by including variables with missing data into the regression
analysis.
The variables for which this was particularly true were
those associated with the seasonal temperature data (V36, V 37, V28,
V39), rainfall data (V40), and spring soil moisture data (V43 through
V49).
Although other variables which appeared in the regression
equations also had missing data, the severity was not. as pronounced
(See Table 7 for the number of missing cases for each variable.
Simple correlations (Simple R ) , however, for each of the variables
in the equation indicated valid relationships to the dependent variable.
It is important to note the strong positive correlation between
structure of the B horizon (V23) and yield in regression No. I, No.
2, and No. 3.
This indicated.that higher yields were associated with
the more strongly structured soils.
Thickness of the B horizon
(V28) was also positively correlated with yield in regression No. 2.
45
Mean annual soil temperature (MAST) was positively correlated
to yield in regression No. I, No. 2, and No. 3.
This suggested
that lower soil temperatures associated with some soils may have been
yield-limiting.
Yield (V09 and V10) in regression No. 4 and No. 5 showed a
strong positive correlation to the rate of applied K (V14 and V15,
respectively).
This indicates that availability of native soil K and/
or low rates of applied K are not adequate in many cases.
Dry consistence of the Ap horizon (V16) was the only variable
to appear in all five of the regression equations.
Its correlation
(Simple R) to yield ranged from slightly positive to slightly
negative.
However, bulk density of the Ap horizon (V19) showed a
consistently strong negative correlation to yield in regression No.
3, No. 4, and No. 5.
The literature generally supported this
observation in that increased bulk densities were associated with
decreased root penetration and nutrient uptake.
In the surface
horizon (Ap) high bulk densities may also have been associated with
increased difficulty of seedling emergence.
Rainfall (V40) was positively associated with yield in regression
No. I 1 No. 3, and No. 5; yet moisture regime (V41), which was coded
wet to dry, was also positively correlated to yield in regression
No. 4 and No. 5.
This latter relationship of moisture regime to
yield seems to be in agreement with the positive association between
46
MAST and yield.
In regression No. 51, MAST and moisture regime were highly
positively correlated to one another.
The warmer soils (> MAST)
would be expected to be associated with the drier moisture regimes
in that soil moisture would have exerted a profound buffering
influence on soil temperature.
Temperature in May (V37) in regression No. I, Nb. 3, and No. 5
as well as temperature in June (V38) in regression No. 4 showed a
strong negative correlation with yield.
This at first seemed to.
have contradicted the relationship between MAST and yield.
However,
the value of the MAST whs determined from long-term temperatue data
and so was an average.
That low
early growing season temperatures
are correlated with greater yields could be explained by the
fact that low soil temperatures have been associated with greater
stored soil moisture.
Crop differences existed with regard to yield.
This was shown
by the fact that crop (V03) was an important variable in regresison
No. I, No. 2, and No. 3.
Varietal and crop differences with regard
to nutrient uptake may, be due to differences in root membrane
activity; the selectivity for certain ion absorption may be an
inherited trait (Mattson, 1974).
Lal and Sharma (1974) found that
two varieties of dwarf wheat differed significantly in their ability
to extract .N, P, and K. , Their experimental results were consistent
47
over three levels' of soil moisture and five rates of applied N-
Regressions No. 6 - No. 10:
Average Percent Yield Response (V42) as
the Dependent Variable.*
2
The regression analyses were performed separately for each crop
(regression No. 6 for winter wheat, Nb. 7 for spring wheat, No. 8
for barley and No. 10 for winter wheat) and combined over all crops
(regression No. 9).
The data subfile for oats was not analyzed
■
separately throughout the regression analysis because there were
only four experiments with oats as the experimental crop.
Regression No. 6 and No. 10 differ only in that VOl and V04
(site and year) were removed as independent variables in regression
No. 10.
The removal of VOl and V04 from regression No. 10 and
subsequent regression analyses was done because it was decided that
no explicit meaningful information could be derived from their
inclusion in the regression equation.
Crop (V03) was also removed as an independent variable from
these and subsequent regression analyses as the data were divided
and analyzed according to crop.
The highest R
2
values were obtained for spring, wheat and barley
2
(regression No. 7 and No; 8).
respectively.
The R
values were .57 and .45,
However, the highest .F-significance level (.005)
was attained for winter wheat (regression No. 6 and No. 10) and in
Regression
No.
Regressions No. 6 — No. 10; A v e r a g e P e r c e n t R e s p o n s e (V42) as the Dependent
Variable; D a t a Fi l e Subdivided on C r o p ; Independent Va r i a b l e s not Restricted,
Step
into eq.
6
Dependent
I
Var. V42
2
Independent
3
Vars. VOl,
4
V04, Vll to
5
V41, V43 to
6
V49
7
Subfiles:
8
Divided on
9
crop; winter 10
wheat (133
cases)
7
Dependent
I
V a r . V42
2
Independent
Vars. VOl,
3
V04, Vll to
4
V41, V43 to
5
V49
6
Subfiles:
Divided on
7
crop; spring 8
wheat (36
9
cases)
10.
Variable
Name
Latitude
Rainfall
Dry Con.B
Bulk Den.B
Site
Struc.A
Struc.B
Slope
Temp.(May)
K-trt.2
Variable
Number
V27
V40
Vl 7
V20
VOl
V22
V23
V31
V37
Vl 2
Constant
93.03
Beta
Simple
R
R2
.241
.210
.313
-.244
-.285
.285
-.196
-.099
-.207
.264
.26
.20
.14
.04
-.01
.13
.06
-.04
.01
.09
.07
.08
.10
.11
.13
.14
.15
.16
.17
.19
F-sig.
level
. .005
.005
.005
.005
.005
.005
.005
.005
.005
.005
oo
Latitude
Soil H O
(30-60)
Dry Con.A
Slope .
Bulk Den.C
Thickness
of B
Rainfall
K-trt.4
Bulk Den.A
Soil H O
(90-122)
-.23
.05
-1.461
-.566
.713
-. 380
-.15
-.16
.06 .
-.14
.09
.14
.20
.25
V28
V40
Vl 4
Vl 9
.458
-1.058
.916
.627
.10
-.17
.04
-.10
.31
.33
.39
.48
.10
.10
.10
.025
V47
.885
i
O
kO
Table 9.
.57
.01
V33
-.163
V45
V16
V31
V21
284.78
—
— ---- —
—
—
—
■
Table 9, Continued. Regressions No. 6 - No. 10; Average Percent Response (V42)
Dependent Variable; Data File Subdivided on Crop; Independent Variable
Restricted
2
R
Simple
Regression
Step Variable• Variable
Constant
Beta
Name
No.
Number
R
into eq.
8
Dependent
I
V a r . V42
2
•Independent
3
Vars. V01,
4
V04, Vll to
.5
V41, V43 to
6
V49
7
Subfiles:
8
Divided on
9
crop; barley
(49 cases)
10
9
Dependent
I
Var. V42
2
Independent
3
Vars. VOl,
4
V04, Vll to
V41, V43 to . 5
V49
Subfiles:
6
Divided on
7
crop: over­ 8
all (222
cases)
9
10
.
as the
not
F-sig.
level
.540
.380
.486
.151
.238
-.347
.235
-.173
.29
.26
.13
.05
.10
-.17
-.19
.07
.08
.15
.25
.29
.31
.34
.38
.39
V48
-.547
-.10
.41
.01
V49
.471
.10
.45
.01
V15
Vl 3
Vl 7
.371
.254
.241
.13
.11
.11
.02
.06
.08
.10
.005
.005
V44
.376
.08
.09
.005
-.486
-.165
-.157
-.04
-.03
.08
.12
.14
.14
.005
.005
.005
V46
.632
.02
.15
.005
V45
V39
— .483
.091
.002
.04 .
.16
.17
.005
.005
K-trt.3
K-trt.5
Temp.(July)
Dry Con.B
Elevation
Year
Site
Aspect
Soil H O
(122-152)
Soil H O
(152-183)
V13
V15
V39
V17
V30
V04
VOl
V32
K-trt.5
K-trt.3
Dry Con.B
Soil H O
(0-30)
Soil H„0
. (90-122)
Year
Bulk Den.B
Soil H 9O
(60-90)
Soil H O
(30-60)
Temp.(July)
V47
V04
V20
221.47
141.19
.05
.025
.005
.005
. .01
.01
.005
.01
Table 9, Continued.
Regressions No. 6 - No. 10; A v e r a g e Percent Response (V42) as the
Dependent Variable; Date Fi l e S ubdivided on Crop; Independent V a r i a b l e n o t
Restricted.
Step
into
Variable
Name
Variable
' Number
Constant .
Beta
'
2
Simple
R
R
.162
.356
-.167
-.121
'.19
.19
.08
-.07
.04
.06
.08
.09
.05
■ .025
.01
.025
.079
-.997
.738
.15
.02
.09
.10
.11
.15
.025
.025
■ .005
O
CM
CO
Regression
No;.
,F-sig.
level
-.06
■ .16
- .005
.335
.192
-.008
; ■ .14
.18
.19
.005
.005
10
10
Geog. Ioc.
Dry Con.B
Bulk Den.B
Depth to Ca
Soil H-O
(0-30)
Temp.(May)
Temp.(June)
Soil. H 9O '
(152-183)
Soil H 9O
■ (122-152)
K-trt.2
.
V27
Vl 7
V20
V29
V44
V37
V38
.
92.99
V49
I
Dependent
I
V a r . V42
2
Independent
3
Vars. Vll to .4
V41, V43 to
5
V49.
Subfiles:
6
Divided on
7
winter wheat. 8 .
(133 cases)
9
-
V48
Vl 2
51
the case of the overall regression analysis (regression No. 9).
Again, it is stressed that when variables with missing data are
included in the analysis the R
2
and F values are probably inflated.
Spring soil moisture (V43 to V49), rainfall (V40), and temperature
(V36 to V39) variables appeared in all the regression equations.
The presence of these variables in the regression equations meant
they were subject to misinterpretations.
It was somewhat contradictory that rainfall (V40) was positively
correlated to percent yield response in regression No. 6 and nega­
tively correlated in regression No. 7.
This difference could
possibly be explained by assuming that the addition of water to the
soil may have produced variable influences op crop response to added
K under different soil conditions.
If the soil was dry so that no
K diffusion could take place, then rainfall should have had a
positive influence on crop response.
Under adequate soil moisture
conditions in which diffusion of native K to the plant root is
sufficient, the general cooler climatic conditions (cloud.cover, etc.) associated with rainfall may possibly have acted to depress
crop response to added K.
Distribution of rainfall and soil perme­
ability may also have been important factors in the explanation of
these results.
Both dry consistence of the B horizon (V17) and bulk density
of the B horizon (V20) were positively correlated to percent
52
response in all cases in which they appeared in the regression,
equations.
High bulk densities are usually associated with less
plant available water..
This is due to the increased raatric potential.
If high bulk densities were restricting diffusion of native K,
one would have expected the observed positive correlation between
bulk density, and crop response to added K.
increases as clay content increases.
zone of clay accumulation;
Dry consistence probably
The B horizon is the major ■
The higher CEC values associated with
increasing clay content (especially considering that 2:1 clays are
dominant in Montana soils) may also be associated with restricting
diffusion of native K.
These results were in accord with observations
presented in the Literature Review section.
In the one'case in which thickness p f ■the B horizon.(V28) was
.
included in the regression equation (regression Ho. 7), it was
also positively correlated to crop response.
Structure of the B
horizon (V23) was also positively correlated to winter wheat response
■(regression No. 6).
This relationship between strongly structured
soils and positive crop response to added K may be due to the fact
that strongly structured soils are.usually associated with soils
that have an abundance of clay and experience a definite dry period
during any given year.
Both of these factors would tend to cause
slower or more tortuous diffusion of K in the soil.
The K-rate variables (Vl2, V13, V15) were positively correlated
53
to crop response, in regression No. 6 through No. 10.
This again
suggested that, in many cases, the availability of native K was not
adequate.
Although subsoil moisture was somewhat variable in its correla
tion to crop response, spring soil moisture in the surface 30 cm.
(V44) was positively correlated with crop response (regression No.
9 and No. 10).
This indicated that adequate' surface soil, moisture
may have been a prerequisite in order for added. K to have had a
positive effect.
The strong negative correlation of latitude (V33) to crop
response is of interest in regression No. 7.
Soils in the higher
latitudes generally warm up more slowly in the spring;
The
potassium requirement of small grains is high early in the growing
season.
If temperature was the limiting factor, a positive
correlation between latitude and crop response would be expected. .
However, other factors may have influenced the effect of latitude.
Perhaps low temperature was still the limiting factor in that the
effects of applied K were not realized, in the colder soils.
Also,
higher rates of K application may have been required in colder
soils to provide a crop yield increase.
Regressions No. 11
No. 14;
Maximum Percent Yield Response (V50)
as the Dependent Variable.
These regression analyses were conducted separately for each
Table 10.
Regressions No. 11 - No. 14; Maximum Percent Response (V50) as the Dependent
Variable; Data File Subdivided on Crop; Independent Variable VOl-VlS Not Included.
Regression
No.
Step
into eq.
Rainfall
Elevation
K-rate for
max. res.
Struc. B
Soil H O
(90-122)
Thickness
of B
. Dry Con.B
Bulk Den.B
Temp.(May)
Temp.(June)
Variable
Number
Constant
Beta
Simple.
R
. R2
F-sig.
level
V40
V 30
.182
.093
.20
.16
.04
.06
.025
.025
V5I
•V23
.189
-.155
.14
-.10
.08
.10
.025
.01
-.158
-.04
.12
.005
V28
Vl 7
V20
V37
V38
-.116
.221
-.168
-.802
.789
-.17
.08
-.02
.03
.10
.14
.15
.15
.16
.19
.005
. .01
.01
.01
.005
V38
V31
V 30
V41
V25
V16
V19
-1.803
-.011
-.245
1.508
-1.436
.837
-.805
-.29
.26
. -.20
.15
-.08
.03
-.15
.08
. .21
.28
.31
.39
.46
.55
.005
.005
V51
V39
.306
1.671
.20
-.28
.61
.64
.005
.005
V45
-.176
O
i—i
I
11
I
Dependent
„ 2
Var. V50
3
Independent
Vars. V16
4 ■
to V41, V43
5 .
to V49, V51
Subfiles:
6
Divided on
crop; winter 7
wheat
8
(133 cases)
9
10
Variable
Name
.66
.005
V47
101.48
12
Dependent
I
V a r . V50
2
Independent
3
Vars. V16
4
to V41, V43
5
to V49, V51
6
Subfiles:
7
Divided on
8
crop: spring
wheat .
9
(36 cases)
10
Temp.(June)
Slope
Elevation
Water Reg.
Tex.cls.
Dry Con.A
Bulk Den.A
K-rate for
max. res.
Temp.(July)
Soil H 9O
(30-60)
184.17
.10
.025
.025
.025
.01
2
Table 10, Continued. Regressions No. 11 - No. 14; Maximum Percent Response (V50) as the
Dependent Variable; Data File Subdivided on Crop; Independent Variable V01-V15
Not Included.
n
Regression
No.
Step
into eq.
13
Dependent
Var. V50
Independent
Vars. Vll
to V41, V43
to V49, V51
Subfiles:
Divided on
crop: barley
(49 cases)
I
2
3
4
5
6
7
8
9
10
14
Dependent
Var. V50
Independent
Vars Vl6 to
V41, V43 to
V49, V51
Subfiles:
Divided on
crop: overall
(222 cases)
I
2
3
4
5
6
7
8
9
10
.
Variable
Name
' Variable
Number
K-trt.5
K-trt.3 '
Elevation
K-trt.2
K-rate for
max. res.
■ Temp.(June)
Geog. Ioc.
Latitude
Soil H„0
(122-152)
Soil H O
(152-183)
Constant
R
F-sig.
level
.40
.31 .
.19
-.04
.16
.26
.32
.41
.005
.005
.005
.005
.207
.242
.220
.219
.33
.01
-.19
.13
.45
.49
.49
.50
.005
.005
.005
.005
V48
-.373
.04
.51
.005
V49
.355
.25
.53 .
.005
.251
-.154
.295
.130
.21
■ -.12
.11
.07
.04
.06
.08
.09
.005
.005
.005
.005
-.196
WV04
.10
.005
. .116
.104
.233
.151
.03
.03
— .01
-.03
.11
.11
.12
.12
.005
.005
.005
.005
-.051
— .08
.13
.005
V15
Vl 3
V30
V12
V51
V38
•V27
. V33
K-rate for
max. res. . V51
Struc. B
V23
V 30
Elevation
V40
Rainfall
Soil H 9O
(90-122)
V47
Soil H 9O
V49
(152-183)
V18
Dry Con.C
V33
Latitude
Geog. Loc.
V27
Thickness
V28
of B .
Beta
.491
.389
. .460
.326
-155.51
-50.16
Simple
R
•
56
crop as in t.he previous section (regression Mo. .11 for winter wheat,..
Mo. 12 for. spring wheat,, and. N o . 13 for barley).
Regression No. 14-.
was the combined overall regression and corresponded to regression
No. 9 in the previous group of regressions.
The inclusion of V51
(the K'-rate at which maximum response occurred) into these regression
analyses was done because the dependent variable (V50) was the one
with which it was associated.
The other Krrate variables were
excluded from these regressions except for in regression No. 13,
Regression No. 12 (spring wheat) and No, 13 (barley) produced
2
the highest R
values,
. 66 and
‘
.53, respectively.
were higher than the corresponding regression. R
group (regression No. 7 and No. 8).
values in the prior
The F-significance levels of
these regressions were also higher overall
being significant at the .005 level.
2
These values
each regression equation
Care must be taken in inter-
preting the significance of these results as the R
2
and F-values were
probably inflated due to. the inclusion of variables with missing data
into the analysis.
However, results indicated a good potential for
utilizing certain of these independent variables to predict crop
,
response to K fertilizer.
The K-rate variables (V51, V13, V15) again showed a pronounced
positive correlation to the dependent variable.
Variable 51 (K rate.
at which maximum response occurred) was postively correlated in all
four regressions.
Variable 15 and V13 (high and moderate rates of
•
.
.
57
.
K application) in regression No. 13 had Sinple-R values of .40 and
.31, respectively, with the dependent variable (V50).
Rainfall was positively correlated to maximum percent response
in both cases in which it appeared in the regression equation
(regression No. 11 arid No. 14).
Elevation (V30) was also positively
correlated to percent response.in three of the four regression
equations.
Elevation may have been an important factor with regard
to crop response to added K in the way that it may have modified
the climatic environment.
Elevation may have influenced soil
moisture and soil temperature in that wind, rain, and storm patterns
may have been affected.
Average temperature in May (V37.) and Jurie (V38) were only
slightly positively correlated to percent response in regression
No. 11 and No. 13.
There was a much stronger negative correlation
between V38 and average temperature in July (V39) in regression
No. 12..
Soil temperature probably.influenced crop response to added K
differently depending on other existing soil conditions.
correlation would be expected to exist,
A positive
between soil temperature and
crop response under conditions of adequate soil moisture.
Iricreasing
soil temperature in this case would have probably increased ion
diffusion of applied K.
On the other hand, a negative correlafion
would have existed between soil temperature and crop response
58
(particularly later in the growing season); if soil moisture was
inadequate.
In this case, greater soil temperatures could have been
associated with drying of the soil profile so that diffusion was
reduced.
Another aspect of soil temperature is that diffusion oj:." ;
K would normally increase with increasing temperature, allowing
native soil K to be more available, thus decreasing the response to
. ... ' ■ ■■
■
.
supplemental K.
'I,
.I fIt'
Spring soil moisture (V45, V47, V48, V49) was not consistently ■
correlated to crop response.
The greatest correlation, which was
.25 between deep soil moisture (V49) and maximum percent response,
appeared in regression No. 13.
Although snring soil moisture
variables appeared in all of the regression equations, they generally
.
2
entered toward the end of the equation, and increased the R values
very little.
Structure of the B horizon (V23) was negatively correlated to
crop response in regression No. 11 and No. 14.
results from previous regression groups.
This.contradicted
It was. difficult to
explain.the effect of soil structure on crop responses, especially
considering the way in which the soil structure variables were
derived and coded.
First, it was impossible to know if grade, size,
or type of structure was the dominant influence on crop response.
Second, because the structure variables were derived from the SCS
soil series description, it was possible that the coding scheme was
59
misleading for these particular soil sites.
Horizon depths and
designations in the soil series description usually did not
correspond to the method of horizon naming and sampling that was
done in the.field during this study.
Also, the soil series
description usually described horizon soil structures in terms of a.
range of structures (e.g. moderate medium angular blocky to strong
medium prismatic).
It was doubtful that a meaningful interpretation
could be derived from the structure variables as they were used in.
these analyses.
For these reasons, structure as an independent
variable was removed from subsequent analyses.
Adams and Wilde (1976) added some insight into this problem.
-■ .
'
:
They studied the variability in 16 morphological soil properties in
a soil mapping unit mapped as one soil series.
Maximum variability
was observed in structure grade and size and wet consistence.
Data
showing the variability within soil mapping units should be included
in all soil surveys to give users a better understanding of the
limits of their ability to generalize.
This may be very important
when making fertilizer recommendations for areas where limited
sampling has led to generalized soil mapping.
Regressions Mo. 15 r No. 19;
V50
or
V42 as the Dependent Variable..
Regression No. 15 through No. 19 were different from the previous
regressions in that the three structure variables, the average
Table 11
Regression
No;
Regression No. 15 - No. 19; Maximum (V50) and Average (V42) Percent Response as
the Dependent Variables; Data File Subdivided on Crop; VOl - V15 and Soil
Temperature , Rainfall , and Soil Moisture Variables Not Included as Independent
Variables.
Step
into eq.
15
Dependent
I
V a r . V50
Subfiles:
2
.Divided on
3
crop: Winter 4
Wheat (133)
•
Independent
Var.V16 to V21,
V25, V26, V28
to V35, V41
16
Dependent
I .
2
Var. V42
Subfiles:
3
Divided on
crop: Winter
wheat (133)
Independent
Vars. Vl6 to
V21, V25, V26,
V28 to V35,
V41
Variable
Name
Thickness
of B
Dry Con.B
Elevation
Bulk Den.A
Variable
Number
V28 .
Vl 7
V30 ,
V19
Constant
. 73.82
Beta
-.186
.145
.164
.147
Simple
R
-.17
.08
.16
.12
R2
F-sig.
level
.03
..05
.07
.09
.05
.05.
.05
.025
§
Dry Con.B
Bulk Den.B
Latitude
V17
V20
V33
153.89
.436
-,312
-.122
.19
.08
-.15
.03
.06
. .07
.05
.025
.025
Table 11, Continued.
R e g r e s s i o n No. 15- No. 19; M a x i m u m (V50) and A v e r a g e (V52) Percent
Response as the D e p e n d e n t Variables; D a t a F i l e S ubdivided on Crop; V Ol - V15 and
Soil Temperature, Rainfall, and Soil M o i s t u r e V a r i ables Not Included as Independent
Variables.
Regression
No.
Step
into eq.
