Retrospective Theses and Dissertations
1981
Nitrogen use efficiency of legume-grass and grass
pastures
Charles Patrick West
Iowa State University
Follow this and additional works at: http://lib.dr.iastate.edu/rtd
Part of the Agricultural Science Commons, Agriculture Commons, and the Agronomy and Crop
Sciences Commons
Recommended Citation
West, Charles Patrick, "Nitrogen use efficiency of legume-grass and grass pastures " (1981). Retrospective Theses and Dissertations. Paper
7014.
This Dissertation is brought to you for free and open access by Digital Repository @ Iowa State University. It has been accepted for inclusion in
Retrospective Theses and Dissertations by an authorized administrator of Digital Repository @ Iowa State University. For more information, please
contact [email protected].
INFORMATION TO USERS
This was produced from a copy of a document sent to us for microfilming. While the
most advanced technological means to photograph and reproduce this document
have been used, the quality is heavily dependent upon the quality of the material
submitted.
The following explanation of techniques is provided to help you understand
markings or notations which may appear on this reproduction.
1. The sign or "target" for pages apparently lacking from the document
photographed is "Missing Page(s)". If it was possible to obtain the missing
page(s) or section, they are spliced into the film along with adjacent pages.
This may have necessitated cutting through an image and duplicating
adjacent pages to assure you of complete continuity.
2. When an image on the film is obliterated with a round black mark it is an
indication that the film inspector noticed either blurred copy because of
movement during exposure, or duplicate copy. Unless we meant to delete
copyrighted materials that should not have been filmed, you will find a good
image of the page in the adjacent frame. If copyrighted materials were
deleted you will find a target note listing the pages in the adjacent frame.
3. When a map, drawing or chart, etc., is part of the material being photo­
graphed the photographer has followed a definite method in "sectioning"
the material. It is customary to begin filming at the upper left hand corner of
a large sheet and to continue from left to right in equal sections with small
overlaps. If necessary, sectioning is continued again—beginning below the
first row and continuing on until complete.
4. For any illustrations that cannot be reproduced satisfactorily by xerography,
photographic prints can be purchased at additional cost and tipped into your
xerographic copy. Requests can be made to our Dissertations Customer
Services Department.
5. Some pages in any document may have indistinct print. In all cases we have
filmed the best available copy.
UniversiW
Microllms
International
300 N. ZEEB RD., ANN ARBOR, Ml 48106
8209189
West, Charles Patrick
NITROGEN USE EFFICIENCY OF LEGUME-GRASS AND GRASS
PASTURES
Ph .D. 1981
Iowa State University
University
Microfilms
• International
m N. zeeb Road. Ann Arbor. MI 48106
PLEASE NOTE:
In all cases this material has been filmed in the best possible way from the available copy.
Problems encountered with this document have been identified here with a check mark V .
1.
Glossy photographs or pages
^
2.
Colored illustrations, paper or print
3.
Photographs with dark background
4.
Illustrations are poor copy
5.
Pages with black marks, not original copy
6.
Print shows through as there is text on both sides of page
7.
Indistinct, broken or small print on several pages
8.
Print exceeds margin requirements
9.
Tightly bound copy with print lost in spine
10.
Computer printout pages with indistinct print
11.
Page(s)
author.
lacking when material received, and not available from school or
12.
Page(s)
seem to be missing in numbering only as text follows.
13.
Two pages numbered
14.
Curling and wrinkled pages
15.
Other
. Text follows.
University
Microfilms
International
Nitrogen use efficiency of legume-grass
and grass pastures
by
Charles Patrick West
A Dissertation Submitted to the
Graduate Faculty in Partial Fulfillment of the
Requirements for the Degree of
DOCTOR OF PHILOSOPHY
Department:
Major:
Agrononw
Crop Production and Physiology
Approved:
Signature was redacted for privacy.
In Charge of Major Work
Signature was redacted for privacy.
r the Major Department
Signature was redacted for privacy.