.17.
Dependent
I
2
Var V50
Subfiles:
3.
Divided on
4
crop; Spring 5
wheat (36)
Independnet.
Vars. V16 to
V21, V25, V26,
V28 to V 3 5 ,
V41
18 .
Dependent
I
V a r . 42
Subfiles:
2
Divided, on
3
crop: Spring 4
Wheat (36)
Independent .
Vars: V16 to
V21, V25, V26,
V28 to V35,
.
V41 .
Variable
Name
• ■
Slope
Elevation
■ Water Reg.
Tex. cls. .
Dry Con.A
Variable
Number
V31
V30
V41
V25-'
V16
■
•'
Constant
94.62
Beta
. .276
-.003
1.256
-1.086
.330
Simple
R
.26
r-.20
.15
-.08
.03
R
2
.07
..12
, .19
.28
.36 •
' F-sig.
level
. .
--- :
—
.10
.05
.025
2-
’
Thickness
of B
Aspect
Water Reg.
Slope
V28
V32
' V41
V3l
99.46
i485
-.511
.293
.219
.32
-.31
.12
.03
.10
.26
.32
. .36
.10
.01
.01
.01
Table 11, Continued.
R e g r e s s i o n No. 15- No. 19; M a x i m u m (V50) and A v e r a g e ( V 4 2 ) .Percent
Response as the D e p e n d e n t Variables; Da t a File Subdivided on Crop; VOl - V l 5
and Soil Temperature, Rainfall, and Soil M o i s t u r e Va r i a b l e s Not Included as.
- Independent Variables.
Regression
No.
Step
into eq.
19
Dependent
I
Va r . V42
'2
Subfiles:.
Divided on
crop; overall
(22%)
Independent
Vars: V16 to
V21, V25, V26,
V28 to V35; V41
Variable
Name •
Dry Con.B
Latitude '
Variable
.Number
V17
V33
- Constant
145.09
Beta
.100
' -.094
Simple
R
.11
. -.11
R2
F-sig.
level
.01
.02
.10
.10
o>
NS
63
monthly temperature data, the spring soil moisture data, and the
rainfall data were excluded from the analyses.. This was done to
try to increase the statistical validity of the subsequent regression
equations, due to extensive missing data in each of these parameters.
Regression No. 15 and No. 16 differ only in that the dependent
variable was V50 (maximum percent K response) in the first case
and V42 (average percent K response) in the second, case for winter
wheat.
The R
2
and F-significance level indicated very little
difference in using V50 or V42. as the dependent variables.
Only
■four variables in the first case and three in the second were, signifi­
cantly (P = .10) related to crop yield responses.
Dry consistence .
of the B horizon (VI7), however, was the only variable common to
both equations.
Of. the seven variables included in both equations;
four of them are related to B horizon characteristics.
In both
equations and in regression No. 19 as well, bulk density and dry
consistence were positively correlated to percent yield response.
Increased, bulk density probably resulted in a reduction of plant
available water, reduced root proliferation, and restriction of
native K diffusion.
Higher values of dry consistence can be
associated with decreased root penetration and utilization of
native K, so that a positive resnonse to added K resulted.
A higher R
2value■'(.36)
. ■vwas
■
. " ■ ..
obtained from the spring wheat
sites (regression No. 17 and No. 18).
But, in regression No. 18
64
the F-value was more highly
significant (.01) than that for
regression No. 17 (.025) with V50 ,as the dependent variable.
Thickness of the B horizon (V28) was positively correlated to
crop response in regression No. 18 and negatively correlated in
regression No. 15.
Again, the effect of the thickness of the B
horizon.on crop, response to added K
may have been determined
by other factors which modified its effect (e.g. clay content,.
stored soil moisture).
-
,
Water regime was positively correlated to crop response in
both cases in which it appeared in the regression equation.
This
suggests that the soil types which are normally associated with the.
drier sites tend to show a greater crop response to added K.
'
Elevation was positively correlated to crop response in
regression No. 15 and negatively correlated in regression No. 17.
This indicated that elevation, by itself, was not a good parameter
.
for predicting crop response.
'
.
:
Its effect on crop response was
probably realized by the way in which it interrelated with other-,
factors.
The negative correlation of latitude to crop response in
regression No. 16 and No. 19 suggests that temperature in the more
northern latitudes may be limiting crop response to added K.
Slope
w a s .positively correlated to crop response in regression No. 17 and
No. 18.
Areas with greater slopes are usually associated with drier.
65
sites in that groundwater tends to move downslope^
If the site
aspect was oriented to the south, this drying effect associated
with greater.slope would be maximized.
Regressions No. 20 - Mo,. 31:
V50
or V42 as the Dependent Variable.,.
Geographic location (V27) appeared as a variable in several of
the previously discussed regression groups.
It was not clear,
however, because of the coding scheme employed, exactly for w h a t .
reason this variable was important to crop responsei
In order to
try to determine this ’’reason", the data for the present set of
regressions was divided and analyzed both in terms of crop type
(V03) and geographic location (V27).
...
The regression equations for this analysis were grouped in
pairs so that each pair concerned the same data set (identified
in the regression tables).
The only difference was that in the
'
first case of a given pair the dependent variable was maximum percent
response (V50), and in the second case it was average percent
response (V42).
Regression No. 30 and No. 31 yielded the highest R
.86.and L72, respectively.
2
values,
For regression No. 30 (maximum percent
response as the. dependent variable) the F-statistic was more highly
significant at the .01 level.. Both.equations concerned barley
experiments in the southeastern part of Montana.
Generally, these
soils tend to. be warmer during the growing season, and this is' the
Table 12.
Regression
No.
Regression No. 20 - No. 31; Maximum (V50) and Average (V42) Percent
Response as the Dependent Variables; Data File Subdivided on Crop
and Geographic Location; Independent Variables Restricted as in Table 11.
Step
Variable
into e q . .
Name
20
Dependent
I .
Var. V50
2
Subfiles:
3
Divided on:
4
Crop -I
Geog. loc.-l
(44)
Independent
Vars: - same as
Regression #15
Variable
Number
Bulk Den.C
Aspect
Dry Con.A
Depth to Ca
V21
V32
V16
V29
Bulk Den.C
Dry Con.A
Aspect
Depth to Ca
Dry Con.B
V21
V16
V32
V29
Vl 7
Constant
168.73
Beta
Simple
R .
R2 -
F-sig.
level
-.319
-.267
-.274
-.251
-.37
-.17
-.01
-.33
.14
.18
.21
.25
.025
.025
.025
.025
-.317
-.431
-.185
-.287
.221
-.35
-.14
-.09
-.26
.12
.12
.17
.22
.25
.29
- .025
.025
.025
.05
.025
21
Dependent.
I
V a r . V42
.2
Subfiles:
3
Divided on:
4 ^
Crop - I
5 Geog. loc.-l
(44)
Independent
Vars: - same as
Regression #15
159.27
Table 12, Continued.
R e g r e s s i o n No. 20 - No. 31;
M a x i m u m (V50) and. Average (V42)
Percent R e s p o n s e as the Dependent Variables; Da t a File Subdivided on Crop
a n d . G eographic Location; Indep e n d e n t Va r i a b l e s R e s t r i c t e d as in Table 11.
Regression
No.
' Step
into eq.
22
Dependent
I
Var. V50
2
Subfiles:
3
Divided on:
■4
Crop-2
•5
Geog.loe.-l
(25)
Independent
Vars: same as
Regression #15
23
Dependent
I
Var. V42
2
Subfiles:
3
4
Divided on:
Crop —2
5
Geog.loc.-l
(25)
Independent
Vars.: same as
Regression #15
■
Variable
Name
Latitude
Dry Con.C
Bulk Den.A
Aspect
Water Reg.
. Variable
Number
V33 .
Vl 8
V19
V32
V41
Constant
-306.00
Beta
.434
-.764
-.899
-.657
-.431
Simple
R.
.33
. -.32
-.30
-.27
..13
R
2
. .11
.27
.44
.73
.79
F-sig.
level
.
--.05
.01
.005
.005
O'
Aspect
.Bulk Den.A
Dry Con.C
Depth to Ca
Slope
V32.
. Vl 9
Vl 8
V29
V31
121.34
-.829
. — .40
-.01
-.037
-.31
. -.749
.781
.07
.347 : -.02
.16
.28
.39
. .54
.60
.05
.025
.025
.005
.005
Table 12, Continued.
Re g r e s s i o n No. 20 - No. 31; M a x i m u m (V50) an d Average (V42)
. Percent Response as the D e p endent Variables; D a t a File Subdivided on Crop
and Geographic Location; Independent Va r i a b l e s R e s t r i c t e d as in Table 11.
Regression Step
No.
into e q .
24
Dependent
I
V a r . V50
2 ■
Subfiles:
3.
Divided on:
Crop - 1,2,3 .
Geog. loc.-3
(26)
.
Independent
Vars.: same as
Regression #15
25
Dependent
-I
Var. V42
.2
Subfiles:
3
Divided on: ■ 4
crop - 1,2,3
Geog. Ioc. - 3
(26)
Independent
Vars: same as
Regression #15
Variable
Name
Bulk Den.C
Latitude
Aspect
Variable
Number
V21
V33
V32
Constant
978.91
Beta
. . .
Simple
R
.655
-.434
.231
.32
-.07
.OS
R
■
2
.10
.21
.26
F-sig.
level
—
.10
.10
CTi
00
MAST
Tex.F a m ,
Bulk Den.A
Slope
V34
V26
V19
V31
-39.99
.745
.474
.231
.331
.30.
.13
.11 .
.07
.08
.25 .
.30
.33
.05
.05
.10
\
Table 12, Continued.
Regression No. 20 - No. .31; Maximum (V50) and Average (V42)
ercent Response as the. Dependent Variables; Data File Subdivided on Crop
and Geographic Location; Independent Variables Restricted as in Table 11.
Regression Step
No.
. into eq.
26
Dependent
I ■
Va r . V50
2
Subfiles:
3
Divided on:
Crop - 1,3
Geog. Ioc. -4
(77)
Independent
Vars: same as
Regression #15
2.7
Dependent
I
Vars. V42
2
Subfiles:
3 .
Divided on: 4
Crop - 1,3
Geog. loc.-4
(77)
. .
Independent
Vars: same as
Regression #15
Variable
Name
Elevation ■
Tex. cls„
Dry Con.B
Variable
Number
V30.
V25
Vl 7
Constant
93.93
Beta
Simple
R
R^
' .385
-.265
.154
.23
-.002
.03
.05
" .07
.09
F-sig.
level
.
.05
.10
.10
o
Elevation
Tex. cls.
Dry Con.B
Latitude
V30
V25
Vl 7
. V33. ■
-71.15
.448
-.355
.358
.151
.15
-.05
.14
— .06
.02
.05
.11
.13
•------------ — —
.05
.05
Table 12, Continued.
Re g r e s s i o n No. 20 - No. 31; M a x i m u m (V50) and Average (V42).
Percent Response as the De p e n d e n t V a r i a b l e s ; Date File Subdivided on Crop
and Geographic Location; I n d ependent Va r i a b l e s R e s t r i c t e d as in Table 11.
Regression Step
Nb.
into e q .
Variable
Name.
Variable .
Number
Constant
Elevation
Tex. els.
Dry Con.B
.V30
■725
Vl 7 ■
92.79
.421
-.307
.208
Elevation
Dry Con.B
Tex. cls,
V30 .
Vl 7
V25
89.54
.396
.336
. -.374
Beta
Simple
R
R^
F-sig.
level
.24 .
-.01 .
.05
.06
.08
.11
.10
.10
.10
.17
.15
-.04
.03
.06
.13
.05
28
Dependent
I
Vars. V50
2
Subfiles:
3
Divided on:
Crop - I
Geog.loc.-4
(64)
Independent.
Vars: same as
Regression #15
29
Dependent
I
Var. V42
2
Subfiles:
3 .
Divided on:
Crop - I
Geog. Ioc. - 4 .
(64)
Independent
Vars: same as
Regression #15
/-
a
e
Continued.
Reg r e s s i o n No. 20 - N o . .31; M a x i m u m (V50) and A v e r a g e (V42)
Percent Response as the Dependent Variables; Data File Subdivided on Crop
and Geographic Location; I n d ependent V a r i a b l e s Restr i c t e d as in Table 11.
O
Regression Step
Variable
No..
into eq. .
Name
30
Dependent
I .
Va r . V50
2
Subfiles:
3
Divided on: 4
Crop - 3
5 ■
Geog. loc.-4
(13)
Independent
Vars: same as
Regression #15
31
Dependent
I
V a r . V42
2
Subfiles:
3
Divided o n : 4
Crop - 3
5
Geog. Ioc. -4
(13)
Independent
Vars: same as
Regression #15 '
.Variable
Number
Latitude
Aspect
Depth to Ca
' Tex. cls.
. Bulk Den.B
V33
V32
V29
V25
V20
Depth to Ga
Temp. Reg.
Latitude
' Dry Con.B
Water Reg.
V29
V35
V33
Vl 7
V41
Constant
. Beta
Simple
R
R
-403.72
.603
-.392
.555
.392
-.258
.63
-.52
.40
.06
-.11
.39
.64
.73
.80
.86
98.30
.181
-.954
-.014
.500
.845
.47
-.43
,37
.41
.03
.22
.40
.57
.63
.72
F-sig.
level
.025
.01
.01
.01
.01
—
.10
.05
•10
•10
72
only part of Montana that has soil, types with a mesic temperature
regime.
Depth to the ca (V29) was positively correlated to crop response
in these equations.
This indicated that as the horizon of calcium
accumulation approached the surface, there was a decreased response
to applied K.
Perhaps the greater calcium concentration associated
with a ca horizon was interfering in some way with K uptake.
The
literature supports this as a.possibility.
Elevation (V30) was a very important variable in this group of
regression equations.
It was positively correlated to crop response
in regression No. 26 through No.. 29, which deal with winter wheat
and barley and winter.wheat alone in geographic location 4.
Because
elevation did not enter regression No. 30 and No. 31 (barley alone),
this effect can be attributed to the winter wheat sites.
The
possible effects of elevation have already been discussed.
Textural class (V25) entered the regression equations five
times (regression No. 26 through No. 30).
No.consistent correlation
to crop response appeared to exist.
Dry consistence of the B horizon (V17) was positively correlated
to crop response in all cases in which it appeared in an equation;
it also appeared most often in the regression equations.
However,
dry consistence and bulk density of the Ap (V16 and V19) a n d .Cca
horizons (.Vl8 and V21) were most often negatively correlated to crop
73
response, Ttiiis negative correlation of the Ap horizon to crop
response could have involved the possibility that textural differ­
ences at the surface of a soil could have governed the effectiveness
of broadcast application of K;
Hard dry consistence and high bulk
densities may not have allowed penetration of surface-applied K.
Aspect (V32) also appeared in six of the regression equations
and was negatively correlated to crop response in all cases except
one.
The interpretation of the effect of aspect on crop response
was difficult because of the coding scheme devised.
The effect of
aspect was no doubt related to how it modified the effect of slope
and incidence of incoming radiation (soil temperature).
If this
was the relationship that existed, then one would have expected a
positive correlation between aspect and crop response if most of
the experimental sites pertaining to the regression equation had
northeast to southeast exposures,
One would have expected the
opposite correlation if most of the experimental sites had northwest
to southwest exposures.
This was due to the coding scheme employed
(see Figure I).
Regression No. 22 and No. 23 (spring wheat in geographic
'
•
V
location I) produced the most highly significant results (.005
level).
It is interesting to note that the equations have only
three variables in common and they appear in a different order.
This
points out that the nature of the dependent variable exerts some kind
74
of unexplainable influence on the factors which relate to the
variability in crop response.
Regressions No. 32 - No. 43:
Average Percent Response (V42) as the
Dependent Variable.
Some of the percent response (V42) data clearly indicated that
a negative response to added K occurred.
In some cases there was
clearly a strong positive response to applied K .
In order to
determine what, factors might be responsible for this variation in
response to added K, the data file for the present set of regressions
was divided riot only by crop type and geographic location but also
be response class (V42).
the following basis:
Three response classes were created on
Subfile I) 95 percent response and less =
negative response; Subfile 2) 96 - .105 percent response = no response;
Subfile 3) 106 percent response and greater = positive response.
Regression No. 41, which described the analysis of the spring
wheat experiments in geographic locations I, 2, and 3 in the "no
response" class, yielded the highest R
2
value (.93).
This was 2
2
significant at the .005 level.
Regression No. 33 produced an R
value of .20, also significant at the .005 level.
This latter
equation described the analysis for all crops in subfile 2 (no
response) over all of the geographic locations.
Response class 2 (no response) was described by regressions No.
33, No. 36, No. 39, and No. 41.
All of the regression equations
Table 13.
Regression
No.
Regres s i o n No. 32 - No. 43; A v e r a g e Percent R e s p o n s e (V42) as the Dependent
Variable; D a t a File Subdivided on Crop, Geogr a p h i c Location, and Percent
Response Class; Independent V a r i ables as in Table 11.
’ Step
into eq.
32
Dependent
I
Var. V42
2
Subfiles:
3
Divided on: ,4
Crop - 1,2, 3 l l _
Geog.Ioc-I,2,3,4
%—ras. — I
(45)
Independent
Vars: same as
Regression #15
33
Dependent
I
V a r . V42
2
Subfiles:
3
Divided on: 4
Crop, 1,2,3, 5
4. Geog.
Ioc. - I,2,3,4
% res.-2
(85)
Independent
Vars: Same as
Regression #15
Variable
Name
Depth to Ca
Latitude
Tex. Earn.
Aspect
Elevation
Variable
Number
V29
V33
V26
V32 .
V30 .
Constant
231.56
Beta
•.171
-.503
-.340
.240
-.234
Simple
R
R2
.24
-.16
-.13
.14
-.08
.06
.09
.15
.18
.22
■ F-sig.
level
—
.10
.10
.10
. ^i
""
Elevation
Slope
Dry Con.C .
Water Reg.
Tex. cls.
V30
V31
V18
V41
V25
97.05
.202
.239
.148
-.295
.253
.32
.23
.11
-.11
.11
.10
•13
.16
.18
.20
.005
.005
.005
.005
.005
Table 13, Continued.
R e g r e s s i o n No. 32 - No. 43; A v e r a g e Per c e n t R e sponse (V42) as the
D e p e ndent Variable: D a t a File Subdiv i d e d o n Crop, G eographic Location, and
Percent Respons e Class; Independent V a r i a b l e s as i n Table 11.
Regression
No.
34
Dependent
Va r . V42 '
Subfiles:
Divided on:
Crop-1,2,3
Geog.Loc-
Step
into eq.
I
2
3
4
5
6
Variable
Name
Tex. cls.
Tex. Fam.
Water Reg.
MAST
Dry Con.C
Bulk Den.C
Variable
Number
V25
V26
V41
V 34
V18
V21
Constant
.116;59
Beta
.536
-.493
.553
. T-.448
.278
-.198
2
Simple
R
R
.
.13
.06
.13
.02
.11
-.05
. .02
.04
.06
.10
.12
.15
F-sig.
level
---- — —
— — --------------
.10
.10
.05
I,2,3,4
% res.-3
(89)
Independent
Vars: Same as
Regression #15
35
Dependent
Var. V42
Subfiles:
Divided on
Crop-1
Geog.Loc.
I
2
3
4
5
I,2,3,4
%res. -I
(24)
Independent
Vars:Same as
Regression #15
'-j
ON
Elevation
Aspect
Bulk Den.A
Temp.Reg.
Water Reg.
V30
. V32
Vl 9
V35
V41
119.31
-.571
-.366
-.251
-.318
.230
. -.43
-.31
-.14
-.05
.09
. .19
.25
.35
.39
.43
.05
.05
.05
.05
.10
Table 13, Continued.
Reg r e s s i o n No, 32 - No. 43; A v e r a g e Per c e n t Res p o n s e (V42) as the
Dependent Variable; D a t a File Subdiv i d e d o n Crop, G eographic Location, and
Percent R e s p o n s e Class; Independent V a r i a b l e s as i n Table 11.
Regression Step
No.
into eq.
36 .
I
Dependent
Var. V42
2
Subfiles:
3
Divided on: 4
Crop - I
5
Geog.loc.-l,2,3,4
% res.-2
(57)
Independent
Vars: Same as
Regression #15
Variable
Name
Depth to Ca
Latitude
Water Reg.
Dry Con,C
Dry Con.A
37
Dependent
I
Water Reg.
Va r. V42
2
MAST
Subfiles
3
Dry Con.B
Divided on: 4
Bulk Den.A
Crop -I
5
Latitude
Geog. Ioc.1 ,2 ,3,4
% res.-3
(52)
Independent Vars:
Same as Regression #15
Variable
Number
V29
V33
V41
V18
V16
V41
V34
Vl 7
Vl 9
V33
Constant
Beta
Simple
R
R^
F-sig.
level
143.01
-.292
-.303
-.279
.240
-.187
-.23
-.22
-.15
.02
-.06
.05
.09
.13
.16
.19
.10
.10
.10
.10
.10
153.41
.581
-.514
.219
.201
-.201
.20
.07
.19
.19
-.15
.04
.08
.13
.17
. .19
— — — .
— — — '
.10
.10
.10
Table 13, Continued.
R e g r e s s i o n No. 32 - No. 43; A v e r a g e Per c e n t R e sponse (V42) as the
Dependent Variable; D a t a File S ubdivided on Crop, Geographic Location, and
Percent R e s pon s e Class; I n d ependent V a r i a b l e s as in Table 11.
Regression
No.
Step
Variable
into e q .
Name
38
Dry Con.B
Dependent
I
Var. V42
2
Tex. Fam.
Dry Con.A
Subfiles:
3
Dry Con.C
Divided on: 4
Crop-1
5
Tex.cls.
Geog.loc.-l
% res.-3
(18)
Independent Vars:
Same as Regression #15
39
Dependent
Latitude
I.
V a r . V42
2
Dry Con.A
Water Reg.
Subfiles:
3
Divided on: 4
Dry Con.C
Crop - I
Geog. loc.-4
A res.—2
(31)
Independent Vars:
Same as Regression #15
Variable
Number
Vl 7
V26
Vl 6
V18
V25
Constant
123.95
Beta
.534
.008
-.247
-.546
-.703
Simple
R
.45
-.40
-.08
-.13
-.32
F-sig.
level
.20
.30
.41
.47
.55
.10
.10
.10
.10
.10
oo
V33
Vl 6
V41
V18
253.04
-.615
-.432
-.334
-.255
-.27
-.18
-.14
-.03
.08
.16
.25
.30
—
.10
.05
.05
a
e 13, Continued.
Re g r e s s i o n No. 32 - No. .43; Ave r a g e Percent R e s p o n s e (V42) as the
Dependent Variable; Da t a File S ubdivided on Crop, G eographic Location, and
Percent R e s pon s e Class; Independent V a r i a b l e s as in Table 11.
Regression
No;
Step
Variable
into eq.
Name
40
Dependent
Dry Con.C
.I
Va r. V42
2
Latitude
Subfiles:
3
Dry Con.B
Divided on: 4 .
Elevation
Crop - I
5
Tex. cls.
Geog.loc.-4
. % res.-3
(27)
.Independ. Vars . r
Same as Regression #15
41
Dependent
I
Dry Con.C
Var. V42
2
Elevation
Subfiles:
3
Bulk Den.B
Divided on:
4
Bulk Den.C
Crop - 2
5
Water Reg.
Gepg.loc.-l,2,3
% res. - 2
(16)
Independent Vars.:
Same as Regression #15
n
Variable
Constant
Beta
■Number_____________;___________
V18
V33
Vl 7
V30 .
V25
V18
V30
V20
V21
V41
Simple
R ______
R
p-sig.
■
level
-972.77
.521
.888
1.599
1.728
-1.400,
.27
.15
.13
.04
.05
.07
.11
.16
.28
.48
—
—
—
--- —
.025
96.33
.649
.246
-.589
.319
.157
.78
.75
-.31
. .34
-.33
.61
.79
.89
.91
. .93
.005
. .005
.005
.005
.005
T able 13, C o n t i n u e d . R e g r e s s i o n No. 32 - No. 43; A v e r a g e P e r c e n t Re s p o n s e (V42) as the
Dependent Variable; D a t a File S ubdivided o n Crop, G eographic Location, and
P e rcent R e s pon s e Class; Independent V a r i a b l e s as in Table 11.
Regression Step
No.
into eq.
Variable
' Name
42
Dependent
I
Depth to Ca
Va r. V42
2
Elevation
Subfiles:
Aspect
3
Divided on:
Crop -3
Geog.loc.-l,3,4
% res.-3
(25)
Independ. Vars.:
Sames as Regression #15
43
Dependent
I
Thickness
V a r. V42
of B
Subfiles:
Divided on:
Crop - 3
Geog. loc.-l
% res. - 3
(16)
Independ. Vars.:
Same as Regression #15
Variable
Number
Constant
Beta
Simple
R
R
2
F-sig.
level
V29
V30
V32
99.33
-.482
.359
.307
-.51
.34
.22
.26
.34
.43
.01
.01
.01
V28
130.03
-.551
-.55
.30
-05
81
included dry consistence of
regime (V41).
the Cca horizon (V18) and moisture
Moisture regime was consistently negatively correlated
to crop response.
This suggests that added K may not produce as
great a response on soils that normally are of the drier regimes.
Dry consistence of the Cca horizon was positively.correlated to
crop response in three of the four regressions.
In regression No.
41., bulk density of the Cca horizon was also positively correlated
to crop response.
Perhaps this indicates that in soils with more
easily penetrable subsoils, a response to added K is not likely to
occur.
This may be due to the fact that native soil K was utilized
more readily.from the subsoil horizons.
Latitude (V33) was negatively correlated to crop response in
regression No. 36 and No. 39.
The reason that the ''no response"
class is negatively correlated to latitude is open for speculation.
It could be that the effect of latitude was temperature related.
At
any rate, it was interesting that latitude was more negatively
correlated in the Hno response1'
1
' class (regression No: 36 and No. 39)
and negative response class (regression No. 32) and positively
correlated in the positive response class (regression No. 40).
Another trend appeared with respect to the depth to the Cca
horizon CV29).
In the negative response class (regression No. 32),
V29- was strongly positively correlated to crop response.
However, in
the no response class (regression No. 36) and positive response class
82
(regression No.42), there was a strong negative correlation between
crop response and depth to the Cca.
This relationship suggested
that added K fertilizer was more likely to cause a positive response
under conditions of high C a C O ^ accumulation nearer to the soil
surface.
The positive response class was described by regression No, 34,
No. 37, No. 38, No, 40, No. 42, and No. 43.
In this, class, moisture
regime (V41) and MAST (V34) were both positively correlated to crop
response.
This again indicated that soils which tended to be warmer
and drier throughout the year showed a greater positive response to
added K.
1
.
■
Textural class (V25) and textural family (V26) appeared to be
inconsistently correlated to crop response between regression
equations in response class 3 (e.g. regression No. 34 vs. No. 38),
However, both variables were consistently correlated within a
regression equation.
This perhaps indicated that other factors were
responsible for modifying the effects of texture and that crop
response was dependent on these other factors as well,
Dry consis­
tence of the B horizon 0/17) was again positively correlated to crop
response in all three regression equations (No. 37, No. 38, and No.
40) in which it appeared.
The implications of this relationship
have already been discussed.
Regression No. 43 included only one significant variable (V28).
83
Thickness of the B horizon was strongly negatively correlated to
crop (barley) response, and this variable alone accounted for 30
percent of the observed variation in response.
Regressions. No. 44 - No. 47:
Maximum Response (V50) as the Dependent
Variable.*
2
The data file for regressions No. 44 through No. 47 was divided
according to the K rate at which 'fhe maximum crop response occurred.
This was done to investigate whether different factors were important
to response depending on rate of applied K.
The data subfiles were created as follows :
Subfile I) 11^39
kg K/ha; Subfile 2) 45-67 kg K/ha; Subfile 3) 81-96 kg K/ha; Subfile
4) 112-128 kg K/ha; and Subfile 5) 134 kg K/ha.
None of the. regres­
sion equations corresponded to subfile 2 (which included only those
experiments in which maximum percent response was attained at 45-67
kg K/ha applied K) since no significant results were obtained. .
The regression equation (No. 47) dealing with the highest rate
of added K produced the highest R
2
.value (.40).
The highest F-
significance level (.025) in this equation was actually achieved in
Step 4 (R
2
= .36). 'Regression No. 45 and No. 46, which dealt with
subfile 3 and subfile 4, respectively, as explained above, also were
significant at the ,025 level.
Textural class (V25) appeared in three of the four regression
Table 14.
R e g r e s s i o n No. 44 - No. 47; M a x i m u m P e r c e n t R e s p o n s e (V50) as the Dependent
Variable; Data F i l e Subdi v i d e d on K - r a t e at W h i c h M a x i m u m Re s p o n s e Occurred
(V51); Independent V a r i a b l e s as in T able 11.
Regression Step
No.
into eq.
44 '
Dependent
I
Va r . V50
2
Subfiles:
3
Divided on
V51 (SI)
4
cases: 43
5
Independent
Vars: V16 to
V21, V25, V26,
V28 to V 3 5 , V 4 1
45
Dependent
I
2
V a r . V50
Subfiles:
3
4
Divided on
5
V51 (S3)
cases: :41
Independent
Vars: V16 to
V21, V25, V26,
V28 to V 3 5 , V41
Variable
Name
Variable
Number
Elevation
Dry Con.C
Thickness
of B
Tex. cls.
Bulk Den.C.
V30
V18
Latitude
Tex.cls.
Aspect
Bulk Den.C
Dry Con.A
V33
V25
V32
V21
V16
V28
V25
V21
Constant
37.26
-323.36
R
2
Beta
Simple
F-sig.
level
.453
.115
.34
.05
.12
.18
.05
.025
-.354
-.274
.242
-.26
.06
.15
.21
.25
.27
.05
.05
.05
.631
.592
-.305
.312
.203
.31
.05
-.14
.12
.03
.10
.16
- .22
.28
.31
.05
.05
.05
.025
.025
Table 14, Continued.
R e g r e s s i o n No. 44 - No. 47; M a x i m u m P e r c e n t R e s p o n s e (V50) as the
Dependent Variable; Da t a F i l e S u b d i v i d e d on K-rate at W h i c h M a x i m u m R esponse
Occur r e d (V51); I n d ependent V a r i a b l e s as in Table 11.
Regression Step
No.
into eq.
46
Dependent
I
V a r . V50
2
Subfiles:
3
Divided on
4
V51 (S4)
5
cases: 42 ■
Independnet
Vars: V16 to
V21, V25, V26,
V28 to .V35, V41
47
Dependent
I
V a r . V50
2
Subfiles:
3
Divided on
4
V51.(S5) .
5
cases:
29
Independent
Vars: V16 to
V21, V25, V26,
V28 -to V35, V41
Variable
Name
Water Reg.
Temp. Reg.
Aspect
Tex. Fam.
Dry Con.C
Variable
Number
V41
V35
V32
V26
V18
Constant
101.50
2
Beta
Simple
R
R
.242
-.020
.208
.349
-.290
.38
.003
.19
.36
-.15
.15
.20
.23
.26
.30
F-slg.
level
.025
.025
.025
.025
.025
co
Ln
Latitude
Tex. cls.
Tex. fam.
Temp. Reg.
Slope
V33
V25
V26
V35
V31
-29.70
.270
1.282
-1.036
.224
.205
.27
.21
-.03
.14
.02
.
.07
.17
.33
.36
.40
—
.10
.025
.025
.05
86
equations and was most strongly positively correlated to crop
response at the highest rate of applied K.
Textural family (V26)
was also strongly positively correlated to crop response in regression
No. 46 which was also concerned with the higher rates of applied K.
This indicates that {the more finely textured soils show a
more
positive response to high rates of applied K „
Bulk density and dry consistence of the Cca horizon appeared in
regression No. 44, No. 45 and No. 46..
Both of these variables seemed
to be more positively correlated to crop response at the lower rates,
of applied K.
In regression No. 46 (higher K rate class), dry
consistence of the Cca was negatively correlated to response.
This
suggests that perhaps the lower rates of applied K have more of a
positive effect on response as the subsoil horizons become more
impenetrable.
Temperature regime (V35) and moisture regime (V41) appeared in
:
'
■
the regression equations (No. 46 and No. 47) at the higher rates of
applied K.
The general positive correlation of these variables to
crop response indicates that the higher rates of K are more effective
on the warmer and drier soil types. .
Aspect (V32) appeared in regression No. 45 and No. 46 (maximum
percent response as the dependent variable) as ah important variable.
It was negatively correlated to crop response in one case and
positively correlated in the other.
The problem, as discussed
87
earlier, with regard to the coding scheme for aspect, could explain
this phenomenon,
It should be pointed out that one class or rate of applied.
K did not dominate the determination of the maximum response
variable (V50).
This indicates that factors such as soil type and/or
soil conditions may dictate the most effective rate of K application.
Regression No, 48;
Average Percent Response (V42) as the Dependent
Variable.*
2
The fact that much of the growing season temperature data, rain~
fall data, and spring soil moisture data were missing, indicated
that statistical validity of previous results in which they were
included in the regression equation was questionable.
To determine
a more accurate interpretation of the importance of these variables
to crop response, a regression analysis option was used to include
only those cases which had a complete set of data for these variables.
Only growing season temperature data (V37, 738* V39), rainfall (V40),
and cumulative spring soil moisture to 122 cm. (V52), V53), 754,
755) were included as independent variables in the analyses.
The results showed that only average temperature in May (737)
and rainfall (740) entered the regression equation.
May by itself produced an R
2
correlated to crop response.
Temperature in
value of ,17 and was strongly negatively
This indicated that low soil temperatures
early in the growing season may have limited uptake of native K.
One
Table 15.
Regression No. 48; Average Percent Response (V42) as the Dependent Variable;
Data File-Restricted to Include Only Cases With Complete Data; Independent
Variables Restricted to Soil Temperature, Rainfall, and Cumulative Soil
Moisture Variables.
,
Regression
No.
I
2
Variable
Name
Temp.(May)
Rainfall
Variable
Number
V37
V40
Constant
126.37
. Beta
-.467
-.172
Simple
R
.
-.42
-.03
R
.17
.20
F-sig.
level
.01
.025
88
48
Dependent
Var. V42
Subfiles:
All;cases:
41 '
Independent
Vars: V37 to
V40, V52 to
V55
Step
into eq.
89
would probably have observed a positive response to added K if
this was the case,
These results were in accord with observations
discussed earlier by Boatwright et^ jal. (1976) .
Regressions No, 49
or No. 50;
Maximum (V50) and Average (V42).
Percent Response as the Dependent Variable.
The same regression analysis was performed.with a different
dependent variable in each case (V50 and V42, respectively).
set of independent variables was the same in both cases.
The
The
independent variable set included all of those variables used.in
regression No. 48 with the addition of those variables (V16 to V21,
i
V25, V26, and V28 to V35) used in the majority of the regression
analyses.
A more statistically valid regression analysis option
was again used to only include those cases for which there was a
complete data set for the variables included.
Dry consistence of the Cca horizon (V18),.temperature in May
(V37), and rainfall (V40) appeared in both regression equations.
Both equations were significant at the .05 level.
The R
2
values for
both equations were similar at .33 and .34, respectively.
All of the variables that entered the two equations which were
related to soil physical properties, dry consistence of the Cca
horizon, bulk density of the Ap horizon, textural family, and
thickness of the B horizon (V18, V 1 9 , V26, and V28, respectively)
were all positively correlated to percent response.
These results
Table 16 , Regression No. 49 and No. 50; Maximum (V50) and Average (V42) Percent Response
- as the Dependent Variables; Data File Restricted to Include Only Cases With
■ Complete Data; Independent Variables as in Table 11 and Table 15 Combined.
Regression
No.
■
Step
into eq.
49
Dependent
I
Var. V50
. 2
Subfiles:
3,
All^cases: 32 .4
Independent
Vars i-1 V16 to. 5
V21, V25, V26, V28 to V35,
V37 to V40, '
V41, V52 to
V55
.
■■ 50Dependent
Var. V42
Subfiles
All»cases: 32
Independent
Vars. - Vl6 to
V21, V25, V26,
V28 to V35, .
V37 to V40,
I
2
3
4 ’
5
Variable
■ Name
Variable
Number
Constant
Befa
Simple • R 2 . •
R
FXsig.
level
-Dry Con.C.
Bulk Den.A
Rainfall
Thickness
of B
Temp.(May)
■
V18
. Vl 9
■ V40 .
V28
V37 •
53.12
.317 . .
,37
.367
.11
-.05
-.315
. .278
-.240
.29
-.33
.13
.17 •
.23 .
.05
.10
,10
.29 ,
.33 .
.05
.05
.■ '
Dry Con.C
Tex.- Fam.
Temp.(May)
.) Rainfall
Soil H O
(0-90)
.Vl 8
V26
V37
V40
.596
-.380
102.36
-.223
, -.344
V54
.244
.
.
•41 .17
.11 . .22
-.41
.26
-.03
.30.
.12
.34
.025
.05
.05
.05
.05
V
0
o
91
suggest that as clay content Increases (associated with hard
consistence of the Cca, greater thickness of the B horizon, and a
positive correlation to textural family as coded coarse to fine),
a more positive response to added K results.
The negative correlation of temperature in May (V37) to percent
response was consistent with regression No. 48 and has been dis­
cussed.
The fact that cumulative soil moisture to 90 cm. (V54)
appeared in regression No. 50 indicated that soil moisture was
important with regard to crop response to added K.
The positive
correlation of this moisture variable to response suggests that
without adequate moisture diffusion of applied K to the plant root
will not occur.
Rainfall, another moisture variable, was not
clearly correlated to crop response; but its appearance in both
equations proved that it did account for some of the variability
in response to added K.
Regression No. 51;
MAST (V34) as the Dependent Variable.
Schaff (1979) determined that MAST was the most important
variable in relating to percent change in yield of winter wheat to
applied K fertilizer.
This factor alone accounted for 56 percent
of the observed variability in percent change in yield.
Because of the importance of this variable in Schaff's .regression
analyses, it was of interest to try to determine why MAST did not
Table 17.
Regression No. 51; MAST (V34) as the Dependent Variable; Data File
on Crop; Independent Variables Restricted as in Table 11.
Regression
No.
Step
into eq.
51 .
Dependent
Var.. V34
Subfiles:
Divided on:
Crop - I
Cases ;133
Independent
Vars. Same
as Regression
#15
I
2
3
4
5
Variable
Name
Water Reg.
Latitude
Tex. Fam
■Temp. Reg.
Dry Con. C
Variable
Constant
Number •
V41
V33
V26
V35
V18
15.56
Beta
.511
-.133
.299
.243
-.162
Simple
R
.82
-.51
.65
.47
.03
Subdivided
R2
F-sig.
level
.68
.73
'.75
.77
.79
.005
.005
.005
.005
.005
93
appear more often in the regression equations in this study.
Regression No, 51 was performed with MAST (V34) as the dependent
variable.
The independent variable set was that utilized in the .
majority of the regression analyses, in other words, that set which
excluded variables with a large amount of missing data.
MAST was highly positvely correlated with moisture regime (V41),
textural family (V26), and temperature regime (V35).
highly negatively correlated to latitude (V33).
It was also
The regression
equation showed that these four variables alone accounted for 77
percent of the observed variability in MAST (.005 significance,
level).
Because of this relationship, it was not surprising that
MAST did not appear more often in the regression equations.
Other
independent variables, particularly water regime (V41), probably
accounted for the same variability in crop response as MAST.
Regression Equations - Summary
Table 18 summarizes the correlations (Simple R) of each
independent variable used in the regression equations with the
dependent, variables,which are yield (V06 to VlO) and average (V42)
and maximum (V50) percent response.
Table 18 also summarizes the
relative overall importance of each independent variable, based on
the number of times that each of them enters a regression equation
and the frequency of entry (number of times included as an independent
variable divided by the number of times of entry into, the equation).
Table 18.
V a r ia b l e
Name
S ite
C ro p
Year
-K -ra te 2
K -ra te 3
K -ra te 4
K -ra te 5
D ry C o n . A
D ry C o n . B
D ry C o n . Cca
B u lk D e n . A
B u lk D e n . B .
B u lk D e n .Cca
S tru c . A
S tru c . B
S t r u c . Cca
T e x . e ls .
T e x . Fam.
Geog. L o c .
T h ic k n e s s B
D e p th t o Ca
E le v a t io n
S lo p e
A spect
L a t it u d e
MAST
Tem p. R eg im e
Tem p. (M a y)
Tem p. (J u n e )
Tem p. ( J u l y )
R a in fa ll
W a te r R egim e
S o il H O
(0 -3 0 ;
S o i l H2 O
(3 0 -6 0 )
S o i l H-O
(6 0 -9 0 )
S o i l H -o
,(9 0 -1 2 2 )
Overall Correlations of the Independent Variables with the
Dependent Variables - Yield (V06 - V10) and Maximum (V50) .
and Average (V42) Percent Yield Response.
S im p le R
(+ )
-
V a r ia b l e
N um ber •
VOl '
V03
V04
V12
V13
V14
V15
V16 ■
Vl7
V18
V19 V20
V21
V22 .
V 23 ■
V24
V25
V26
. V27
V28
V29
V 30
V31
V 32
V33
V34 V35
V37
V38
V39
V40 ’
V41
•(-)-
•
.
I
9
'
4
2
4 '
4
15
7 '
6
• 7.
6
3
■
3
4
2
6 •
■
11
N o. o f R e g r e s s io n s i
in c lu d e d a s a
V a r ia b l e
9
3
3
2
'.■2.
i
2
2
3
2
4
6
16
10
4
■ 4
5
I
4
S im p le R
.
8
I •
5
8
3
,
4
2
6
3
3
5
5
■4
2
8
10
3
6
2
I
4
4
5
9
7
7
7
7
.
50
50
50
50
50
50
15
15
15
50
50
15
50
50
50
' 50
50
50 '
49
.50
18
18
18
18
• 50
N o . T im e s V a r ia b l e
E n te r e d
.
R e g r e s s io n ..
' 4 .
3
4
'■
4
.
3 .
2
4
14
17
15
12
' 7
9 .
I
'
6
I
15
7
5
9
9
19
.9
14
17
■ 6.
6
9
6
3
10
15
E n t r y F re q u e n c y
i n t o R e g r e s s io n
E q u a t io n . (%)
•
44 '
60 .
44
57
43
29
57
28
34
30
24
14 '
18
7 ■
40
7
30
14
33
18
18
38
18
28
34
12
12
50
• 33
17
56
30
V44
3
I,
15
4
27
V45 •
I
2
15
3
, 20
V46
I
15
I
V47
2
.4
15
6
I.
■■2
15
3
■ 20
4
4
. . 2
.1 5
• 4
6
4
40
100
3
I
33
S o i l H-O
(1 2 2 -1 5 2 )
V48
S o il H O
(1 5 2 -1 8 3 )
V49
K -ra te o f
V51
M a x . R e sp o n se
S o i l H-O
(0 -9 0 )
• V54
'
I'
■
7
\
The' variables entering the regression equations most often were
elevation (V30 - 19 times), latitude (V33) and dry consistence of the
B horizon (V17
17 times), dry consistence of the Cca horizon (V18),
textural class (V25), and dry consistence of the Ap horizon (V16)
and aspect (V32 - 14 times).
,
The most consistently.correlated
variables in the regression equations were the K-4ate variables
(V13, V14, V15, V5l), MAST (V34), dry consistence of the B horizon
.
(V17), elevation (V30),, slope (V31), and moisture regime (V41).
The. K-rate variables were consistently positively correlated to
crop response, indicating that applied K generally caused increased
.yields.
MAST and moisture regime were also consistently positively
correlated to crop response, suggesting that, applied K was generally
more effective on soil types associated.with the warmer and drier
sites.
Slope was also consistently positively correlated to crop
response,
This relationship may be related to moisture availability,
(drier sites) or to a slope-aspect interaction in which the incidence
of incoming radiation may have affected soil temperature trends.
Overall, dry consistence and. bulk density throughout the soil
profile seemed to be very important with regard to crop response to
added K,
Dry consistence of the B horizon (V17) was the most
important factor relating to crop response and was also the most
96 .
consistently correlated variable (Simple R) to crop response.
Dry
consistence of the B and Cca horizons are generally positively .
correlated to crop response whereas dry consistence and bulk density
of the A horizon appear slightly negatively correlated to crop
response.
.
If one assumes that increased dry consistence values can be
associated with greater clay contents and decreased ease of K
diffusion, root penetration, and exploitation of native soil K
reserves, then the positive correlation between crop response and
the harder subsoil horizons seems reasonable.
The fact that bulk
density and dry consistence of the Ap horizon appeared more
negatively correlated to crop response perhaps suggests that the
effectiveness of applied K (usually broadcast) was limited by
compacted surface horizons which would have retarded diffusion of
applied K into the soil profile.
Table 19 summarizes the responses to applied K by crop and
geographic location for all of the experiments included in this
study.
Although oats showed the highest percentage of experiments
with a positive response to applied K, there were only four
experiments which used pats as the experimental crop.
The other
three crops are similar in the percentage of experiments of each
which responded positively to applied R,
Winter wheat experiments in geographic location 4 had the
Table 19.
Crop
Winter
Wheat
Spring
Wheat
Barley
Oats ■
Sunnnary of Small Grain Response to Applied K Fertilizer; Montana Statewide
Study, 1968-1980.
No.
of
Experiments
133
36
49 .
4
No. with
Positive
K-Response
90
22
31
4
%
Positive
Response
68
61
63
100
No. with
Negative
K-Response
43
14
18
0
■ %
Negative
Response
32
39
37
0
Range of
% K
Response
81-122
85-121 •
88-117
83-153
Geo. Average
Lo g .
% K (V27) Response
I
2
■3
4
.102
101
.101 .