Iowa State University
Ames, Iowa
1981
TABLE OF CONTENTS
Page
INTRODUCTION
1
Objectives
3
REVIEW OF LITERATURE
4
The Role of Nitrogen in Pasture-Livestock Production
4
Legume N vs. Fertilizer N
5
Nonsymbiotic Pathways of N Input
8
Contribution of N Fixation to the N Economy of Legume-Grass
Mixtures
9
Measuring N fixation in pastures
Factors affecting N fixation in pastures
Ecophysiology of legume-grass mixtures
Amounts and proportions of N fixed
Nitrogen and the Productivity and Quality of Cool-Season
Grasses
9
16
23
27
28
The effect of N on productivity
28
The effect of N on forage quality of cool-season grasses 29
Fate of applied fertilizer N
30
Reed canarygrass as a pasture species
32
Animal Use of Pasture N
Animal productivity on pastures
Utilization of herbage N in ruminant digestion
Fate of excreted N in pastures
The effect of N-forage system on forage quality and
N utilization
PART I.
THE CONTRIBUTION OF N FIXATION TO THE N ECONOMY
OF ALFALFA-GRASS PASTURES
34
34
35
39
41
43
INTRODUCTION
44
MATERIALS AND METHODS
47
ii i
Site Characteristics
47
Plot Establishment
47
Sample Preparation
48
Calculations of Plant Measurements
49
RESULTS
51
Annual Means
51
Seasonal Trends
51
Components of DM and N Yield
61
Correlations Among Plant Factors
67
Correlations Between Plant and Soil Factors
69
Percentage Legume and N Fixation
72
Percentage legume N fixed
Amount N fixed
Percentage herbage N fixed
DISCUSSION AND CONCLUSIONS
PART II.
THE NITROGEN USE EFFICIENCY OF ALFALFA-GRASS
AND REED CANARYGRASS PASTURES
73
77
79
81
85
INTRODUCTION
86
MATERIALS AND METHODS
89
Soil Analysis
89
Vegetative Analysis
90
Animal Analysis
91
RESULTS
93
Total Availability of Forage DM and N
93
Vegetative Responses to N Input
94
IV
Animal Responses
Return of N to the Soil
96
102
DISCUSSION AND CONCLUSIONS
106
GENERAL DISCUSSION AND CONCLUSIONS
111
LITERATURE CITED
115
ACKNOWLEDGMENTS
133
APPENDIX
134
1
INTRODUCTION
Food production from plants and animals depends on the management
of land, labor
and capital to exploit the available radiant energy,
water, nutrients and genetic potential.
Competition for land for non-
agricultural uses is causing a decline in the arable land base, while
the demand for food by an ever-growing population is increasing (Thomas,
1981).
As labor, capital and cultural inputs become more costly, the
efficiency of resource utilization per unit land area will need to be
improved to meet the growing demand for high quality food (Henzell,
1981).
This entails improvement in the conversion efficiency of environ­
mental growth factors (light, water and nutrients) into plant growth,
and subsequently into animal growth.
Concurrent with improved living
standards worldwide, there has been a growing per capita demand for
animal protein in the diet (CAST, 1980).
Since ruminants are ecologically adapted to harvesting their own
feed by grazing, animal production is possible on lands too steep, rocky,
wet, dry or erosive for production of grain crops directly consumable by
man (Hodgson, 1976).
Pastures and rangeland account for more than 22%
of the total land area worldwide (Fitzhugh et al., 1978) and 30% of the
total land area in the U.S. (Wedin et al., 1980).
Much of the area is
extensive, semi-arid to arid rangeland with low productivity per unit
area, but which is important in maintaining the breeding herds for beef
and sheep (Wedin et al., 1975).
Pastures consisting of introduced species
are found primarily in the subhumid to humid regions, where competition
2
for land use by edible grain crops is more likely to occur (Hutton, 1970).
It is in these regions that the potential for increased land use effi­
ciency through more intensive management is greatest (Hodgson, 1981).
Once a pasture is established, its perennial nature and dense growth
stabilize
the soil from wind and water erosion, allowing animal produc­
tion from land too hilly for row crops.
In a grazed grassland ecosystem,
the cycling of native or applied nutrients results in low rates of nutri­
ent loss; in fact, there can be a net increase in some nutrients.
Nitrogen (N) is a major elemental constituent of protein and nucleic
acids, both of which play crucial roles in the structure, function and
internal regulation of all organisms.
In subhumid to humid environments,
dry matter (DM) production is most limited by N availability (Black,
1968).