106
80-115
HO
96-120
I
100
2
HO
106
61-146
I
3
4
106
2
90-118
89-118
102-109
108-110
3
4
104
104
’I
106
109
2
3
VO
98
Table 20.
Number of Experiments at the Various Rates o f Applied K
at Which Maximum Percent K Response Occurred.
Rate at Which
Maximum K-^Response
Occurred (V51)
Number o f ■
Experiments
-■ kg K/ha -
11
3
13
2
22
12
27
2
28
22
■ 34
I
39
1
45
37
48
11
54
2
56
11
57
4
67
2
81
2
90
32
96
7
112
32
115
9
128
I
134
29
222
99
highest average percent K response (106 percent) for that crop.
Spring wheat showed a higher average percent response in geographic
location 3 (106 percent) than in geographic location I (100 percent).
Barley showed a.greater average percent response in geographic
locatipn I (106 percent) yet also showed a much greater variability
in response in this area (61 - 146 percent).
Table 20 shows the number of experiments which produced maximum
percent K response at each rate of applied K.
It was interesting to
note that in about half of the experiments the crops responsed to
their maximum at applied K rate of 90 kg K/ha and above.
This is
conclusive evidence that in many cases availability of native soil
K is not adequate for the nutrition of small grains. It further points
odt that crop responses to added K fertilizers may not be measured
in many field experiments due to inadequate rates of application
in the experiments.
Chapter 5
SUMMARY AND CONCLUSIONS
From 1968 to 1980, numerous small grain soil fertility experiments
in which applied K fertilizer treatments were involved were establish­
ed around the State of Montana.
Two hundred twenty-two of these
experiments established on 127 different site locations were selected
to study the influence of soil profile and site characteristics and
soil classification parameters on yield and crop response to applied
K fertilizer.
From two to five rates of K, ranging from 0 (control),
to 134 Kg K/ha, were applied in the experiments studied.
Sites were classified as to their soil series using the legal
description of the site along with the appropriate SCS county soil
survey, or by using the series name associated with an experiment
as reported in the Montana Agricultural Experiment Station Annual
Report.
Eight variables (classification parameters) were included
in the regression analyses as determined from the SCS soil series
description.
Soil samples from each of three horizons identified as the A p ,
B, and Cca horizons were taken at each of the 127 sites.
Dry
consistence and bulk density measurements were obtained for each
soil horizon sample.
Site characteristics including slope, aspect,
elevation and latitude were also recorded and introduced as variables
in the regression analyses.
Spring soil moisture data, rainfall data, and soil temperature
101
data were obtained from the Montana Agricultural Experiment Station
Annual Report and introduced as variables in the regression analyses.
However, many of the experiments did not include soil moisture and
temperature data..
A total of 48 independent variables iri varying combinations
were inserted into multiple linear stepwise regression programs with
actual crop yield dr percent change in yield as the dependent
variable..
Percent change in yield was determined both as an
average and maximum percent figure.
Over 60 regression analyses were performed with varying combin- ■
ations of independent and dependent variables.
The data file was
divided in many cases so that the regression equations usually only
pertained to a selected part of the data file.
subdivided according to:
The data file was
I) crop (winter wheat, spring wheat, barley,
and oats ) ; 2) crop and four geographic locations (see Figure 2);
3) .crop, geographic location and percent response class (see Results
and Discussion section - page 74 under Regressions No. 32 - No. 43);
and 4) K rate at which maximum response occurred (V51).
Fifty-one of
these regression equations produced significant results ranging from
.10 to .005 significance level.
These.regression equations are
presented in the Results and Discussion Section of this thesis in
Table No. 8 through Table No. 17.
Regression equations No. I through No. 5, which utilized actual
102
crop yield as the dependent variable and included ten variables each,
all produced R
2
values significant at the ,005 level.
Regression equations No. 6 through No. 10 (average percent
response as the dependent variable) and regression equations No. 11
through No. 15 (maximum percent response as the dependent variable)
also included ten variables each; and the regression equations also
all produced R
2
values significant at the .005 level.
The significance
of these results is probably inflated due to the inclusion of spring
soil moisture, rainfall, and soil temperature variables which had
extensive amounts of missing data.
Regression No. 4 and No. 5
(actual crop yield as the dependent variable) are of special interest,
though, in that the first two variables into the equation (K-treatment
and bulk density of the Ap horizon) by themselves produced R
2
values
of ,76 and .6 8 , respectively.
Regression No. 15 through No. 17 are similar in that the same
restricted set of independent variables was used in these analyses.
The set of independent variables was restricted in that soil moisture
(V43 - V49 and V52 ^ V55), soil temperature (V36 - V39) and rainfall
(V40) variables were removed from the analyses due to the large
amount of missing data for these variables.
The dependent variables
used in these analyses were either the average (V42) or maximum (V50)
percent response,
The most significant results (,005 level) were
produced by regression equations:
No. 22, which measured the
103
relationship of maximum percent response of spring wheat in
geographic location 1—
No. 23, which measured the relationship
of average percent response of spring wheat in geographic location
I; No. 33, which measured the relationship of average percent
response in experiments which included all four crops over all four
geographic locations and that had average percent response values
between 96 and 105; and No. 41, which measured the relationship of
average percent response in experiments which dealt with winter
wheat in geographic locations I, 2, and 3 that had average percent
response values between 96 and 105.
These regression equations
2.
produced R
values of .79, .60, .20, and .93, respectively, with
five variables included in each equation.
The only variable to
appear in all of these equations was dry consistence of the Cca
horizon (VIS).
Regression No. 49 and No. 50 (V50 and V42 as the dependent
variable, respectively) were produced with a restricted regression
program option in which only those cases (experiments) with a
complete data set were included in the analysis.
The independent
variable set was again expanded to include soil temperature, rain-fall, and cumulative spring soil moisture variables,
equations produced R
— See Figure 2.
2
These regression
values of ,33 and .34 respectively (.05
V
104
significance level).
Dry consistence of the Cca horizon (V18),
rainfall (V40), and average soil temperature in May (V37) were the
only variables to appear in both regression equations.
Elevation (V30) entered the regression equations more than any
other variable and was most often positively correlated to crop
response to applied K.-
The effect of elevation on crop response may
be related to its influence on climatic factors such as moisture and
temperature, but ascertaining its importance at this point is merely
speculative.
Dry consistence of the B horizon (VI7) is consistently positively
correlated to crop response to added K.
This relationship may have
to do with the decreased ability of roots to penetrate into subsoil
horizons with a hard dry consistence or for native soil K to readily
diffuse to crop roots.
If this is so, then the plant's ability to
utilize native soil K would be limited and a response to added K
would be likely.
Consistence is no doubt a measurement that integrates the
effects of other factors such as clay content, clay type, and bulk
density.
It will be important in the future to quantitatively
determine exactly how these various factors interact to influence the
consistence measurement.
Schaff .(.1979) determined that clay content of the ca horizon
was one of the five most important factors influencing winter wheat
105
response to added K.
The importance of the dry consistence of the
Cca horizon (VlS) in the present study corroborates his observation
and gives credence to the idea that clay content and consistence are
directly related to one another.
The relationship of soil structure to crop response to added K
is not clearly defined in this study.
However, the structure of
the B horizon (V23) appears to be very important with regard to
yield variability.
In regression No. I, No. 2 and No. 3, it is
strongly positively correlated to actual crop yield, which was the
dependent variable in these regression equations.
Because of the coding scheme used for the soil structure
variables, it was difficult to ascertain the exact effect of soil
structure on yield; but the indication was that greater yields were
associated with the stronger, more coarsely structured soils.
Perhaps
this effect was related to better root penetration and exploitation
of the soil environment along the planes of separation between
adjacent peds associated with well structured soils.
Other soil classification parameters accounted for much of the
variability in crop response to added K.
Moisture regime and mean
annual soil temperature were two classification parameters that.were
consistently positively correlated to crop response.
That is, the
warmer and drier soils were associated with greater response to
added K,
Temperature regime was also an important soil classification
106
parameter that appeared in the regression equations, but it was not
as consistently correlated to crop response.
Textural class and textural family both were significantly
related to crop response to applied K in a number of the regression
equations.
The general trend was for the fine textured soils to
respond to a greater degree than the more coarse textured soils.
This was consistent with most o f .the literature which indicated that
as clay content of the soil increased, more total soil K was
required for adequate supply to the plant to be maintained.
The above soil classification parameters may be easily determined
in the field or from a soil series description.
The indication was
that all or some of these parameters could be incorporated into a
reliable K soil test procedure.
The K rate variables were the most consistently correlated
variables to crop response.
The high positive correlation of K
rate to crop response indicated that, in many soils, adequate
availability of native soil K did not exist.
However, approximately
one-third of the total number of experiments in this study either
showed no response or a negative response to applied K.
cases the negative response was pronounced;
In some
This suggested that
for a given set of existing soil conditions, there was an optimum
rate of K which should be applied to achieve maximum crop response.
Tf the availability of native soil K was adequate, it is possible
107
that added K may have interfered with the uptake of some other nutrient.
It is important to know when arid where this type of adverse effect is
likely to occur.
Schaff (1979) found a positive correlation between the con-
++
centration of extractable Ca
percent yield response.
in both the 1st and 2nd horizons and
It was the most highly related chemical factor
to percent yield response of winter wheat to K fertilizer.
In the
present study, depth to the Cca horizon (V29) was an important factor
relating to crop response to applied K.
It is possible that the
calcium concentration in the soil solution may have affected diffusion
of soil K by influencing the ionic concentration of K.
A high calcium concentration may have interfered with other pri­
mary nutrients also.
Phosphorus, for example is considered to be more
limiting for small grains than potassium.
Calcium is known to convert
phosphorus into forms unavailable to plants and, thus, could indirectly
influence plant K needs as well.
This indirect influence may be related
to decreased demand and uptake of K with decreased phosphorus avail­
ability.
Rainfall during the growing season was also important with
regard to crop response to applied K.
Rainfall was not consistently
correlated to yield response because its influence was probably
modified by other existing soil conditions, such as amount of
spring soil moisture.
The time and distribution of rainfall was also
108
probably important and could explain variations in response to
applied K.
These considerations were not part of this study.
Soil temperature and moisture were important with regard to crop
response to added K.
Temperature in May (V37) was most often negative­
ly correlated to crop response, suggesting that low spring temperatures
controlled and probably retarded diffusion of native soil K to the
plant root.
Added K would have increased the soil solution K con­
centration and increased K diffusion under these circumstances.
Rather than soil moisture or temperature being the primary
influence on crop response by itself, it was probably an inter­
action of these two factors which partially governed the effect of
added K.
An abundance of soil moisture would tend to prolong low
spring soil temperatures by increasing the specific heat of the
soil.
On the other hand, the absence of adequate soil water would
allow the positive effect of temperature on diffusion to be realized
yet may have adversely affected the K concentration in the soil
solution and limited K diffusion.
This study has shown the importance of soil temperature and
moisture upon crop response to added K.
However, significance of
these relationships were limited by the fact that many of the
experiments did not include soil moisture, rainfall, and soil
temperature data collection.
All future experiments should include
the collection of these data as part of the standard experimental
109
procedure.
These factors are critical to dryland agriculture in
Montana,and indications are that these factors highly influence crop
response to added K fertilizer.
The effects of soil moisture and temperature on crop response
will undoubtedly also be influenced by soil texture and other
soil physical properties.
Clay content, clay type, and bulk density
are known to influence ion diffusion in soils.
It is the interaction
of all of these factors which will ultimately determine crop response
to added K.
It is essential that future studies be directed toward
quantitatively determining the nature of these interrelationships.
The relative importance of each of these factors
will vary, and each
of them could probably be considered to be limiting K diffusion and
crop response depending on a given set of soil conditions.
The
regression equations substantiate this conclusion in that the
correlation (simple R) between crop response and these climatic and
physical factors varied from one regression to another in many cases.
Pt is important to realize that this study was limited in
that many factors which could affect crop response to added K were
not considered or included as variables.
Management techniques,
for example, may have a direct influence on crop response.
The fact
that bulk density of the Ap horizon was an important variable in
many of the regression equations suggested that compaction problems
HO
on some soils may have existed as a result of time, amount, and kind
of cultivation employed.
Also; time of fertilizer application
(spring vs. fall) and seeding date are two important variables which ■
were not considered in this study.
..
.1.
Raychaudhuri (1976) found that in high potassium-fixing soils,
recovery of applied K by the crop was appreciably lower from broad­
cast than from band application.
Application of K in two or three
split doses was found to. be superior to a single basal dressing,
depending on the soil texture and predominant clay type.
Method and
time of K fertiliser application, as well as type of K fertilizer
applied (e.g. KCl vs. KgSO^), were variables not included in this,
study. •
+
Kowalenko and Rpss (1980) observed that the presence of K
4*
depresses
'
.
fixation, and that fixed ammonium established some
+
kind of equilibrium with exchangeable NH^ over time.
4“
4“
It. is known
that NH^ and K
act similarly in the soil environment under certain
circumstances.
This is due to the fact that the ions are of equal
valence and approximate ion size.
The type and amount of N ferti­
lizer was variable from one experiment to another in this study,
and the influence of this difference was unknown.
McLean (1976) made some interesting observations with regard to
K fixation and release;
Potassium release, which became progressively
smaller in magnitude with time, was eventually exceeded greatly by
Ill
the capacity of the clay to fix added K in non-exchangeable form.
He believed both processes occurred simultaneously, and it was not
always known which predominated for a 'given soil.
Therefore, there
was always uncertainty as to how to take them into account when making
a K recommendation based on soil tests.
He concluded that it was
particularly the fine textured alluvial or lakebed soils with a
predominance of micaceous illitic clay that varied the most in K
release and fixation, depending on the degree of original weathering
and recent cropping conditions.
The weathering of clay particles probably varies from year to
year in Montana in that soil moisture, snow cover, and the depth to
which a soil freezes vary.
Temperature effects have been shown to
be important with regard to K fixation and release.
A better under­
standing of these mechanisms will enhance the predictability of crop
response to added K.
Willis and Power (1975) point out that soil temperature has a
dominant influence on plant growth, both directly and indirectly.
Whether temperature has a direct or indirect effect is, however, a
moot question, particularly because we are primarily interested in
some yield function; and the plant acts as an integrator of many
individual factors.
Diffusion is the primary mechanism by which K reaches the plant
root.
Diffusion of K in the soil environment is no doubt governed
.
112
by the interaction of many important factors.
This study has
established that known or easily determined soil and site character­
istics are significantly correlated to crop response to K fertilizer
and do indeed influence the process of diffusion.
The results of this study indicate that many of the significant
variables could be incorporated for use in a K soil test.
The
parameters which accounted for much of the variation in crop
response to applied K fertilizer can be easily determined in the
field, from a soil series description, or from a soil sample taken
for limited laboratory analysis.
Moisture content and soil temperature are known to influence
diffusion.
The way in which site characteristics (e.g. slope and
aspect) and soil physical properties, such as bulk density, con­
sistence, structure, and texture, modify the influence of moisture
and temperature will allow for better understanding the soil
environment in which K diffusion must occur.
APPENDICES
APPENDIX I
115
Site Numbers, Cooperators, County, Legal Descriptions, and Soil Series
for the 127 Small Grain Sites, 1968-1980.
Site
No. Fertility
Experiments
Legal Description
Cooperator
County
Stevenson
Berkrum
Benge
Kelly
Rowland
Hill
Glacier
Powder River
Bighorn
Carbon
I
I
I
2
2
S * . Sec. 12, T MN , RlOE
SEll of SW,, Sec. 20, T M N , R5W
NVfI,, Sec. 7, TlS, R50E
SVR, of SMI, Sec. 26, TlS, R M E
NVf,, Sec. 10, T4S, R22E
Absorokee
6
7
8
9
10
Wieler
Bates
Daum
Toreke, W.
Erickson, L.
Stillwater
Gallatin
Yellowstone
Bighorn
Bighorn
I
I
I
2
2
NEl, of SVRi, Sec. 18, TIN, R20E
SEl,. Sec. 32, TIN, RlE
NVf, of NEtl1 Sec. 7, T4S, R26E
SEI,, Sec. 7, T2S, RllE
SE!, of NVf,. Sec. 11, T4N, R23E
Amsterdam
Shaak
Gilt Edge
Bainville
11
12
13
14
15
Cooper
Franeen Bros.
Lakey
Filing
Relnoweki
Gallatin
Liberty
Hill
Hill
I
I
I
8
2
SVf,, Sec. 27, TIN, RlE
NVf, of NEl,, Sec. 24.T32N, RlM
SE), of NE!,, Sec. 10, T U N , R5E
SMliof NVf,, Sec. 5. T32N, R9E
NEl, of NVf,, Sec. 4, T U N , R U E
16
17
18
19
20
Warnick
Gregoire
Rolston
Redekopp
Coulter
Hill
Hill
Hill
Valley
Garfield
I
I
6
2
3
SE), of SE),. Sec. 19, TU N , R U E
NVf,, Sec. 18, T U N . RISE
NVf,, Sec. 21, T32N, R U E
SMI,, Sec. 30, T U N , R45E
SVf,, Sec. 4. T19N, R M E
Williams
Telstad
Dooley
Cherry
21
22
23
24
25
Erickson, K.
Nissley
Fadhl
Halside
Obergfell
McCone
Dawson
Rosebud
Roosevelt
Richland
2
I
2
2
2
NVf,
SE),
NVf,
sEl,
NE!,
Vida
Farnuf
Ghamn
Williams
Chama
26
27
28
29
30
Mocassin Sta.
Metcalf
Nemec
Lee
Bates
Judith Basin 11
Judith Basin 3
Fergus
I
Judith Basin I
Gallatin
I
SVf1 of S E V Sec. 16, TU N , R14E
SE!, of SVf,, Sec. 14. T16N, R14E
NVf,, Sec. 6, T18N, R U E
SEl, of SM!,, Sec. 7. T17N, R U E
S E V Sec. 32, TIN, RlE
Judith
Judith
Danvers
Winifred
Amsterdam
31
32
33
34
35
Toreke
Wieler
Daum
Benge
Erpelding
Bighorn
Stillwater
Yellowstone
Powder River
Rosebud
I
I
I
I
I
SVfl,
SVf,.
SVf,,
SHI,,
SEV
Absarokee
Farland
Fort Collins
36
37
38
39
40
Dyk
Patterson
Holland
Holland
Miklovich
Gallatin
Stillwater
Rosebud
Rosebud
Bighorn
2
I
3
I
I
NW*, Sec. 6, TlS, R3E
SW* of SW*, Sec. 23, T4S, RlOE
NElllSec. 20, T6N, R42E
N E V Sec. 20. T6N, R32E
NEl,. Sec. 33. TSS, RISE
Manhattan .
Absarokee
41
42
43
44
45
Torske, L.
Dyk
Herzog
Erpelding
Brinkman
Bighorn
Gallatin
Stillwater
Rosebud
Fergus
I
I
I
I
I
SHV
SHV
NVf1,
SEV
NWV
Gilt Edge
Manhattan
Tanna
Fort Collins
Danvers
46
47
48
49
50
Boling
Cash
Daura
Pidwerbeckl
Redekopp
Fergus
Musselshell
Yellowstone
Valley
Valley
2
2
I
I
I
SW*, Sec. 33, TlSN, R14E
NHl,. Sec. 30, TSN, R25E
SVf,. Sec. 18, T4S, R26E
SW*, Sec. 14, T27N, R40E
Nlj of NVf,, Sec. 32. T U N , R45E
Coffee Creek
Bainville
Absarokee
Evanston
Martinsdale
51
52
53
54
55
Erpelding
Erpelding
Helm
Hauks
Perry
Rosebud
Rosebud
Garfield
Gallatin
Choteau
I
I
I
I
I
S E V Sec. 15, T6N, R41E
NEk. Sec. 15, T6N, R41E
NEl,, Sec. 32, TlTN, R43E
SE), of SHl,, sec. 32, T2S, RSE
SE), of S E V Sec. 31, T23N, RlOE
Fort Collins
Fort Collins
Cherry
Amsterdam
Gerber
56
57
58
59
60
Works
Miklovich
Pehl
Schaff
Sime, Inc.
Choteau
Bighorn
Prairie
Golden Valley
Gallatin
2
I
3
I
I
SVfi of SEl,. Sec. 24, T28N, R9E
NVf, of SVf,, Sec. 27, T6S, R36E
SE), of SVf,. Sec. 20, T12N. R52E
E*! of NE V Sec. 6, T5N, R22E
SVfl. Sec. 31, T2S, RSE
Evanston
Savage
Chauta
Wormser
Bozemae
I
2
3
4
5
of
of
of
of
of
NVfll
SEV
NVf,,
SEV
SHV
Sec.
Sec.
Sec.
Sec.
Sec.
Sec.
Sec.
Sec.
Sec.
Sec.
Sec.
Sec.
Sec.
Sec.
Sec.
I, T24N, R49E
12, T19N, RSlE
28, T12N, R44E
19, T U N . R55E
13. T23N, R57E
7, T2S, R U E
18, TIN, R20I
18, T4S, R26E
6, TlS, RSOE
16, T6N, R41E
7, T2S, R U E
7, TlS, R3E
21, TIN, R20E
15, T6N, R41E
31. T19N, R U E
Soil Series
Kevin
Vona
Unanmed loam
Gilt Edge
Edgar
Richfield
116
Site Numbers, Cooperators, County, Legal Descriptions, and Soil Series
for the 127 Small Grain Sites, 1968-1980.
61
62
63
64
65
Lasella
Visaer
Works
Dralne
Holland
Cascade
Madison
Choteau
66
67
68
69
70
Dahlman
Koch-Pox
Todd
Kronebusch
Kronebusch
71
72
73
74
75
Rosebud
I
2
I
I
I
Still. Sec. 33, 12IN, RSE
NE**, Sec. 9, T3S, RlW
SE*!, Sec. 24. T28N, R9E
NE1I1 Sec. 8, T7S, RSSE
SVA of NWS, Sec. 21, T6N, R43E
GcirKnr
Evanston
Evanston
Kremlin
Ploweree
Rosebud
Gallatin
Gallatin
Pondera
Pondera
I
I
I
I
I
NWS, Sec. 13, T6N, R41E
S * of NEll. Sec. 22, T2S. R6E
NWiI , Sec. 21, T2S, R4E
NWiI, Sec. 8. T29N, R2U
SEiI of SEiI, Sec. 16, TION1 R3W
Unwrn
Bozeman
Amerstdam
Kevin
Kevin
DeStaffany
DeStaffany
Hufflne
Kell
Gettel
Pondera
Pondera
Gallatin
Pondera
Cascade
I
I
I
I
2
StiiI, Sec. 29, T28N, Rlti
ES of SES. Sec. 30, T28N, Rlti
NES, Sec. 7, T2N, RSE
StiS. Sec. 23. T30N, R2W
SES, Sec. 8 T22N, RlE
Scobey
Amsterdam
76
77
78
79
80
Sommerfeldt
Swlnland
Swinland
Goldenstein
Huffine
Cascade
Park
Park
Gallat in
Gallatin
I
I
I
I
I
NES. Sec. 17, T22N, RlE
StiS. Sec. 17, TSN, R9E
SES, Sec. 27, T SN. R9E
StiS of SES, Sec. 27, T2S, RSE
NWS, Sec. 7, T2N.RSE
81
82
83
84
85
Teymoes
McOmber
pPsrson Bros.
Barber, L.
Martens
Teton
Teton
I
I
I
I
3
NES, Sec. 25, T23N, R3W
SUS, Sec. 14, T21N, RSti
NWS, Sec. 32, T21N, Rlti
NEH, Sec. 18, T19N, R14K
NtiS of SVA, Sec. 33, T28N, R12E
86
87
88
89
90
Jurenka
Hill
Bltz
Hill
Spicher
Hill
Res.Site Hingham Hill
Res.Site Rudyard Hill
I
5
I
2
3
SES, Sec. 27, T35N, R9E
StiS of StiS, Sec. 10, TlON, Rl2E
NtiS of NES, Sec. 12, TI2N. RlOE
SES of SES. Sec. 6, T32N, RlOE
NWS of NWS, Sec. 27, T2SN, R9E
Assinniboine
91
92
91
94
95
Vermulum
Vermulum
Johnson
Lakey
Kaimnerzell
Glacier
Glacier
Glacier
Liberty
Liberty
?
5
3
2
I
NES of SES, Sec. I,
StiS of NWS, Sec. 6,
StiS of NWS,Sec. 26,
SES, Sec. 35, T13N,
SES of NES, Sec. 5,
Pendroy
Pendroy
Brockway
96
97
98
99
100
Cady
Kaercher
Donovan
Tviet
Chrlstofferson
Liberty
Hill
Hill
Richland
Roosevelt
2
10
3
2
2
NtiS
StiS
NtiS
StiS
NWS
of
of
of
of
of
StiS. Sec. 26, T34N, R7E
StiS, Sec. 2, T12N, R14F.
SES. Sec. 13, T34N, R13E
SES.Gec. 25, T25N, R57E
StiS, Sec. 17, T29N, R56E
Vida
Williams
101
102
103
104
105
Iloye
Benson
Waters
Holland
Hansen
Roosevelt
Roosevelt
Roosevelt
Rosebud
Sweetgrass
2
4
2
I
4
SES
NES
SES
NES
StiS
of
of
ol
of
of
SlA,
SES,
StiS1
NES,
SES,
Sec.
Sec.
Sec.
Sec.
Sec.
12,
22,
16,
19.
36,
Williams
Parshall
Williams
Floweree
Floweree
106
107
108
109
HO
Mosdal
Larsen
Herzog
McFarland
Stillwater
Rosebud
Bighorn
Stillwater
Yellowstone
2
I
I
I
I
NES
SES
NtiS
StiS
StiS
of
of
of
of
of
StiS,
SES,
StiS,
StiS,
SES,
Sec.
Sec.
Sec.
Sec.
Sec.
12, TIN, R22E
24, T6N, R41E
7, T2S, R34E
21, TIN, R20E
10, TIN, R24E
111
112
113
114
115
Becker
Warren
Sire
Larsen
Eastlick
Yellowstone
Bighorn
Yellowstone
Rosebud
Stillwater
I
I
2
I
1
NES of StiS. Sec. 26, T2S, R26E
SES, Sec. 18, T2S, R34E
StiS. Sec. 22. TIN, R29E
SES of NWS, Sec. 24, T6N, R41E
NWS of NWS, Sec. 11, TIN, R22E
Absorokee
Richfield
Danvers
116
117
118
119
120
Lee
Larsen
Keller
Logan
Warren
Yellowstone
Rosebud
Yellowstone
Yellowstone
Bighorn
I
2
I
I
3
NWS of NWS, Sec. 3, T IS. R26E
StiS of SES, Sec. 24, T6N, R4I#
NWS, Sec. 29, T4N, R32E
SWt. Sec. 36, TIN, R28E
StiS1 Sec. IR, T25, R34E
Absarokee
Degrand
Kobar
Shaak
Richfield
121
122
123
124
125
126
127
Haines
Keller
2
I
I
2
I
I
i ____
- 222
StiS of NWS, Sec. 16, TON, R38E
SUS, Sec. 21. T4N, R32E
NES of NWS, Sec. 36, TIN, R38E
NWS of SVA, Sec. 24, T6N, R21E
SVA of NWS, Sec. 15, T2N, R28E
NES of NES, Sec. 30, T2N, R28E
NWS of StiS, Sec. 17. T12N. Rl3E
Vanstel
Lonna
Choteau
Rosebud
Yellowstone
Yellowstone
Schaff
Golden Valley
Huntley Sta.
Yellowstone
Michael
Yellowstone
Rolston
"Hi
.
Total Experiments w/Yleld Data
T14N,
T14N,
T32N,
RSE
T31N,
RSti
RSti
RSti
Cargill
Cargill
Bridget
Bozeman
Amsterdam
Rothiemay
Rothiemay
Rothiemay
Marias
R6E
T ION, R56E
TlON1 RSSE
TION1 RS6E
T6N, R43E
T SN, R22E
Evanston
Yaraac
Bainville
Marias
Thurlow
APPENDIX II
118
Soil Series, Textural Class, Textural Family, Classification, Water
Regime, and Temperature Regime of the 127 Experimental Sites.
Series Name
Textural Class
Absarokee
Amsterdam
Assinniboine
Bainville
Bozeman
Bridger
Brockway
Cargill
Chama
Chanta
Cherry
Coffee Creek
Danvers
Degrand
Dooley
Edgar
Evanston
Farland
Farnuf
Floweree
Fort Collins
Gerber
Gilt Edge
Havre
Joplin
Judith
Reiser
Kevin
Kobar
Kremlin
Lonna
Manhattan
Marias
Martinsdale
Marvan
Parshall
Pendroy
Richfield
Rothiemay
Savage
Scobey
Shaak
Tanna
Telstad
Thwrlow
Unnamed
Vanstel
Vida
Vona
Wages
Williams
Winifred
Wormser
Yamac
clay loam
silt loam
fine-sandy
clay loam
silt loam
loam
silt loam
silty-clay
silt loam
loam
silty-clay
clay loam
clay loam
loam
fine-sandy
Textural Family
fine-raontmorillonitic
fine-silty
fine-loamy
fine-silty
fine-silty
fine-loamy
fine-silty
fine-silty
loam
fine-silty
fine-loamy
loam
fine-silty
fine-montroorlllonitic
fine-montaorillonitic
fine-loamy
loam
fine-loamy
fine-loamy
fine-loamy
loam
fine-silty
silt loam
fine-loamy
loam
fine-silty
silt loam
fine-loamy
loam
silty-clay loam
fine-roontmorilIoni tic
fine-montmorilIonitic
silty-clay loam
fine-loamy
loam
fine-loamy
loam
gravelly-clay loam fine-loamy
silty-clay loam
fine-silty
clay loam
fine-loamy
silty-clay loam
fine-montmorilIonitic
fine-loamy
loam
silt loam
fine-silty
coarse-loamy
fine-sandy loam
fine-raontmorillonitic
clay
fine-loamy
loam
fine-montmorillonltic
clay loam
coarse-laomy
fine-sandy loam
clay
fine-montmorillonitic
silty-clay loam
fine-montmorillonitic
fine-loamy
loam
fine-montmorillonitic
silty-clay loam
clay loam
fine-montmorillonitic
fine-montmorillonitic
silty-clay loam
fine-montmorillonitic
clay loam
fine-loamy
fine-montmorillonitic
clay loam
fine-loamy
clay loam
fine-montmorillonitic
fine-loamy
fine-sandy loam
coarse-loamy
fine-loamy
fine-loamy
loam
clay loam
fine-montmorillonitic
clay loam
fine-montmorillonitic
fine-loamy
loam
loam
Classification
Typic Argiboroll
Typic Cryoboroll
Arldic Argiboroll
Ustic Torriorthent
Argic Pachic Cryoboroll
Argic Cryoboroll
Borolllc Calclorthld
Borollic Calclorthld
Typic Haploboroll
Aridic Haploboroll
Typic Ustochrept
Typic Haploboroll
Typic Arglboroll
Aridig Argiboroll
Typic Argiboroll
Ustolllc Camborthid
Aridic Argiboroll
Typic Argiboroll
Typic Argiboroll
Aridic Haploboroll
Ustollic Haplargid
Vertic Argiboroll
Haplustollic Natrargid
Uatic Torrifluvent
Aridic Argiboroll
Typic Calciboroll
Ustollic Haplargid
Aridic Argiboroll
Borollic Camborthid
Aridic Haploboroll
Borollic Camborthid
Typic Calciboroll
Ustertic Torriorthent
Typic Argiboroll
Ustertic Torriorthent
Pachic Haploboroll
Ustertic Torriorthent
Aridic Arglustoll
Aridic Calciboroll
Typic Argiboroll
Aridic Argiboroll
Abruptic Argiboroll
Aricid Argiboroll
Arldic Argiboroll
Ustolllc Haplargid
Typic Cryoboroll
Ustolllc Haplargld
Typic Argiboroll
Ustollic Haplargid
Aridic Argiustoll
Typic Argiboroll
Typic Haploboroll
Aridic Argiustoll
Borollic Caraborthid
Water Regime
Temperature Regime
UBtic
udic
ustic— ssridic
aridic— kistic
udic
udic
aridic— fustic
aridic— ♦ustic
ustlc
ustic— xiridic
ustic
ustlc
ustlc
ustic— +aridic
ustic
aridic—*ustic
ustic— +aridic
ustic
frigid
cryic
frigid
mesic
cryic
cryic
frigid
frigid
frigid
frigid
frigid
frigid
frigid
frigid
frigid
ustic-+aridic
aridic— +ustic
ustic
aridic— +ustic
aridic— Rustic
ustic-*aridic
ustlc
aridic— +ustic
ustlc-»aridic
aridic-'+ustic
ustic— +aridic
aridic-+ustic
ustic
aridic— +ustic
ustic
aridic— +ustic
ustic
aridic-tustic
ustic-+aridic
ustic-^aridic
ustic-aaridic
ustic-taridic
ustic
ustic— +aridic
ustic— +aridic
aridic-+uatic
udic
aridic-*ustic
ustic
aridic-+u8tic
ustic— +aridic
ustic
ustic
ustic— +aridic
aridic-»ustic
frigid
frigid
frigid
frigid
frigid
frigid
frigid
frigid
frigid
frigid
frigid
frigid
frigid
frigid
frigid
frigid
frigid
frigid
frigid
frigid
frigid
frigid
frigid
frigid
mesic
cryic
frigid
frigid
frigid
frigid
frigid
APPENDIX III
120
Site Number, Experiment Number, Average I K response. Maximum % K
Response, and K Rate at Which Maximum Crop Response Occurred.
Slti
No.
I
2
3
4
5
5
6
7
8
9
10
10
11
12
13
14
14
14
14
14
14
14
14
15
15
16
17
18
18
18
18
18
18
19
19
20
20
20
21
21
22
23
23
24
24
25
25
26
26
26
26
26
26
26
26
26
26
26
27
27
27
28
29
30
31
32
33
34
35
36
36
37
38
38
38
39
40
No.
I
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
SI
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
X K res.
101
107
103
103
94
107
124
111
104
104
107
101
100
102
101
111
83
107
95
109
107
103
103
108
95
103
90
98
112
105
HO
121
123
101
108
99
97
103
89
118
HO
102
107
107
105
113
96
102
99
111
106
95
94
102
144
102
109
111
105
116
111
121
100
104
106
104
104
104
120
108
104
98
104
85
98
111
134
102
Maximum
% K ree.
120
107
111
107
101
111
124
119
106
109
108
107
102
103
106
115
86
HO
109
126
114
109
HO
117
96
113
95
98
115
109
HO
129
127
104
109
99
101
105
90
118
HO
107
109
108
109
114
100
104
101
111
107
96
97
102
146
103
109
111
108
121
114
133
114
HO
106
106
106
105
124
109
106
100
106
95
100
116
127
104
K rate of
Max. res.
112
112
28
28
112
28
112
28
112
112
112
28
28
28
112
28
112
45
45
45
. 134
134
134
45
28
28
112
112
28
28
56
134
90
134
112
28
112
28
112
90
112
112
112
112
28
112
112
112
28
90
45
112
90
34
134
28
90
90
22
112
112
90
134
112
28
28
28
112
28
112
112
28
28
112
22
45
134
134
Site
No.
41
42
43
44
45
46
46
47
47
48
49
50
51
52
53
54
55
56
56
57
58
58
58
59
60
61
62
62
63
64
65
66
67
68
69
70
71
72
73
74
75
75
76
77
78
79
80
81
82
83
84
85
85
85
86
87
87
87
87
87
88
89
89
90
90
90
91
91
92
92
92
92
92
93
93
93
94
94
Exper.
Z K ree.
79
80
81
82
83
84
85
86
87
88
89
90
91
93
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
HO
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
Average
X K res.
Maximum K rate of
X K res. Max. res.
103
100
116
109
115
102
107
HO
89
119
115
96
124
112
92
93
104
114
100
118
98
114
97
83
109
106
121
85
117
107
98
101
HO
86
98
90
107
93
104
101
96
130
103
HO
108
107
97
89
90
87
81
102
94
61
112
97
99
109
90
93
93
122
90
107
80
81
120
146
86
70
100
92
103
108
106
107
102
105
HO
112
120
113
115
J03
108
HO
107
119
115
105
124
122
106
105
103
113
101
116
99
120
98
101
112
112
124
98
119
108
100
105
HO
99
104
107
116
102
112
104
100
133
108
HO
109
105
HO
97
100
95
88
104
101
62
130
99
101
115
95
95
95
138
100
116
95
96
143
173
92
70
104
112
115
117
127
114
121
114
134
134
134
90
134
90
45
90
134
45
90
22
45
90
134
45
134
134
45
90
134
90
45
134
22
22
45
45
22
22
90
112
11
134
56
90
134
134
22
22
11
56
56
56
45
90
45
90
45
45
90
67
67
45
45
45
134
90
45
45
56
56
56
134
45
45
134
90
121
Site Number, Experiment Number, Average % K Response, Maximum % K
Response, and K Rate at Which Maximum Crop Response Occurred
site
No.
95
96
96
97
97
97
97
97
97
97
97
97
97
98
98
98
99
99
100
100
101
IOi
102
102
102
102
103
103
104
105
105
105
105
106
106
107
108
109
HO
111
112
113
113
114
115
115
HS
116
117
117
118
119
119
119
Exper.
No.
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
107
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
X
Average
K res.
95
87
132
93
100
101
100
109
115
90
100
118
107
109
90
133
88
116
120
108
107
HO
97
90
100
97
97
117
91
HO
120
99
112
153
118
111
112
100
99
107
HO
100
111
102
111
122
117
104
105
101
95
103
97
99
Maximum
% K res.
98
91
162
106
102
105
106
111
116
100
102
126
107
120
111
139
92
115
128
113
109
HO
112
98
104
97
108
135
98
111
135
100
114
160
118
117
116
101
101
116
115
102
111
103
113
123
117
105
105
106
97
105
100
100
K rate of
Max. res.
56
45
45
45
56
90
56
45
90
90
45
45
45
134
45
90
22
90
134
90
112
112
134
45
22
112
134
45
115
115
115
48
48
57
54
115
57
115
57
96
115
115
54
57
115
48
48
96
48
48
48
48
27
127
Site
No.
Exper.
No.
120
120
120
121
121
122
123
124
124
125
126
127
211
212
213
214
215
216
217
218
219
220
221
222
Average Maximum
% K Res. % K res.
103
93
95
104
95
118
108
96
105
104
97
92
107
98
100
104
102
122
HO
96
106
109
98
96
K rate of
Max res.
96
81
81
96
96
96
48
48
48
115
96
39
APPENDIX IV
123
COMPLETE DATA SET USED IN THESIS PREPARATION
FORMAT STATEMENT (CP6 -SPSS)
INPUT FORMAT
NOTE:
FIXED (F3.0, F1.0, F1.0, F2.0, X, FI.O 5 X 5
5F4.0, X 5 FI.O 5 F2.0, F3.0, F2.0, F3.0, X 5
3F2.1, 3F3.2, X 3F3.0, F2.0., F2.0, F.1.0, X 5
F2.0, F2.0 / IOX 5 F4.0, X 5 F2.1, F2.0, X 5
F4.2. F3.I 5 X 5 FI.O 5 X 5 F2.0, X 5 F2.0, X 5
F2.0X, F2.0, X 5 F3.1, F2.0, F3.0, F4.1, X 5
F3.1, X 5 F3.1, X 5 F3.1, X, F3.1, X 5 F3.1, X 5
F3.1 / IOX f3.0, F3.0, 6F4.0)
Refer to the next page. Format statement begins in column I 5
located 2 columns after the dash mark.
It ends for each card
2 columns before the column with all number ones. The column
with all number ones and the next two digit column (beginning
with number 81) are not part of the data file. The numbers
in these columns are values for V27 and V42 and are located
elsewhere in the data file. They were also moved to the end
of each line for the creation of subfiles for manipulation
of the data. Table 8 may be used as a cross reference.
It
includes the variable names, numbers, and format for each one.