Of the global N potentially able to exchange with the biosphere,
99.4% is atmospheric dinitrogen, a form that is not directly assimilable
by higher plants (Burris, 1980).
Legumes can derive substantial portions of their N nutrition from
an association with Rhizobium bacteria possessing an Ng-fixing system.
Most of the N requirement of legumes can be met via symbiotic N fixa­
tion and legumes usually show little or no yield response to N fertil­
izer (Black, 1968).
There are numerous published reviews describing the N cycle in
pasture ecosystems (Henzell and Ross, 1973; Jones and Woodmansee, 1979;
Mott, 1974).
It has received the most research attention of all the
nutrient cycles because (1) a large quantity of N input is needed to
sustain a highly productive system, (2) substantial N losses can occur
from intensively managed systems, to the detriment of profit and the en­
vironment, and (3) it is the most complex of the major nutrient cycles.
Objectives
The agronomic and nutritional merits of legumes and grasses are
well documented, but little is understood of the dynamics of N fixation
in a legume-grass pasture or of the efficiency of N use by herbage and
grazing animals as affected by N source.
The objectives of this research were:
(1)
To determine the contribution of symbiotic N fixation to the N
economy of herbage available for grazing in an alfalfa (Medicago
sativa L.)-smooth bromegrass (Bromus inermis Leyss.)-orchardgrass
(Dactylis qlomerata L.) mixture and how the level of contribution
is affected by time of season, percentage legume composition and
soil factors.
(2)
To compare the N use efficiency in herbage and beef steer growth of
two N-forage systems:
(a) the above-mentioned alfalfa-grass mixture
receiving no N fertilizer vs. (b) reed canarygrass (Phalaris
arundinacea L.) receiving 180 kg/ha-yr of N.
(3)
To characterize the general N status of the two N-forage systems.
The data from this research are presented in two parts as manu­
scripts to be submitted for publication, in correspondence with the
first two objectives.
The third objective is treated in the General
Discussion and Conclusions.
4
REVIEW OF LITERATURE
The Role of Nitrogen in Pasture-Livestock Production
Most plant species must depend on soil reserves of mineral, combined
N, as nitrate (NOg") or ammonium
(Henzell and Ross, 1973).
to meet their N requirements
The total reserves of N in the volume of
soil exploitable by plant roots (4,500 to 24,000 kg/ha) is much larger
than the annual plant requirements, but 90 to 99% of it is tied up in
stable organic forms that are only slowly released (Henzell and Ross,
1973).
Because the rate of release is usually too low to sustain high
plant growth rates, N fertilizers are coimonly applied to increase
yields and obtain a profitable return (Black, 1968).
In intensively managed, temperate grassland systems, total N up­
take of 250 to 680 kg/ha annually can be achieved, resulting in DM yields
of up to 22,000 kg/ha-yr (Heath, 1976).
About two-thirds of the N
taken up is translocated to the shoot, where it is available for grazing
by livestock (Walker, 1956).
Nitrogen comprises a much larger proportion of DM in animals than
in plants (ca. 8% vs. 1 to 3%) because protein is the major nonaqueous
constituent of muscle, skin, hair, organs and blood.
A growing or
lactating ruminant has a high daily N intake requirement to meet the
tissue demands for muscle and milk synthesis (Kay, 1976).
Unlike
plants, which until senescence retain most of the N assimilated in one
growth cycle, animals are constantly excreting up to 75 to 99% of their
daily N intake (Whitehead, 1970).
3
The N input level into a grassland ecosystem, whether as combined
N or symbiotically fixed N, is a sensitive controlling agent of vegeta­
tive productivity.
For temperate grasses, applications of N fertilizer
will result in linear increases in DM yield up to a level between 100
to 300 kg/ha of N fertilizer (depending on species and availability of
HgO and other nutrients), beyond which the increases diminish and level
off (Templeton, 1976).
There is a concomitant increase in the pasture
carrying capacity^ and animal product yield per ha, although product
yield per animal is changed little, if at all (Blaser, 1964).
The N status of pasture ecosystems also influences the production
of root biomass, the rate of accumulation of plant litter on the soil
surface and the level of soil organic carbon (Whitehead, 1970).
Legume N vs. Fertilizer N
Symbiotic N fixation via legumes and N fertilizer are two sources
available to the forage manager to meet the N requirements of the crop.
Numerous assessments have been made of the relative merits of the two
N sources (Mulder, 1952; Reed, 1981; Templeton, 1976).
In summary, the
source chosen depends on fossil energy costs, adaptability of legumes
to local soil and climate conditions and the relative profitability
of alternative grass or grain crops.
The commonly cited advantages of a legume-based forage system are
(Templeton, 1976; Reed, 1981):
^Defined as stocking rate at the optimum grazing pressure (Mott,
1960).
6
(1)
More even distribution of DM productivity within a season;
(2)
Higher concentration of protein, calcium and magnesium;
(3)
Higher rate of digestion in rumen;
(4)
Lower cell wall concentration, allowing greater rate of passage
through rumen and thus a higher intake potential;
(5)
Use of a "free" and renewable resource, solar radiation, as its
energy source in the N-fixing process, whereas N fertilizer pro­
duction consumes costly and non-renewable natural gas;
(6)
Lower incidence of hypomagnesemia, alkaloid and nitrate poisoning;
and
(7)
Higher conception rates in grazing cows.
The attributes of an N-fertilized grass system are (Wedin, 1974;
Tempieton, 1976):
(1)
Greater manager control over the amount and timing of vegetative
growth;
(2)
More stable production level from year to year;
(3)
Simpler to manage for persistence and high productivity;
(4)
Higher DM yield potential than legumes except, perhaps, alfalfa;
(5)
Better adapted to low soil pH, potassium (K), and phosphorus (P)
levels; and
(6)
Generally more competitive and persistent.
The primary difference between the two N-forage systems affecting
the manager's selection is cost.
Initial establishment costs may be
higher for the legume-based system, for seed, and to ensure adequate
lime, P and K, but grass production must be sustained by high annual
7
expenditures for fertilizer N ($0.50/kg of urea-N; August, 1981).
Furthermore, the cost of N fertilizer is expected to continue increas­
ing.
When considering national and international resource allocation, the
difference in fossil energy consumption between the two systems influ­
ences the choice.
The chemical synthesis of ammonia (NHg), in which
natural gas is the primary substrate, requires an expenditure of 20 to 24
Mcal/kg of N (Slesser, 1973).
Supplies of natural gas are finite, un­
evenly distributed among nations and availability is unpredictable in
the long term.
In New Zealand and Australia, which depend largely on
external sources for meeting domestic needs of petroleum and natural gas,
legume-grass pastures form the basis of highly profitable milk, meat and
wool production systems.
There is increasing interest in the U.S. in
basing profitable grazing systems on legume-N because of the climbing
costs of N fertilizer (Tempieton, 1976) and to improve the efficiency of
fossil energy use in food production systems (Heichel, 1978).
An indirect decision on the choice of N source can be made on the
basis of forage quality.
Forage quality is defined in this disserta­
tion as the sum of the effects of nutritive value (including digesti­
bility, fermentation end-products and crude protein concentration),
intake potential, and the presence of antimetabolites (e.g. alkaloids,
nitrates) on animal performance.
At a given maturity stage, legume
forages have a higher intake potential (Van Soest, 1965) and digesti­
bility (Reed, 1981) than grasses.
This disparity is explained by the
lower concentration of legume cell wall fiber and the higher rate at
8
which legume tissue is digested and cells are ruptured.
These char­
acteristics allow a greater rate of passage of legume mass through the
rumen, thus stimulating the appetite.
The net effect is a greater flux
of digestible nutrients to the target tissues, and increased productiv­
ity.
The higher the percentage legume composition of a legume-grass
mixture, the greater will be the difference in quality between the
legume-based and N fertilizer-based forages (Napitupulu and Smith,
1979).
Nonsymbiotic Pathways of N Input
Three of the several documented pathways of N input into an unfer­
tilized pasture are (1) fallout as particulate matter or as solutes in
precipitation, (2) absorption of atmospheric NHg by soil and plants,
and (3) nonsymbiotic biological N fixation (Jones and Woodmansee, 1979).
Ammonia loss from industrial centers and large feedlots can augment
the N fallout on nearby pastureland, but the influx rarely exceeds
30 kg/ha-yr (Henzell and Ross, 1973).