i
1
34 I I 68
I
375632892767
2
8 4 I I 71
8 4 I I 71
2
1082
5 3 4725
88
11
252020292164
967
15 8 4 8 3 4
86
112
J
4
3
I
141171
14 1 1 7 1
5
2
3
I
141171
8 3 1 I 78
831178
2
3
I
11
811177
I 2
I 3
I 4
SI I I 77
1 6 1 I 71
2
3
I
161171
2
I 5
161171
16
I 7
821178
821173
8 2 1 I 78
872177
3
I
2
3
I
872177
872177
973172
2
3
I
973172
973172
2
I 5
I 9
20
21
22
23
24
25
26
27
I 271 I 80
1271130
1271130
873177
873177
3
I
2
3
I
28
29
-
30
31
32
33
34
-
873177
971172
2
3
I
971172
971 I 72
151 I 7J
2
3
I
35
-
151170
2
36
37
-
151170
751176
3
I
38
39
-
751176
751 I 76
2
3
22
443933 178157158
2
81
2
2
028112
1 1 1 3 1 6
242632
24602332221821302265
I 181
70 4 4731
67 2
95
22
022
36093508319932733515
1213
10 8 4 7 4 3 6 7 2
97
134
022
202917071935
028112
14 1 7 1 7
392829
50335047469746574549
1295
30 4 4 7 3 4 6 7 2
1 00
22
022
401619
313229842648
812
10 5 4 8 2 2
95
45
045
876 10
95
112
3
4825
2
72
798
106
00
45
4590134
4593134
045
90
243923452292
853 10 S 4834
96
28
2
23592292209723652285
1146
10 8 4 7 4 0 58 2
100
90
166156177
89
374328
43
41
23
38
I
46
66
39
216222125
72
2
21
15
25
216222125
2
77
74
75
21
15
25
297
71
166144154
90
3
337
57
153140146
79
73
I 13
75
54
216222125
2
21
5
039
00
39
362824
3
90
I 92638
3
I 57160168
I I 6222
8
I
160152133
66
76
I
13
28
216222224
2
21
IS
30
116222
I
21
66
76
92
I 57160168
8
93
028112
283540
163166162
4
254
64
I 20132139
96
I
83
83
87
87
I
I
I 5
21
I 7 51 4 9 1 38
I 75 1 4 9 1 3 8
93
I 5281 5
81
83
I
I
I
I
I
I
90
3 74328
4590134
I
I
I
I
90
022
81
81
I
I
I
90
95
I
I
I
I
I
I
25
90
045
2
76
I 92638
2
4834
3
3
90
150138163
62
5
87
383136
I 23
045
72
151734
3
2
126310011263
798 00
4834
2
1 00
90
21842097
21571908
876
10 8 4 8 3 2 69 2
96
39
340032393105
812
10 5 4 2 2 2
95
45
12101284
961
4590134
142158155
83 2 7 9
63
316212122
294
85
I
66
68
216222
85
70
51
7
4
54
13
28
I
30
55
51
31
15
30
I
I
I
I
I
I
I
I
87
89
89
89
90
90
90
90
90
90
90
90
90
90
90
90
92
92
92
93
I
I
93
93
93
93
93
95
I
I
95
95
I
96
96
96
I
I
I
I
124
10
8 3 1 I 78
81 I I 7 7
011
67
40
41
42
-
43
44
-
-
-
2
3
I
871177
871177
-
11170
2
3
I
237924062312
812
10 5 4 8 2 2
101
45
I 24310081492
-
I I I 70
11173
121171
121171
2
3
I
902
5 7 4843
120
112
I 774 I 7 1 4 1 8 8 2
72
2
69
1 2 1 I 71
3
I
1008
15
106
112
55
-
56
57
-
58
59
-
60
61
-
461174
4 6 1 I 74
-
151171
151171
151171
-
761 I 68
761 I 68
-
761 I 68
2 8 1 I 71
281171
2 8 1 I 71
551
551
551
181
181
I
I
1
I
I
73
73
73
71
71
4822
72
2
67
2
72
2
201619691982
914
10
98
112
I
2
4827
4831
11
045
2
3
I
1 1 4 6 10 8 4 7 4 0
108
11
I 7671 7341 949
58
2
3
I
1 1 27
5
H O
112
64
2
-
262175
262175
78
79
-
262175
261 I 75
3
I
80
-
261175
2
3
I
2
35
28
7
4831
69
2
28963138311129972869
1295
108
5
22
8
4704
65
116222
68
57
8
I
59
21
49
15
2
21
75
13
I
I
I
46
I
I
I
022
4590134
028112
12 14
022
19
4590134
022
4590134
371
I 58
101
46
145
294
60
15 I 64
101
338
151
61
216222224
91
71
39
66
1
I
233745
163155161
7 0 3 101
369
63
216222115
2
81
66
65
21
293419
316212122
41
15
30
I
30
51
64
156142136
163166102
353
40
I 281 3 3 1 4 0
60
66
216222
102
7
216212232
233843
157145154
85
2 1 0 4 1 64
43
316222224
333833
216222224
3
4 5
50
4
35
31
15
30
103
314634
163164163
2 104
91
5
30
102
103
4
2
27152856291028432849
I 295
5 8 4704 65 2
25
282029
028112
32633129329933763268
1 0 4 8 20 8 4741
64 2
103
90
194221241969
880
109
26
I
I
I
I
2
2
98
98
98
99
99
76
98
283540
22
97
97
21
2
O il
97
I
I
I
3
16
143153165
I
S
36
76
028112
13 16
76
116222
19
2
66
I 92638
I 57160168
3
99
028112
12 17
862
10 I
4834
113
28
383341223767
21
I
I
I
2
2
3
I
I
20
3
4590134
8
21
10
96
2
I 95
4590134
116222
216324225
2
68
76
84
86
21
022
2
3
I
90
I 57160168
97
214022
18
2
31183205311132123152
1173
5 8 4 7 1 7 67 2
103
90
I 794 20291 848
4720
15
028112
11 1 7
2
3
I
4
192638
028112
022
2
3
I
90
3
35823730359536363636
1 0 4 8 80 2 4 8 2 0 64 2
104
22
75
76
77
1 8 1 I 71
5
160157152
105
262
48
41
5
213121
149146143
2 106
2162221 I 5
216222115
15
30
21
10
65
20
4
27
2
44
213121
149146143
2 105
41
OO
Kl
73
74
74 I I 7 6
461174
10
90
OO
r\j
70
71
72
741176
74 I I 7 6
876
99
5
5
21
21
10
10
99
101
101
101
101
101
101
101
I
I
I
I
101
101
102
I 02
I 02
103
I
I
103
103
I
I
103
I
I
103
103
104
I
I
104
104
I
I
I
I
I
I 04
I 04
I 04
25
I
I
I
25
I
I
I
105
105
105
105
105
105
106
106
125
68
69
045
171171
171171
8 7 1 I 77
48
49
65
66
67
229822112265
*71176
171171
45
46
47
62
63
64
I
2
3
I
-
50
51
52
53
54
871176
371176
81
-
82
-
83
84
-
85
-
261 I 75
291171
291171
2 9 1 I 71
87
-
61 I I 74
61 I I 7 4
6 1 I I 74
88
-
21 I 70
21 I 70
21 I 70
86 89
-
90
91
-
92
-
71 I I 7 5
71 I I 75
93
94
-
95
-
96
97
3
I
2
3
I
2
3
I
2
107
45
143815251525
1264
10 8 4 7 1 5
106
28
64
2
1 2 0 6 6 0 2 4 8 5 7 58 2
107
112
29532904341929392937
2
1075
71 I I 75
9 0 1 I 75
3
I
-
9 0 1 I 75
9 0 1 I 75
872176
3
I
116
45
1546178815721599
899
10 2 4 8 4 6
116
45
I 821 I 8 8 2 2 0 9 7
98
99
-
872176
872176
I 00
101
-
972172
972172
102
-
I 03
104
-
972172
131171
3
I
131171
2
1 3 1 I 71
2 6 1 I 72
3
I
261 I 72
261172
262172
2
3
I
262172
262172
2 7 1 I 71
2 7 1 I 71
2
105
-
106
107
-
I 08
I 09
-
110
-
111
-
112
-
1 1 3 -
114
-
1 1 5 1 1 6 1 1 7 -
271171
151 I 70
181170
181 I 70
1 1 8 119 -
561 I 73
561173
I 20 -
561 I 73
631174
I 21
-
2
2
3
I
2
3
I
2
3
I
876
115
10
90
5
4809
4822
67
72
022
4590134
I 20
023112
IC I 3
022
4590134
5
41
I 54149 1 34
316222224
4
94
41
216222135
5
73
72
73
21
316322132
5
79
72
73
41
045
90
2
045
25802738274828762580
1295
5 8 47 0 4 65 2
111
90
022
25832621278228562688
I 295
5 8 4 7 0 4 65
2
111
90
022
229225002607
1242
5 I
4709
114
112
028112
65
2
69
2
192222182083
874
15 4 4 8 3 2
3
I
115
28
16171540182917751614
2
893
113
72
022
I 8
4590134
I 9
4590134
2
20312131241722982162
022
77
147152129
107
34
402
167
107
396
66
96
105
151
I 5
66
4590134
334036
143170158
45
111
306
58
97
15
30
156
10
I
106
I
I
I 06
106
I 06
18
56
89
69
116222
58
8
I
58
21
I
58
54
59
15
I
I
13
I
15
77
30
213121
149146143
2 111
2162221 I 5
5
21
10
25
283917
I 50142 1 56
2 111
216222115
5
21
13
25
333833
I 6 0 1 5 71 5 2
133
3 112 322
77
216222224
I 72617
I 5 9 1 501 5 C
91
3 114 167
25
216222
33
42
7
304324
216222
7
147144137
10
28
216222115
61
21
76
213121
149146143
2 111
58
5
33
56
56
66
I
I
I
I
I
I
I
I
20
68
I
374328
175149138
109
4 5 90 1 34
028112
11
16
625
192638
I 57160168
3 109
90
028112
12 14
106
331920
I 99
2
207623862211
1013
10 8 4 8 3 8
115
28
4810
216222224
393523
164151140
I 35 3 107 360
74
33
6790
2
2
8
I 5
2
3
045
I 0 0 1 I 1 0 9 1 0 75
798 00
4834
111
45
10
45
137
106
3 2 3 4 36
2
2
3
I
2
I 53
25
I 06
I 06
106
107
I 07
107
107
107
I
I
I 07
107
107
107
I
I
I
I
I 09
109
109
I
I
I
I
109
109
I I I
I I I
I
I
I
I
I I I
I I I
109
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
1
I
I
I
I
I
I I 2
I I 2
I I I
I I I
2
64
21
10
62
20
2
21 2 8
I 4
43
I
I
I I 2
I I 4
I I 4
21
36
I
I
I
53
2
23
114
I I 7
126
3
I
8
15
I
25272789276226482822
1027
2 0 8 4 7 3 2 64
2
1 1 2 I 34
240625672574
15
17
028112
I 22
-
631174
2
I 23
I 24
-
631174
974172
3
I
974172
974172
2 7 1 I 74
2
3
I
2 7 1 I 74
2
271174
891175
891 I 75
3
I
-
I 25
I 26
I 27
I 28
I 29
I 30
131
I 32
I 33
I 34
I 35
I 36
I 37
I 38
I 39
143
I 44
145
I 46
147
I 48
149
I 50
I 51
I
I
I
I
52
53
54
55
I 56
157
I 58
159
I 60
161
I 62
-
2
891175
621 I 75
621175
3
I
-
621175
681175
3
I
681 I 75
6 8 1 1 75
2
-
541173
-
541173
541173
801177
801177
801177
361171
361171
2
3
I
2
3
I
2
3
I
2
3 6 1 I 71
421172
3
I
421 I 72
421172
11 I I 7 0
2
3
I
111170
11 I I 7 0
71 I 7 0
71 I 70
71 I 70
2
3
I
-
3 0 1 I 71
301171
-
3 0 1 I 71
-
362171
2
3
I
362171
362171
2
3
-
2
3
I
798
126
00
45
72
045
4834
8
4539
90
64
022
4590134
04 5
6790
14 2 0
4590134
283917
54
48
50
I 751491 38
I 501421 56
216222115
5
203127 157156132
I 97
122 4 4 7
71
84
81
79
022
4590134
I
276527702646
1 4 2 9 30 2 4 5 4 7
028112
103
28
231923652455
I 3 7 9 10 7 4 5 4 8
106
112
312933083175
1379
20 2 4 5 4 8
64
64
28
4 5 90 1 34
4590134
2425
172187
216222
7
216222
2
I I 8
121
121
121
I 22
122
13
25
I
I
23
58
I
I
I
I
I
74
8
30
2
2
2
90
103
7 3
I I 3
32 10 23
109
90
2
2
32
2
2
323128 142127131
111
I
93 2 3 3
52
216222
40
37
7
373015
148133130
99 I
97 342
51
216222
56
58
7
231613
126124
7
I
126124
7
I
74
2
022
4590134
22
124121124
3
35
4
57
23
36
38
33
32 30 46
59
61
12
20
33
98
452307
74
2
149137134
1 00
470
83
86
77
023112
I 50406
1 451 341 4 0
101
I
102
028112
242209 142138135
114
I 1 04
216222
028112
222305
I 371 201 23
I
104
216222
231613 I 24121124
2 104
126124
I
I
2
I I 8
I I 8
21
85
114225
144141135
I
86 5 8 7
82
2
028112
67
I
I
28
12 23 38
75
75
2
22
28
38
7
3
32
8
20
7
3
32
18
33
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
28
212222532181
1 4 2 9 30 2 4 5 4 7
I I 7
13
28
33043359338934223709
1455
50 I
4 5 4 6 67 2
112
134
158716321609
1371
10 I
4 5 4 9 64
I
I
I
I
121
3
022
106
I 52
I
022
15521324142516061700
I 4 62 20 I
4 5 5 6 64 I
110
134
106
374328
2
022
67
117
118
31812717334126883072
I
I 5 6 9 4 0 3 4 5 3 8 64
105
45
100
3
2
I 7 0 0 1 9 7 6 2 0 5 0 2 2 5 8 2 1 71
1242
5 I
4 7 0 9 65 2
133
90
1962205027082419
9 3 7 50 8 4 8 3 3
2
138
67
34263308292329943341
72 2
1523
30 2 4 5 3 7
98 134
43854345386338834357
1554
50
99
134
I 50
2
7
I
12
20
33
2
2
2
I I 7
122
85
85
85
86
86
86
93
93
93
97
97
97
98
98
98
I 00
I 00
I 00
102
102
102
104
I 04
104
I 04
104
104
I 04
104
I 04
127
I 40
I 41
142
-
893
10 2 4 8 1 0
119
45
8871122 968
-
163
164
I 65
I 66
167
I
I
I
I
-
68
69
70
71
-
I 72
I 73
I 74
-
I 75
I 76
I 77
I 78
I 79
I 80
181
I 86
187
I 88
I 99
I 90
I 91
192
I 93
I 94
I 95
I 96
197
198
I 99
2
3
I
791 I 77
791 I 77
601174
2
3
I
601174
601174
2
991172
2 0 1 I 71
2 0 1 I 71
-
531173
531 I 73
-
200
201
202
-
203
-
-
2 0 1 I 71
531173
2 4 1 I 71
241171
2 4 1 I 71
1 0 3 1 I 72
1031172
1031 I 7?
211171
21 I I 71
211171
201170
201 I 70
201 I 70
2 2 1 I 71
221171
2 2 1 I 71
21 I I 7 0
022
52624805539655245000
1520
5 2 4 53 9 67 I
105
90
39744256408644514257
022
1569
112
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
20
90
I
4537
67
124
30
134
2
4536
72
4590134
022
4 59 01 34
022
265423522392
1 0 0 5 60 6 4 7 2 6
90
112
0281 12
53
4590134
298427692984
625
10 4 4 8 2 6
10
11
0281 1 2
4590134
2
022
164015991747
028112
2
235924862365
60 6 4 7 2 6
28
53
2
2
3
245926072668
800
20 5 4 7 2 5
109
112
67
2
I
277124932507
64
2
67
2
2 1005
3 105
21 I I 7 0
2 1 11 70
2
701
35
3 I 10 112
24 I I 70
24 I I 7 0
2
2
4752
10
4
4826
4590134
363315
I 4213 61 26
71
I
1 09 4 50
87
2425
3
172187
1 21
23
26
31
028112
37
1 36
92
98
136
93
74
26
3
71
32
3
32
78
226222135
3
76
74
72
216222
80
134
560
78
7
68
7
75
2
91
225232215
3
33
225232215
22
21
4
64
30
3
3
88
89
3
3
3
3
89
89
92
92
3
3
92
96
38
3
3
3
96
96
97
97
97
28
3
3
3
373727
216324225
23
2
23
97
2 3 2 3 1 8 1 561491 4 5
2 102 510
93
216232232
23
225232215
3
93
98
67
33
243524
146133131
2 1 07 5 0 3
78
216222134
78
23 23 38
77
77
232318
99
216232232
2
101
88
82
23 15 28
85
83
182822
145146136
2 113 5 70 115
216324225
2
108
77
101
2 3 3 6 51
84
85
26 1 48
103 502
134
92
156149145
1 10
538
91
97
89
96
2
83
2
23
73
70
15
121
88
88
2 3 3 6 51
79
81
I 69160162
109
109
3
3
46
15
2
9
107
107
109
121
121
76
33
2
2
I 04
I 04
104
107
2
2
216324225
2
84
85
76
2
2
2
2
2
2
13
79
13
2
2
2
22 30
104
23
93
33
3 2 3 0 51
74
67
2
111
36
67
82
216232232
103
25
70
182822
145146136
2
96 511
106
2
028112
148
89
2
2
023112
417
332928
161148160
264
2
88
268
028112
I 9 0 2 2 1 1 0 2 1 71
625
226222135
111
37303609365637094045
625
10 4 4821
67 2
108
134
2
454645
I 25157142
102
I
107 373
61
2
022
701
35
107
112
216222
85
81
2
234 I 248 421 7 3 2 2 1 0 2 4 0 6
7 7 7 20 8 471 I 53 2
106
22
67
292117 153137138
161
1 104 462
88
96
40923770370935883662
695
30 3 4 7 5 3 64 2
92
22
64
I
4590134
2
4752
I
022
4 59 0 1 34
I
15371751190018041911
2 1523
991 I 72
-
-
3
I
621174
621174
621174
9 9 1 I 72
32793683354134143562
1 4 6 2 20 3 4 5 5 7 64 I
112
22
81
13
82
3
3
3
3
3
3
3
3
102
102
I 02
103
103
103
107
107
107
3
I I 0
3
3
3
H O
I I O
3
I I 3
I I 3
128
I 82
183
I 84
I 85
I
731 I 76
731 I 76
731176
791 I 77
-
208
209
-
210
241 I 70
1032172
3
I
1032172
1032172
5 91 I 74
2
591174
591174
3 7 1 I 71
2
2
3
I
2
3
I
1 0 4 1 I 76
-
21 8
219
-
1041176
41171
4 1171
41171
220
221
222
223
224
-
I 181 I 73
1181173
1181178
-
121 I 1 3 0
I 21 I I 3 0
121 I I 3 0
-
1241173
I 2 4 1 I 73
124117%
225
226
227
228
229
230
231
232
233
234
-
581175
581175
-
I 2 6 1 I SO
I 261 I 30
235
236
237
-
238
239
240
241
242
-
243
244
-
581175
I 261
381
381
381
581
I
I
I
I
I
80
71
71
71
73
581173
581 I 7J
651175
6 5 1 I 75
6 5 1 I 75
2 5 1 I 71
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
8
4613103
305427732900
1 3 5 6 10 4 4 5 3 4
3
2
69
2
279824262838
990
15 2 4 5 3 2 1 0 6
101
112
3
424141173963
899
IC 4 4605
97
48
997 8761017
20
96
4
4617
232722452223
1 1 2 7 30 8 4 6 1 6
96
48
3
364250
2
057115
8 12 17
69
2
0 4 8 96
7 10 16
67
2
0 4 8 96
9 1 9 21
048
64
411818
23
028112
2
35073275332534353351
7 7 7 10 4 4 6 4 7 58 2
98
90
292428042669
929
20 7 4 5 5 5 92 3
98
96
309030752994
23
5
44
49
111314232
I 9
147185192
121314214
5
44
28
48
5
34
I 3
25
147
83
131
68
23
38
20
85
5
022
20
25
13
I 5
21
45
30272757288629862901
777
10 4 4 6 4 7 58 2
99
90
022
4590134
34193214329033733413
820
10 5 4 6 1 6 6 9 2
I 0 0 I 34
336034073279
022
20
22
6
4616
94
3
4590134
028112
I 5 0 1 34 1 3 4
91
246
107
382241
149161171
1 9 3 4
9 5 2 4 5 1 4 9
86
4
154744
4590134
0 4 8 96
6 15 16
3
1 73624
96
9
117
303936
149176170
4
94
022
792
100
16
2
112
2 9 0 6 2 4 3 9 2 8 53
8 2 3 20 4 4 6 1 6
98
115
838
102
19
I 33
023112
58
95
4590134
216324225
212
4
I 381491 57
95
312
76
118178167
96
309
164
433428
I 701 381 3 8
3
97 305
80
323630
138150149
131
4
97 312
79
322338
162155148
4
98
4 3 3 4 28 I 7 0 1 3 8 1 3 8
I 49 3
98
79
9
373340
I 521451 5 I
3
98 4 96
69
192426
128134121
216322234
62
42
35
7
5
34
25
41
326222325
4
44
25
46
1 1 6322
38
22
316224
36
7
5
44
I 5
28
316334334
6
44
33
46
126224225
24
25
46
111222132
5
83
81
69
44
8
23
211224
2
24
I 5
25
2
24
25
46
89
100
7
126224225
42
28
216322234
91
2
36
72
125222225
3
3
3
4
3
3
34
18
100
34
10
36
7' 4
23
I I 7
4
4
83
83
83
4
4
4
4
85
85
85
9 I
4
4
4
4
91
91
4
4
4
4
94
94
94
95
95
95
4
4
95
95
4
4
95
96
4
4
96
96
9 7
97
4
4
4
4
4
4
4
4
4
4
4
90
I I 3
I I 7
I I 7
4
4
4
4
4
97
97
97
97
98
98
98
98
98
98
98
98
98
99
129
-
35
45
022
373727 169160162
268 2 117
•o
O
215
216
217
1213
101
4590134
rxi
-
3 7 1 I 71
371171
I 04 I I 7 6
3
I
3
022
00
-
112
21642446292322982406
625 10 4 4821
67 2
1 35
45
14851320150410971359
OO
OO
211
212
213
214
3
I
114
LW
-
IW
-
205
205
207
OO
I*
204
245
-
251171
2
246
247
-
251171
-
1 0 5 1 I 78
3
I
248
249
-
250
251
-
1051178
1051173
I 101 I 76
2
3
I
11011?6
I 101 I 76
101171
3
I
101171
2
101171
3
I
252
253
254
255
256
257
258
259
260
261
262
263
264
3
I
661175
661175
I 171 I 30
1171130
2
3
I
1171180
2 5 1 I 70
3
I
251 I 70
2
3
I
-
270
271
-
275
276
277
278
279
-
282
-
283
284
-
285
-
280
281
251 I 70
1141177
I 14 I I 7 7
2
2
2
I 14 I I 7 7
3 1 I 70
3 T l 70
2
31 I 70
41170
3
I
4 1170
2
41 I 70
1191178
I 191 I 73
1191173
3
I
3
I
2
3
69
1158
20 4 4 5 5 1 1 0 6
101
57
252525682501
1158
10 2 4 6 0 7 1 0 6
102
28
306230193095
I 2 4 9 40 I
4600
101
115
318131523238
1181
10 4 4 5 4 9
102
115
61
64
2
2
0 4 8 96
2
8 13
3
057115
0
3
9
212021842120
326
10 4 4 6 1 6 60
103
57
229325462157
I 0 3 6 80 6 454 5108
111
28
205521922055
990
15 2 4 5 3 2 1 0 6
107
28
384240393888
1194
5 4 4548
1 05
48
67
95
83
216322234
165160165
I 3
90
34
10
52
131224135
54
3
84
81
5
34
I 3
25
49
028112
203833
I 55160166
4 I 00
131224135
5
34
20
30
342527
216222132
2
24
13
25
2
0571 I 5
5
9 14
2
057115
I
8 14
2
022
203833
I 58160166
I 7 2 4 1 01
4590134
0 4 8 96
11
17 21
22
436
75
168154159
4
99
4
99
99
99
4
4
4
4
4
41
131224135
34
20
30
7
4
24
55
63
192426
128134121
2 102 520
86
125222225
3
94
89
87
34
10
23
84
434526
116222125
24
169156156
86
4
149134149
32
2
80
25
46
4
4
4
4
4
4
4
4
4
4
4
4
4
4
22
99
99
99
99
I 00
I 00
100
I 00
I 00
I 00
I 00
I 00
I 00
101
101
101
101
101
101
101
101
101
102
102
102
102
216132135
I
14
15
30
116322
5
34
25
41
4
4
4
I 02
102
103
103
103
103
103
30
4
4
103
103
4
103
103
103
166168159
4
4
4
4
63
1 5 6 3 1 0 1 2 5 2
474641
I 9
23
41
303936
149176170
I 50 4 103
3
44
28
181
028112
216212232
5
94
89
67
5
4
4
4
4
4
4
24
171824
028112
84
116222125
2
71
64
54
23
3
0 4 8 96
4
7 17
101
292421
27
75
126
4
15
I 551 571 6 0
I 8
057115
2
82
3 3 4 1 34
2
2
163148158
I 7
0 2 8 H 2
61
I 57
107
92
443534
028112
60
114
I 7
3
1158
10 2 4 6 0 7 1 0 6
3
107
28
34883453337736543543
801
10 8 4 6 1 7 6 7 2
105
90
257425802681
701
10 4 4 7 4 5
104
112
30
564
4
277529772648
149915821453
829
20 5 4 6 1 6
I 06
48
99
7
221322122
75
85
4
94
44
15
4
4
4
130
101 I 70
6 6 I I 75
61
O
oo
-
4745
OO
101170
101 I 70
-
273
274
1131177
2
3
I
-
268
269
272
1131177
2
3
I
4
O
266
267
-
1091176
1091176
1 0 9 1 I 76
1131177
10
OO
265
-
2
701
101
28
239023872360
1 4 0 9 45 8 4 6 0 8
100
48
236823912314
-
286
287
288
289
290
291
292
81 I 70
31 I I 71
3
I
31 I I 71
31 I I 71
2
3
I
-
296
297
-
298
299
-
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
-
321171
3 2 1 I 71
321171
351171
3 5 1 I 71
351171
1161178
I 161 I 78
1161178
1211179
I 21 I I 7 9
I 2 1 1 I 79
1251177
I 25 I I 77
1 2 5 1 I 77
2 3 1 I 71
2 3 1 I 71
2 3 1 I 71
2
23
344742
028112
67
2
377639843896
935
15 2 4 5 4 0 1 0 6
106
28
3
363236953828
253642
163159164
97 3 103 172
87
220
25
283725
162148141
4 104
216222125
33
4
4
4
4
364444
121314214
30
4
4
4
2
0 4 8 96
7 12 17
22
63
350035043775
920
10 I
4556
109
115
92
3
61
2
2
3
I
2
3
I
2
3
I
2
I 90720161 976
I I 27 30 8 4616
106
48
267729712747
1249
10 4 4 5 3 0
11 I
28
2
I
58
2
2
I 7
0 4 8 96
6 10 15
4
8
96
14
028112
3
183
207
73
322335
155141140
1 1 0 4 1 0 4 4 4 6 21 2
201
3
4
462
96
168154159
1 05
I 54744
72
141137145
105
292421
I 9
443
I 381491 57
I 04
142327
19
163177185
104
1 73624
2
028112
4540106
368339453965
884
45 5 46 4 7
057115
6
7 11
048
187619872033
935
15
108
112
96
10
028112
64
104
15
67
2
4
4
4
44
048
oO
30
5
228121152378
838 20 4 4 6 1 7
104
96
3 0 3 9 3 1 9 2 3 1 83
829 20 5 461 6
105
48
15
216222232
398
4647
44
423923
164151149
3 104
I 8
5
103
104
I 04
104
4
028112
14
45
28
4
221322122
4
4
4
6
884
109
103
103
51
0281 I 2
163317741667
4
4
30
2
2
3
I
33
44
58
2
3
I
13
4
316832363335
1274
10 8 4 5 3 7
105
96
2
3
I
44
25
216313224
3
4550
23
414242
164159148
4 104
297430433158
827 20 8 4 6 1 6 1 1 4
106
112
2
3
I
4
2
I 45161166
4
37
028112
I
1241179
I 24 I I 7 9
9 1 I 70
04 8 96
11 1 6
2
2
3
I
231 I 70
2 3 1 I 70
2 2 6 4 2 2 6 1 24 70
1302
25 2 4 5 3 0
109
112
3
67
1171178
1171178
I 24 I I 7 9
70
70
70
70
70
4539106
1 508
50
105
112
I 171 1 7 8
I
I
I
I
I
8
2
I
2
3
I
51
51
51
91
91
30
96
290
148
I 18178167
105
141
59
404626 166157155
117
107
354141
I 32
4
157171171
93
87
316224
50
2
S
24
44
18
20
80
7
36
5
44
15
28
48
1 1 1 2 2 2 1 32 5
84
89
61
44
125222225
3
85
81
77
34 18 30
64
59
116222125
24
55
62
2
8
28
23
41
25
4
4
4
4
4
4
4
4
4
4
4
4
4
4
316334334
6
29
33
20
44
33
46
121314214
5
44
5
15
216313224
4
44
20
30
125222225
3
34
18
30
107
4
4
4
4
4
4
4
4
4
028112
61
2
142327
2
141137145
107
536
99
102
97
88
83
67
104
104
104
I 04
I 04
104
I 04
104
I 04
I 04
104
I 04
104
I 04
104
I 04
104
104
105
105
105
105
105
105
105
105
105
I 07
107
107
107
107
4
107
107
4
107
131
308
309
996
107
-
295
305
306
307
219221662349
S I I 70
81 I 70
-
303
304
I
2
3
I
-
293
294
300
301
302
I 201 I 79
1201179
1201179
327
231170
3
108
328
329
1111179
111 1 1 79
I
2
228222412645
121 I 20 8 4 5 3 8
330
331
332
I 11 I I 7 9
34 1 1 7 i
3
i
I 16
96
1 58717071 722
3 4 1 I 71
3 4 1 1 71
2
3
1 0 3 6 30
1 09
112
333
334
335
336
337
•
338
339
340
■
•
•
343
344
•
■
345
■
346
347
■
-
34 8
349
-
350
351
-
352
353
354
-
35 5
356
357
-
358
359
-
S
4545
048
58
2
67
2
64
2
6
96
9
16
0 4 8 96
6
7 16
2 9 4 141
I 38156163
85
2 108 145
50
216212232
5
13
53
29
44
I 5
33
4
4
4
21 2 8 3 7
131224135
34
13
28
25802467247428372689
I 2 0 3 20 4 4 6 0 9 1 0 6
3
I 10
90
022
1051176
I
388442404308
0 5 7 1 15
1 051 I 76
1051 I 76
1121177
2
3
I
1 4 0 9 45 8 4 6 0 8
111
115
257227032957
1121177
1 121 I 77
2
3
9 9 6 20
1 15 1 15
3
251629932609
381173
381 I 73
I
2
32473691375335463471
7 9 2 20 6 4 6 1 6 94 3
381 I 7 3
1071176
3
I
1 I 6
45
230624022704
1071176
1071176
1151177
1151177
2
3
I
2
829
10 4 4 6 ' 6
117
115
I 70218351931
1211
20 8 4 SS1
1151177
3
113
36 0
361
1051179
1051179
— 1051179
1081 I 76
I
2
3
229426242510
1 4 0 9 45 3 4 6 0 8
114
48
I
I 81 3 2 1 1 2 1 9 5 9
362
363
364
—
2
3
967
116
581174
I
20352316225122752447
365
-
3 66
367
-
581174
581 1 74
2
3
7 7 7 10
120
1 34
1151179
I
210524662458
1081176
1081176
4
4550
5
7
I 10
30
3
154
I 57
110
505
50
158
113
22
474542
I 63150161
1 1 3 3 I 10 3 6 2 1 72
216
223718
145168163
9 9 3 111
2
022
I 39150163
12
023112
67
4
39
0571 1 5
6 13 20
1 508
119
40
28
43
0
I
20
4 59 0 1 34
2
2
3
I 08
108
4
4
4
I
2
3
61170
107
107
46
471173
47 1 1 73
471 1 73
6 1 I 70
611/0
4
4
4
28
209722972253
4539106
67
38
34
1 1 88
110
8
161
23
3
I
69
107
44
115222235
2
3
4548
2
121314214
5
33
34
27
383525
I 50166162
2 108
1231179
3
70
I 36167149
028112
1 2 3 1 1>9
1231179
10
48
264224
4590134
72
2
0 5 71 1 5
8 '3
20
75
2
0571 15
7 10 1 6
69
2
96
8 1 4 2 0
3
0571 1 5
6 11 1 5 2 1
322338
162155148
1 2 6 4 1 1 1
94
15
27
132918
I 59163172
97 3 111
274
126
18
424039
I 571631 73
1 0 7 4 I I I 384 171
55
5
49
4
216322234
3
110
89
90
34
4
44
73
60
10
I 3
33
57
5
44
20
30
211224
2
24
I 5
25
7
23
216222
61
63
31
2
24
24
20
30
215
5
87
89
44
I 5
25
216322234
3
25
24
21
34
216122
64
61
2
70
24
13
33
126224225
2
67
44
17
24
25
46
216
37
7
115
25
57
I
048
4540100
39
30 I 57
9 2 3 1 1 2 1 4 7
158
77
403233
I 59158168
237
3 112 332 137
7
4
4
4
4
4
4
4
216222232
26
4
4
10
4647
58
022
4 59 0 1 34
2
4 3 3 4 28
3
I 701 3 8 1 3 8
I 14
200
51
21
96
424039
I 571631 73
216
215
5
44
25
H O
I I O
I I O
I I O
n o
I I O
H O
4
4
111
111
4
111
111
4
4
4
4
4
4
4
4
4
4
4
4
4
4
I 5
I C8
108
108
108
4
4
4
048
107
I I O
I I O
111
111
4
4
107
4
111
111
111
111
111
111
I I 2
I I 2
I I 2
I I 2
I I 2
I I 2
I I 4
I I 4
I I 4
I I 7
132
341
342
112
1151179
- 1151179
57 1 1 7 3
571173
571 I 73
- 1221179
- 1221179
- 1221179
481173
48 I I 73
481173
331171
33 1171
331171
- I 051 I 77
- 1051177
- I OS I I 77
- 1151178
- 11 51178
- I 151 I 78
5117 1
5117 1
51 I 71
51 I I 73
51 I I 73
51 1173
391 I 74
39 1 1 7 4
39 1 1 7 4
- I 061 I 76
- 1061 I 76
- 1061176
901275
901275
901275
921272
921272
921272
701275
70 1 2 7 5
701275
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
121 I 20 8 4551 75 2
117
48
25582603296329432771
1 0 6 0 30 6 451 7 69 2
116
45
235926672889
896 10 4 4 6 0 5 64 2
122
96
I 61 7 1 4 3 9 1 9 2 1 I 801 I 697
1 2 64 25 I 4 5 2 9 58 2
119
45
134016631549
12 64 30 2 4 5 2 9 58 2
I 24 I 28
182919042470
1 4 0 9 45 8 4 6 0 8 69 2
I 35 115
225827882721
1211 20 8 4551 75 2
1 23
48
246130433050
1 249 10 4 4 5 3 0 58 2
I 24 112
31663473373839203725
827 10 3 4 6 1 6 1 1 4 3
I 24
90
28732697348135403662
792 20 6 4 6 1 6 94 3
127 I 34
135121671975
I 2 8 0 10 5 4601 67 2
160
57
2439231215661989
899 10 2 4 8 4 6
2
95
45
23921908
27552540
1161
5 8 4 8 44 69 2
92
45
30553259300927422901
1 1 3 3 20 8 4 8 1 9 58 2
107
22
5
9 14 20
58
4 117
210
80
41
47
42
3 4 4 5 4 3 I 541 5 5 1 6 9
47 2 113 233
28
1 3 1 2 2 2 1 3 2 4 44 23 43
64
46
SC
45
0 4 8 96
7 14 18 22
274138 151161159
93 4 118 175
67
215215
34
40
022
3 7 3 9 3 9 I 561581 60
47 2 119 123
23
121314214
71
29
5 44
28
43
364630 153157159
2 120
121314214
5 44
13
25
022
45901 34
45901 34
028112
057115
6 10 16 18
39
I 10
0 4 8 96
5
7 1 51 6
424039 157163173
2 7 4 4 1 22 47 8 201
028112
30 I 57
3 I 20 351
158
1 37
7
3 34 41
34
61
2 1 6 3 2 2 2 3 4 3 34
68
70
76
10
2 1 5 5 44
94
76
I 5 25
216
107
404626 166157155
2 I 24
121314214
5 44
2 24
9
5 I 5
022
4590134
1 43037 I 70159165
I 26 4 I 24
90
22
216212232
38
21
022
4590134
2 9 3 3 2 4 I 58 I 4 8 1 5 0
I 46 4 I 34 28 6
48
211224
7 2 24 13 25
74
52
56
56
057115
O 5 10 15
354346 148165163
1 38 3 I 53 31 0 1 72
216222232
72
66
045
33
I 99
6790
34
167
151
80 396
66
89
69
I 34
5 44
58
I
58
6 41
23
33
38
53
15
56
045
00 45
4C4521 1 6 7 1 6 0 1 4 7
4
86
214
022
4590134
27 35 30 I 35134162
3
90 4 8 2
87
2 1 6 2 2 2 1 3 2 5 21 25 41
91
88
76
72
I58
10 20
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
I
I
I
I
I
I
I
I
I
I I 7
I I 7
I I 8
I I 8
I I 8
I I 8
I I 8
118
I I9
I I 9
I I9
I 20
I 20
I 20
I 20
I 20
I 20
122
I 22
I 22
I 24
I 24
I 24
I 24
I 24
124
I 34
134
I 34
153
I 53
153
80
80
80
86
86
86
90
90
90
133
368
36 9
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
383
38 9
390
391
392
393
394
395
396
397
398
399
4 00
401
402
403
404
405
406
407
408
-
89 1 2 7 5
89 1 2 7 5
891275
881275
881275
381275
26 1 272
261272
261272
141273
14 1 2 7 3
141273
261271
261 2 71
261271
561273
56 1 2 7 3
56 1 2 7 3
92 1271
92 1271
9 2 12 71
971271
97 1 2 7 1
971271
973272
973272
973272
971272
971272
971272
85 1271
85 1271
851271
941273
941273
941273
14 22 72
14 2 2 7 2
142272
142273
14 2 2 7 3
I
2
3
1
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
1633147816261304
937 50 8 4 8 3 3
2
100
67
2500223823652386
90 2 30 6 4 8 3 3
2
95
90
28362486251327622728
1295
5 8 4704 65 2
97
90
8 2 0 894 692
746
975 15 8 4834
2
109
45
148513841425
1 295
5 8 4 7 04 65 2
96 112
892 853 890 88? 897
893 10 8 4 8 10 72 2
101 1 34
18351908
20161962
1161
5 8 4844 69 2
104
56
21372097
I 7471 788
79 3 00
48 34
2
102
56
9 8 3 1 0 0 8 995
79 8 00
4834
2
1 02
45
108210551136
798 00
4834
2
105
90
20972097
23252412
8 3 0 10 I 4 8 0 8 64 2
104
56
128 128 I 08
155
1 0 0 5 80 3 4 8 34
2
121 134
974 9 0 7 1 0 4 2
1062
9 7 5 15 8 4834
2
109
13
887
80 6 840 773
9 7 5 15 8 48 34
2
045
6790
14 20
022 4 5 9 0
19 21
203127 157156132
I 97
90 4 4 7
71
84
81
79
35
237
66
66
74
36
I 64
93 399
148
61
022
4590134
213121 149146 143
2
94
045
90
242632
I 34
028112
142158155
95 328
64
2162 22 1 I 5 5
41
30
51
213121 149146143
2
95
2162221 I 5 5
022
4590134
1 7261 7 I 591501 50
91 3 100 I 54
33
21 6 2 2 2
7 2
38
48
25
056
00 56
404521 167160147
4 I 00
214
056
00 56
374328 175149138
100
I 34
6
2 9 4 3 2 6 I 541491 4 5 2 1 6 3 3 4 3 3 4
4 102 345
6
045
90
374 328 1 751491 38
100
04 5
90
3 7 4 3 2 8 I 751491 38
101
05 6
045
OC 56
90
I 34
353430 I 49150157
102 364
66
045
90
I 34
242632 142158155
103
045
90
I 34
2 4 26 32 142158155
103 328
64
71
66
64
41
30
51
8 23 I
58
I
I
I
15 I
66
66
I
I
21 10 25 I
I
I
I 46 66 I
71
71
I
I
21 10 25 I
I
I
21 28 43 I
I
10
I
41 10 20 I
I
I
I 13 28 I
I
I
i 13 28 1
I
I
I 13 28 I
I
I
I
I
1
I
5 20 I
I
46
51
I
1 46 66 I
I
I
I 46 66 I
71
71
I
I
74
O
f\>
*
-
90
90
90
93
93
93
94
94
94
95
95
95
95
95
95
I 00
I 00
I 00
I 00
I 00
I 00
I 00
I 00
I 00
I 00
I 00
I 00
101
101
101
102
102
102
I 02
102
102
103
103
103
103
103
134
409
410
4I I
4I 2
41 3
414
41 5
4I 6
4I 7
418
4I 9
4 20
421
422
423
424
425
426
427
428
42 9
4 30
431
432
4 33
434
435
436
4 37
438
439
440
441
442
443
444
445
446
44 7
448
449
14 22 73
942273
942273
942273
931273
931273
931273
14 1 2 7 2
1 4 12 72
14 12 72
93 2273
932273
93 2273
974272
974272
974272
931272
931277
931272
86 1 2 7 2
86 1 2 7 2
861272
972272
972272
972272
671275
671275
671275
501273
501273
501273
192271
192271
192271
1021272
1021272
1021272
191271
191271
191271
1022272
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
110 134
141 128 161
155
1 005 80 3 4834
2
114
90
110914111122
ions
1 1 0 7 20 4 4 8 30 64 2
1 27
45
1142
114212571250
975 15 8 4834
2
110
45
116913371149
1 2 57
1 1 0 7 20 4 4 8 30 64 2
114
45
9 0 7 9 7 4 961
7 9 8 00
4834
2
1 07
45
161316871673
1 8 88
1 1 0 7 20 4 4 8 30 64 2
117 134
531 692
47 7 437
853
5 I 4846 72 2
130
45
98811161149
7 9 8 00
4 8 34
2
I 16
90
43574689481247344650
1 5 5 4 20 4 4 5 3 9 67 I
110
45
20592155208919632138
865 20 7 4 8 2 3 64 2
105
22
9 0 0 833 907
868 10 2 48 24 64 2
101 112
32262473320531523622
603
5 5 4 8 2 0 64 2
112 134
167316601653
8 6 8 10 2 4 8 2 4 64 2
99
28
35623716349435823420
04 5 90
134
353430 149150157
105 364
66
71
216
36
I 22 3
74
74
I 22 3
74
74
045
90
I 34
233713 150143148
4 1 0 6 413
71
045
90
I 34
242632 142158155
107
04 5
90
I 34
2 337 1 3 I 5014 31 4 8 216
4 107 4 1 3
71
36
04 5
90
045
90
045
66
64
374328 I 75149138
107
I 34
233713 150143148
4 108
216
30
216222
17 I 55
3 112
130
I 22
3
7 2
045
90
3 7 4 3 2 8 I 7 51491 38
115
022
4590134
4 13947 I 42166164
I I 10 465 I 29
226222135 3
I I 7 113 106
022
4590134
214616 I 38161144
80 2
96 169
34
316322121 2
47
23
SI
153025 160183163
2
97 2 2 9
82
216222122 I
34
29
46
023112
022
4590134
20 22 30 165173151
578 2
97
135132135
I
0231 I 2
I 53025 160183 163
2
99 366
73
216222122 I
79
37
68
02 2
202230
135132135
4590134
165173151
I
I
I
5 20 I
46
51
I
I
31
5 15 I
79
79
I
I
I 46 66 I
I
I
31
5 15 I
79
79
I
I
I 13 28 I
I
I
31
5 15 I
I
I
21
15 I
I
I
I 13 28 I
I
I
32 74 99 2
2
2
23 41 56 3
I4
3
3
23 43 56 3
38
CO 3
3
13 13 28 3
3
3
23 43 56 3
50
49
3
3
13 13 28 3
103
105
105
105
I 06
106
106
107
107
107
107
107
107
107
107
107
108
108
108
I I 2
I I 2
I I 2
I I 5
I I 5
11 5
I I O
HO
I IO
96
96
96
97
97
97
97
97
97
99
99
99
100
135
*
*
-
O
O
450
451
4 52
4 53
454
45 5
4 56
457
45 8
459
460
461
462
463
464
465
466
467
468
469
470
4 71
47 2
4 73
4 74
475
4 76
477
4 78
4 79
480
481
482
483
484
485
486
487
488
489
490
-
10 22272
10 22272
19127]
1 9 12 70
191 2 7Q
1002272
1002272
1002272
491273
49 1273
491273
991272
99127?
991272
1001272
1001272
I 001 2 72
852371
85 2371
85 2371
921373
921373
921373
901375
90 1 3 7 5
901375
961 373
961373
961373
982372
982372
982372
92 13 71
92 13 71
92 1371
72 13 75
72 1 3 7 5
721375
85 1371
85 1 3 7 1
851371
2
3
I
2
3
I
2
3
I
2
3
I
2
3
1
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
603
5 5 4 8 2 0 64 2
104
22
I 9 2 2 2 0 5 0 2 1 03
868 10 2 4 8 2 4 64 2
I 09 112
20762258207623392215
625 1 5 7 4 8 1 5 6 7 2
113
90
I 881 21 0421 0 4 2 1 6 8 1 91 £
6 5 5 20 I 4 8 0 6 72 2
115
90
29703185338734073084
695 30 3 4 7 5 3 64 2
115
90
32053709385737364106
6 2 5 15 7 4 8 1 5 67 2
128 I 34
1 9251 199
1 7531043
8 3 0 10 I 4 8 0 8 64 2
62
56
28552000
1161
5 8 4 8 44 69 2
70
45
1935184912531591
899 10 2 4 8 4 6
2
96
45
1 83 167 I 51
I 61
1002
5 8 4840
2
91
45
112912581016
769
8 1 0 15 3 4 8 4 0
2
111
45
22201613
21562425
1161
5 8 4 8 4 4 69 2
112
56
37993775386535013617
1 0 8 2 30 8 4 8 0 9 67 2
102
45
27042339
25752613
830 10 I 4 8 0 8 64 2
101
56
373
2 100
153025 160133163
2 103 4 3 6
68
028112
216222122 I
91
71
95
23 43 56
66
45
022
4590134
3 3 4 4 39 I 721581 63
2 7 6 2 108
216324225
022
4 5 9 C 1 34
1 0 0 6 1 9 I 46 1651 64
1033115140
38
2 1 62 22
7 2 23 41
4
40
38
20
02 2
4590134
332928 161148160
264 2 I 16
21 62 32232
2 23
022
45 90134
3 3 4 4 39 I 721 581 63
2 7 6 2 I 20
21 63 24225
2 23 48
056
00 56
294326 154149145
4
61 3 5 3
69
3 1 6 3 3 4 3 3 4 6 41 10 25
67
67
70
80
404521 167160147
4
70
214
045
045
045
045
6790
90
90
I 34
I 34
33
I 99
34
44
26
203030
I 67
81 396
I 34
2 23 48 64
6 41
56
30 46
64
10 20
151
66
89
69
58
I
58
15
56
I 57
142
37 300
43
43
33
48
I
71
15
61
155154164
90
1 25 4 1
056
00 56
404521 167160147
4
92
214
022
45 90134
404127 142135159
3
93 425
79
3 1 6 3 2 2 1 3 2 5 41 18 33
71
72
69
59
75
056
00 56
I 34
6 41
10 20
2 9 4 3 2 6 I 54 1491 4 5 3 1 6 3 3 4 3 3 4 6 41 10 25
4
94 316
61
47
90
55
63
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
1
I
I
I
I 00
I 00
I 08
108
108
108
I 08
108
11 5
I I 5
I I 5
I I 6
I I 6
I I 6
I 20
120
120
61
61
61
70
70
70
81
81
81
87
87
87
90
90
90
92
92
92
93
93
93
94
94
94
136
491
492
49 3
494
495
496
49 7
498
499
50 0
501
502
503
504
505
506
507
508
509
510
51 I
512
51 3
514
515
516
517
518
51 9
52 0
521
522
523
524
525
526
527
528
52 9
530
531
532
533
534
535
536
5 37
-
535
539
-
540
541
542
-
543
544
-
545
-
54 6
-
-
550
551
-
5 52
-
5 53
554
-
555
5 56
557
-
558
559
-
560
561
562
5 63
564
-
565
5 66
567
—
-
568
569
-
570
571
5 72
-
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
26612607
26772451
951 25 6 4 8 2 8
2
98
56
49825181511348405038
1 0 8 3 15 2 4 8 1 8 58 2
104
22
16241769173616401855
I 24 2
5 I 4 7 0 9 65 2
114 134
23822258
22582398
798 00
48 34
2
106
56
185517691903
1 9 30
84 7 35 7 4831 69 2
104 I 34
213421772161
1 2 95
5 8 47 04 65 2
102
34
18821710
19192204
1161
5 8 4844 69 2
115
56
505 564 4 8 9
575
975 15 8 4834
2
114 I 34
20592204222621832107
1173
5 I 4 7 1 7 67 2
108
45
484 564 468
532
9 7 5 15 8 4834
2
117
45
171021611887
1532
975 15 8 4834
2
126
45
7 6 9 758 83 9
925
810 15 8 4 8 4 0
2
120 134
26402914
8 8 0 35 7 4831 69 2
110
56
12531285131713871441
1143
5 8 4 7 22 64 2
05 6
00 56
95
02 2
4590134
02 2
4590134
314
216222132
141 2 100
056
00 56
100
045
90
134
3 101
034112
056
213121 149146143
2 102
00
56
4 103
04 5
90
134
107
022
300
64
300
64
4590134
I 4 5 2 I 07
045
90
I 34
I 08
045
90
I 34
109
045
90
I 34
109
056
12 14 19
02 2
I 46
41
61
333833
4590134
I
I 601 571 52
56 I
I
I
5 21
15 I
I
I
21 13 25 I
I
I
I 13 28 I
I
I
21 10 20 I
I
I
21 10 25 I
I
I
41 10 20 I
I
I
I 46 66 I
48
51
I
I
41 15 30 I
I
I
I 46 66 I
48
51
I
I
I 46 66 I
I
I
I 25 41 I
I
I
21 10 20 I
61
I
I
41 28 43 I
I
95
95
95
98
98
98
I 00
I 00
I 00
I 00
I 00
I 00
101
101
101
102
102
102
103
103
103
107
107
107
107
107
107
108
108
108
109
10 9
109
I 09
109
109
I 10
HO
I 10
I I 5
I I 5
137
54 7
548
549
95 13 71
951371
95 13 71
691375
691375
691375
271372
271372
271372
97 13 71
971371
97 13 71
183372
183372
183372
261371
261371
261371
92 2 3 7 1
92 23 71
922371
141373
141373
141373
461372
461372
461372
14 2 3 7 3
142373
142373
141372
141372
141372
981372
981 3 72
981372
181371
181371
181371
451372
451372
-
-
-
-
-
-
-
451372
262371
262371
2623 71
91 I 373
91 1373
91 I 373
181372
1 8 13 72
18 13 72
1 8 23 72
18 2372
1 8 23 72
751371
751371
751371
962373
962373
962373
983372
983372
983372
2 6 13 72
261372
26 1 3 7 2
91 23 73
91 2 3 7 3
912373
1021373
1021373
1021373
1022373
1022373
1022373
1011373
10 11373
1011373
1012373
10 12373
10 12373
201 3 72
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
2
3
I
115 134
122613551484
1 295
5 8 4 7 0 4 65 2
121 112
559
39 2 446 403
1161 20 4 4844 69 2
143 134
155917851876
20 16
8 4 7 35 7 4831 69 2
129 134
2038
160718552038
8 4 7 35 7 4831 69 2
I 27
90
84910751129
I 1 46 10 8 4 7 4 0 58 2
133 112
183 296 210
220
1002
5 8 48 40
2
162
45
731 7 9 6 1 0 1 6
1113
2
8 1 0 15 8 48 40
1 39
90
22582446279629033306
1295
5 8 47 04 65 2
146 134
387
2 3 7 242 40 9
1161 20 4 4 8 44 69 2
1 73
90
278527362279
603
5 5 4 8 20 64 2
98
45
322031073113
603
5 5 4 8 2 0 64 2
97 112
160216611753
6 3 2 15 6 4 8 2 2 67 2
109 112
195721402156
6 3 2 15 6 48 22 67 2
I 10 112
24532770284528842876
034112
213121 14 9146143
2 116
2162221 I 5 5 21
10
25
045
90
I 34
403437 162157150
4 I 20 404
58
214
64
045
90
I 34
333833 160157152
3 I 21
216222224
2 21
10
04 5 90
I 34
333833 160157152
3 I 23
216222224
2 21
10 20
1 52815 I 20132139
4 I 30
21 62 22
4 31
15
028112
045
90
134
44
26 157
142
1 32 300
43
43
1 34 6 41 10 20
69
48
66
99
7
33
48
I
71
20
30
15
61
04 5
90
134
2 0 3 0 3 0 I 551 5 4 1 6 4
I 33
022
4590134
213121 14 9146143
2 144
216222115
04 5
90
4 0 34 37 162157150
4 146 404
58
214
64
045112
2 0 2 2 3 0 16 51 73151
77 2
90 254
13 51 32135
I
13
13 28
045112
20 22 30 165173151
77 2
97 287
135132135
I
13
13 28
045112
0 7 3 5 1 7 151159162
I 35 2 107 272
216324225
2 23
18
33
045112
073517 151159162
1 3 5 2 110 328
216324225
2 23
18
33
225232215
3 33
022
I 34
45901 34
23
26
1 48
1 34
I
5 21
25
41
10 25
1 34 6 41 10 20
69
4 8 66
99
13
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
3
3
3
3
3
3
3
3
3
3
3
3
3
I I 5
116
116
116
I 20
I 20
120
121
121
121
123
123
123
I 30
I 30
130
132
I 32
132
I 33
I 33
I 33
144
144
144
146
146
146
90
90
90
97
97
97
107
107
107
I I 0
I 10
I I 0
11 8
138
573
574
575
576
577
578
579
580
581
582
583
584
585
586
5-8 7
588
58 9
59 0
591
592
593
594
595
596
59 7
598
59 9
600
601
602
603
604
605
606
607
608
60 9
610
611
612
61 3
619
619
620
621
622
625
624
625
626
627
628
629
630
631
632
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
2C1372
201372
-
471373
-
471373
471373
I 201 3 8 0
I
I
I
I
2 0 1 3 90
2 0 1 3 80
202390
202380
I 20 2 3 8 0
1191380
1191380
1191380
2
1005
3
I
118
90
10671100
60
2
3
I
1 203
107
20
13
6
4726
988
4
53
9741137
4609106
027
3
2
98
81
178215891782
9 96 30 8 4 5 3 9 1 0 6
3
3
I
100
81
217621822049
3
I
2
1194
I 00
27
217121692139
5 4
4548
1194
I 192380
I 192380
1192330
401372
2
3
I
-
401372
401372
411372
411372
2
100
27
40124111413340644156
1036
5 2 4521106
3
3
I
104
134
I 7 0 4 1 581 I 785 I 7 6 3 1 8 7 1
2
411372
641375
641375
3
I
2
935
20 2 4 5 4 0 1 0 6
3
110
134
29412929294531693046
1 0 3 6 10 2 4 5 1 5 69 2
641375
441372
3
I
108
90
27692875274031182880
-
441372
-
441372
I 131376
1131376
I 131376
521373
521373
2
3
I
-
-
2
3
I
-
521373
431372
2
3
I
-
431372
2
431372
1061376
3
I
1061376
1061376
2
3
-
827
10
4
4
4616114
2
4590134
022
4590134
022
022
022
4590134
29263171329832643503
1249
35 I
4 60 0 67 2
124
135
131224135
5
56
39
15
I 63159164
4
144
34
144
13
28
12
44
13
33
163159164
4
44
13
33
I 66168159
221322122
4
44
15
30
97
474641
81
3
113
90
35053903
1181
10 4 4 5 4 9 64 2
111
54
23362411251426222576
827
10 4 4 6 1 6 1 1 4
3
112
90
120
474641
2
108
95
2
027
4 5 4 8 6 7
78
93
3
81
733
212837
139150163
43 4
89 I 78
57
253642
81
2
-
118
3
027
67
2
253642
SI
027
3
I
5
4590134
022
3
191516931882
996
30 8 4 5 3 9 1 0 6
2
117
2
166168159
221322122
4
44
4
4 4 2 3 41
103 115
15
30
99
445041
I 51168156
87
3 102 619
98
106
100
97
463340
I 73171 I I 6
63 4 103 513
87
216313224
383236
150139158
3 1 0 7 511
86
226322132
2
105
87
75
24 4 3 58
79
79
374739
168178180
I 70 4 109 267
65
216222125
2
78
72
52
24
92
94
4
80
44
30
62
46
103
46
66
054
3 3 4 1 34
67
2
118
89
4
4
89
4
4
4
4
93
93
95
95
4
4
95
97
4
97
97
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4590134
022
4590134
022
I 551 5 7 1 6 0
216212232
5
98
72
66
44
274543
I 63147164
1 2 6 4 112 112
54
216212232
33
19
2
5
24
404242
160154163
1 0 8 3 I 16 3 3 6
84
216222232
5
44
354346
148165163
I 3 8 3 1 18 2 7 7 1 5 6
216222232
69
52
2
054
4601
I I 8
3
4
4
4
I I I
422
186
81
58
60
23
41
0
5
10
15
5
4
4
18
33
10
30
I
53
I 34
24352882
1280
10 5
118
54
3
44
33
53
4
4
4
4
4
4
4
4
4
4
89
93
99
99
99
102
102
102
103
103
103
107
107
107
109
I 09
109
I I I
I I I
I I I
I I 2
I I 2
I I 2
I I 6
I 16
I I 6
I I 8
I I 8
I I 8
139
633
634
-
W
615
616
617
■T*
m
O
O
614
655
656
657
-
261471
I
203220932054
2
-
261471
261471
261472
I 295
103
658
659
-
660
-
661
6 62
665
664
66 5
666
-
261472
261472
781477
781477
5
I
2
3
I
781 4 7 7
771 4 77
2
3
I
771477
7 71 4 77
2
3
5
28
8
4704
028112
65
2
25162362240127422566
I 295
5 8 4 7 0 4 65 2
109
90
022
313634303347
I 721
5 5 4609
109
56
056112
3380
67 I
25702537279926702831
1721
5 5 4639 67 I
I 10 I 34
022
4590134
224
4590134
213121
149146143
2 I 02
216222115
213121
149146143
2 109
216222115
5
21
I o
3935
I
147153
108
216222
2
22
61
2
4039
I
132163
I 10
216222
6 1
2
2
2
5
21
10
25
102
102
7
7
2
22
25
2
102
109
109
109
108
108
108
II0
II 0
I I0
140
LITERATURE CITED
142
LITERATURE CITED
Adams, J . A. and R.H. Wilde.
1976. Variability within a soil mapping
unit mapped at the soil type level in the Wanganui district.
I.
Morphological variation. New Zealand Journ. Agric. Res. 19(2):
165-76.
Bajpai, M. R. and H . S. Mertia.
1977. A note on the irrigation and
fertility levels on the yield and nutrient uptake of dwarf barley
cultivar RDB-I. Annuals of Arid Zone 16(1): 153-6.
Barber, S. A. 1962. A diffusion and mass-flow concept of soil
nutrient availability.
Soil Science 93:39-49.
Beckett, P . H . T. 1964.
Studies on soil potassium.
II. The
immediate Q/l relations of labile potassium in the soil.
J. Soil Science 15:9-23.
Beckett, P. H. T. and M. H. M. Nafady.
1967. Studies on soil
potassium.
IV. The effect of K-fixation and release on the
form of K:(Ca + Mg) exchange isotherm. J . Soil Science 18(2):
233-43.
Black, C. A. (ed.). 1965. Methods of Soil Analysis - Part I. pgs.
381-3. American Society of Agronomy, Inc.; Madison, W i s .
Boatwright, G. O., H . Ferguson, and J. R. Sims. 1976. Soil tempera­
ture around the crown node as it influences early growth,
nutrient uptake, and nutrient translocation of spring wheat.
Agron. Journ. 68(2):227-31.
Bolton, J. 1977. Liming effects on the response of potatoes and
oats to phosphorus, potassium, and magnesium fertilizers.
.J. Agric. Sci. 89:87-93.
Burkart, N. and A. Amberger. 1978. Effect of potassium fertilization
on K-availability in potassium fixing soils over the vegetative
period.
Z. Pflangenernaehr. Bodenkd. 141:167-79.
Byrd, C. W. and D. K. Cassel.
1980. The effect of sand content upon
cone index and selected physical properties.
Soil Science
129(4):197-204.
143
Cannell, R. Q., F. B . Ellis, D. G. Christian, J. P. Graham, and J. T.
Douglas.
1980. The growth and yield of winter cereals after
direct drilling, shallow cultivation, and ploughing on noncalcareous clay soils, 1974-1978. J . Agric. Sci. 94:345-59.
Dass, G. and S. D. Shankhayan. 1979. Forms of soil potassium in acid
soils of Palam Valley of Himachal Pradesh and the response of
wheat to application of potassium.
Indian J. Agric. Sci. 49(6):
434—40.
Drew, M. C. and L. ,R. Saker. 1978. Effects of driect drilling and
ploughing on the root distributipn in spring barley, and pn
the concentrations of extractable phosphate and potassium in
the upper horizons of a clay soil. J. Sci. F d . Agric. 29:201-6.
Drew, M. C. and L. R. Saker. 1980. Direct drilling and ploughing:
their effects on the distribution of extractable phosphorous
and potassium, and of roots, in the upper horizons of two clay
soils under winter wheat and spring barley. J. Agric. Sci. 94:
411-23.
Fedak, G. and A. R. Mack.
1977. Influence of soil moisture levels
and planting dates on yield and chemical fractions in two barley
cultivars.
Can. J. Plant Science 57:261-7. .
Fisher, N. M., P. T. Gooderham, and J. Ingram.
1975. The effect on
the yields of barley and kale on soil conditions induced by
cultivation at high moisture content. J. Agric. Sci. 85:385-93.
Haby, V. A. 1975. Evaluation of soil test methods and development of
a potassium fertilizer recommendation system for Montana. Ph.D.
dissertation. Plant and Soil Science Dept., Montana State Univ.,
Bozeman.
Hodgson, D. R. ,. J. R. Proud, and S. Browne.
1977. Cultivation systems
for spring barley with special reference to direct drilling
(1971-1974). J. Agric. Sci. 88:631-44.
Hopkins, D. G. (ed.). 1979. Soil series of Montana: A key to
taxonomic classification. Montana Agric. Exp. Sta. and SCS.
Miscellaneous Report 16.
Hoyt, P. B. and. W. A. Rice. 1977 . Effects of high rates of chemical
fertilizer and barnyard manure on yield and moisture use of six
successive barley crops grown on three Gray Luvisolic soils.
Can. J. Soil Sci. 57:425-35.
144
Jankovic, V. M. and K. Nemeth.
1979. Influence of long-term ferti­
lizer trials with increasing K and P levels on K and P dynamics
in the soil and on yidld. Landwirtsch. Forsch 32(3):283-91.
Kahn, K. L. and E. W. Toop. 1970. Influence of soil temperature and
potassium fertilization on magnesium content of tomato plants.
Can. J. Plant Sci. 50:740-2.
Karlen, D. L., R. Ellis, Jr., D. A. Whiteny, and D. L. Grunes. 1978.
Influence of soil moisture and plant cultivar on cation uptake
by wheat with respect to grass tetany. Agron. Journ. 70(6);918-
21.
Kowalenko, C. G. and C. J. Ross. 1980. Studies on the dynamics of
"recently" clay-fixed NH^ using N^-*. Can. J. Soil Sci. 60:61-70.
Labanauskas, C. K., L. H. Stolzy, and R. J. Luxmoore.
1975. Soil
temperature and soil aeration effects on concentrations and total
amounts of nutrients in 'Yecora' wheat grain. Soil Science
120(6):450-4.
Lai, P . and K. C. Sharma. 1974. Uptake of N, P, and K in two
varieties of dward wheat grown at different levels of soil
moisture and nitrogen.
Indian J . Agric. Sci. 44(8):499-503.
Lai, R. 1976. No-tillage effects on soil properties under different
crops in Western Nigeria.
Soil Sci. Soc. Am. J . 40:76208.
Liebhardt, W. C. and L. Cotnoir. 1979. Potassium fertilizer
recommendations and changes in potassium soil test values as
influenced by additions of potassium. Commun. in Soil Science
and Plant Analysis 10(5):831-40.
Mack, A. R. 1971. Yield of bromegrass and removal of nitrogen,
phosphorus, and potassium under modified soil-temperature field
conditions. Can. J. Soil Sci. 51:195-309.
Massee, T. W. 1973. Soil characterization by diffusion measurements.
Ph.D, dissertation. Plant and Soil Science Dept., Montana State
Univ., Bozeman.
Mattson, S. 1973. Ionic relationships of soil, and plant.
IV.
uptake in relation to membrane activity and soil moisture.
Agricultures Scandinavica 23:11-16.
Ion
Acta
145
Mattson, S . 1974. Ionic relationships of soil and plant. V. Ion
uptake in relation to valence, membrane activity, and soil
moisture. Acta Agriculturae Scandinavica 24:33-6.
McLean, E. 0. and V. C. Bittencourt. 1974. Complementary ion effects
on potassium and calcium displacement from triionic bentonite
and illite systems as affected by pH dependent charges. 'jjpil
Science 117(2):103-9.
McLean, E. 0. 1976. Exchangeable K levels for maximum crop yields
on soils of different cation exchange capacities.
Commun. i p ,
Soil Science and Plant Analysis 7(9):823-38.
Mengel, K. and T. Aksoy. 1971. The potassium concentration of the
soil solution and its effect on the yield of spring wheat.
Z.Pflanzenernaehr Bodenk. 128(1):28-41.
Mengel, K. and B. Wiechens.
1979. Effect of non-exchangeable soil
K fraction on the yield and production of rye-grass.
Z. Pflan­
zenernaehr Bodenk. 142(6):836-47.
Munn, D . A. and E . 0. McLean.
1975. Soil potassium relationships as
indicated by solution equilibriums and plant uptake. Soil Sci.
Soc. Am. P r o c 39:1072-76.
Nie, -N. H., C. H. Hull, J . G. Jenkins, K. Steinbrener, and D. H. Bent.
1975. Statistical Package for the Social Sciences. McGraw-Hill
Book Company; New York, N.Y.
Niekerk, B. J. and J. J. N. Lambrechts.
1977. Morphological criteria
and their interpretation for agriculture in terms of soil,
behavior and land quality. Technical Comm. No. 146, p. pgs. Dept,
of Agricultural Technical Services, Republic of South Africa.
Oliveira, F. R. de., G. C . Vieira, M. Pinheiro, and J. B. Filho. 1979.
Soil temperature at a depth of 2 cm. as a response to air tempera­
ture. Revista Ceres 26(144):205-15.
Olsen, G. W.
(Chairman of Committee 6 ). 1977. Status of evaluation
of interactions between soils and fertilizer responses in the
United States and some other areas. Cornell Agronomy Memio 77-2.
Cornell Univ., Ithaca, N.Y.
146
Petinov, N. S., V. P. Ivanov, V. I. Kirillina, V. G. Golovatyi, and
K. K. Kalimulina.
1977. Dynamics of dry matter and plastic
compound accumulation in barley plants as a function of soil
moisture and mineral nutrition.
Soviet Plant Phys. 24(3):479-85.
Phillips, S . J. 1973. Potassium availability and mineralogical
relationships of selected Montana soil clays and silts. M.S.
Thesis. Plant and Soil Science Dept., Montana State Uhiv.,
Bozeman.
Raney, R. A. and G. D. C . Hoover.
1946. The release of artifically
fixed potassium from a kaolinitic and a montmorillonitic soil.
Soil Sci. Soc. Am. Proc. 11:231-3.
Raychaudhuri, S . P . 1976. Potassium fertilizers for increased crop,
production. Bulletin Indian Soc. Soil Sci. 10:306-16.
Reddy, D. S., D. R. Reddy, and G. V. N. Chary. 1978. A note on the
effect of sulphate of potash on the hydraulic conductivity,
water holding capacity, and crop yields under rainfed agriculture
at Anantapur. Annals of Arid Zone 17(2):236-8.
Samra, J. S . and N. N. Goswami. 1978. Response of wheat to soil ■
physical parameters at. varying levels of soil fertility. J.
Indian Soc. Soil Sci. 26(1):38-43.
Sawhney, B . L. 1972. Selective sorption and fixation of cations by
clay minerals: A review. Clays Clay M i n . 20:93-100.
Schaff, B . E. 1979. Influence of soil profile and site character­
istics on the response of winter wheat to K on Montana soils.
M.S. Thesis. Plant and Soil Science Dept., Montana State Univ.,
Bozeman.
Schuurman, J . J . 1971. Relation between soil density, root growth,
and uptake of minerals and water by oats. In Structure and
Fucntion of Primary Root Tissues: Proceedings of a Symposium.
Tatranska Lomnica, Czechoslavakia.
Ed. by Joseph Kolek, pgs.
351-56.
Scott, T. W. and F. W. Smith. 1957. Effect of drying upon avail­
ability of potassium to soil moisture. . Soil Sci. Soc. Am. Proc.
20:45-50.
147
Scott, A. D . , F. T. Ismail, and R. R. Locatis. 1972. Changes in
interlayer potassium exchangeability induced by heating micas.
Proc. Intr. Clay Con., Madrid, 1972. pgs. 467-79.
Sharda, A. K. and J. P. Gupta. 1975. Influence of different water
regimes on oxygen diffusion rates, nutrient uptake, growth and
yield of wheat. Annals of Arid Zone 14(4):373-5.
Sharma, R. B . and G. P. Verma.
1971. Effect of compaction of black
cotton soils on the uptake of N, P , and K by wheat.
Indian J .
Agric. Res. 5(3):165-71.
Shrader, W. D., F.l'F. Riecken, and A. J. Englehorn. 1957. Effect of
soil type differences on crop yields on Clarion-Webster soil in
Iowa. Agron, Journ. 49:254-7.
Singh, Y., R. Singh, and G. S . Sekhon, 1977. Uptake of primary
nutrients by dryland wheat as influenced by N fertilization in
relation to soil type, profile water storage, and rainfall.
J. Indian Soc. Soil Sci. 25(2):175-811.
Skogley, E. 0. 1976. Potassium in Montana soils and crop require­
ments. Montana Agric. Exp. Sta. Res. Rept. 8 8 , Bozeman.
Smith, G. D., F. Newhall, L. H. Robinson, and D. Swanson. 1964.
Soil-temperature regimes - their characteristics and predict­
ability. U.S.D.A. Bulletin, Soil Conservation Service, SCS-TP144, 14 pgs.
Sparks, D. L., D . C . Martens, and L. W. Zelayny. 1980. Plant uptake
and leaching of applied and indigenous potassium in Dothan soils
Agron. Journ. 72:551-5.
Steel, R. G. D. and J. H. Touie. 1960. Principles and Procedures
of Statistics. McGraw-Hill Book Company, Inc.
Talha, M., A. Amberger, and M. Burkart. 1979. Effect of soil
compaction and soil moisture level on plant growth and potassium
uptake.
Z . Acker-und Pflanzenbau 148:156-64.
U.S.D.A.
1951.
Soil Survey Manual.
Agriculture Handbook No. 18.
Soil Conservation Service.
148
U.S.D.A. 1975. Soil Taxonomy.
ture Handbook No. 436.
Soil Conservation Service.
Agricul­
I
V a r m a , S . K.
1976. Nutrient absoprtion under soil moisture stress
and nitrogen deficiency in wheat (Triticum aestivum). Indian
J. Agric . Res. 10(3):174-8.
Varma, S. K . , B. S . Malik, and M. C. Agarwal. 1976. Effect of
. nitrogen application under varying soil moisture regimes on the
uptake of nutrients in barley.
Indian Journ. Agton. 21(3):
318-20.
Verma, G. P . and R. B . Sharma. 1972. Effect of compaction on yield
and uptake of N, P, and K by maize (Zea mays L.). J N K W Res.
Journ. 6(1):16-20.
,
j
I
Von Braunschweig, L. - Chr. 1980. K availability in relation to
clay content - results of field experiments. Potash Review;
Commun.. by the International Potash Institute.
Subject 16, No. 2.
Wang, J. S . 1975. Potassium availability as influenced by application
rates and incubation time in Montana soils. M.S. Thesis. Plant
and Soil Science Dept. , Montana State Univ., Bozeman.
Wells, N. and M. L. Leamy. 1977. Genetic-properties of Singapore
soils and their implications for soil management. Malaysian
Society of Soil Science, Conference of Classification and
Management of Tropical Soils, Kuala Lumpur, 1977.
Wicke, H. J . 1973. On the potassium requirement of cereals. The
dependence of the potassium effect on light and temperature as
well as on soil potassium content. Archiv fur Acker - und
Pflanzenbau und Bodenkunde 17(I):55-63.
Willis, W. 0. and J . F. Power. 1975. Soil temperature and plant
growth in the Northern Great Plains. In Prairie: A multiple
view. (M. K. W ali, editor). Northern Plains Research Center,
Univ. of North Dakota Press, Mandan, N.D.
N378
V518
cop.2
DATE
118.
Veeh, R. H.
The influence of
selected soil physical
properties, soil type
and site characteristics,
soil temperature, and...
ISSUED
TO
N A W UBL
I/ C W
A
_