In six Iowa locations, Tabatabai
and Laflen (1976) reported annual N fluxes from precipitation of 10 to
14 kg/ha, equally divided between NO^'-N and
Soil sorption of atmospheric ammonia was estimated by Hanawalt
(1969) to attain up to 74 kg/ha-yr in New Jersey, where atmospheric
NHg concentrations were three times the world average.
Nonsymbiotic N fixation generally contributes less than 20 kg/ha-yr
in temperate pastures, for which blue-green algae are mainly responsible
(Lockyer and Cowling, 1977).
9
Contribution of N Fixation to the N Economy of
Legume-Grass Mixtures
Most estimates of quantities of N fixation under field conditions
are based on legume monocultures.
Most legume-based pastures, however,
are managed as mixtures including at least one grass species.
Thus, N
fixation rates and proportions of legume N derived from fixation can be
affected by the botanical composition and by competition for light,
water, and nutrients (Vallis, 1978).
Measuring
fixation in pastures
In order to adequately assess the factors affecting the contribu­
tion of N fixation to a pasture, methods of measurement must be employed
that are practical and accurate.
use, acetylene reduction and
Two methods that have come into field
isotope dilution, will be reviewed
here as they relate to pasture conditions.
In-depth comparisons of N-
fixation methodologies are given by Hardy and Hoi sten (1978), Bremner
(1977), and Burris (1974).
Nitrogenase is the enzyme system in effective legume nodules that
catalyzes the reduction of Ng to NHg, which is subsequently assimilated
into amino acids (Burris, 1975).
Oilworth (1966) was the first to
report that nitrogenase also catalyzes the reduction of acetylene
(CgHg) to ethylene (CgH^), in a molar equivalent ratio of 3:1, com­
pared to Ng reduction.
Hardy et al. (1968) proposed the routine use
of acetylene reduction as a qualitative and quantitative tool in measur­
ing N fixation by virtue of its sensitivity, the low cost and wide
availability of acetylene and gas chromatographs, its universal
10
adaptability to field and laboratory conditions, and the relatively low
level of technical skill and equipment sophistication required.
As a
result, a large number of samples can be rapidly analyzed at a low unit
cost.
In practice, the method has enjoyed extensive use in pasture
research.
Seetin and Barnes (1977) and Smith et al. (1981) reported
progress in selecting for higher N fixation capacities in alfalfa
(Medicago sativa, L.) and crimson clover (Trifolium incarnatum L.),
respectively, on the basis of acetylene reduction.
Halliday and Pate
(1976) investigated the effects of temperature, shading, and defolia­
tion on N fixation capacity of white clover (Trifolium repens L.) and
estimated 268 kg/ha-yr at 55% legume content.
Sinclair (1973) proposed
a non-destructive method of in situ assay of nitrogenase activity in
white clover, using acetylene reduction.
Briefly, the method entails incubating nodulated root mass (soilfree or as intact soil core) in an air-tight container with an atmospher­
ic acetylene concentration of 10% (V/V), usually at room temperature
for 1 to 2 hours.
Gas samples are periodically withdrawn and the
ethylene content determined by gas chromatography.
as wg ethylene produced per hr on a plant,
Results are reported
core, or nodule-mass basis.
The results are often converted by a molar ratio (determined for the
specific conditions) to the amount of Ng fixed per unit time on an
area basis (Hardy et al., 1968).
The main drawbacks to this method are that:
n
(1)
It is an indirect measurement of N fixation, and the molar ratio
used for conversion to actual N fixation rate varies widely with
species, water status (Sinclair et al., 1978), and duration and
temperature of incubation (Goh et al., 1978).
Furthermore, it
measures total electron flux potential of the nitrogenase system,
including hydrogenase activity (H2 evolution), a process that com­
petes with nitrogen fixation for electron flux (Schubert and Evans,
1C
1976). Therefore, calibration is required with
Ng to establish
the unique conversion factor for an experimental situation, which
most potential users are ill-equipped to do (Bremner, 1977).
(2)
The method measures nitrogenase activity at a point in time, which
is strongly regulated by diurnal fluctuation, irradiance, temper­
ature, physiological status of the plant (Goh et al., 1978), and
the plants' exposure to animal excretion (Ball et al., 1979).
Therefore, many repeated measurements on the same experimental unit
must be performed throughout the plants' growth cycle, which in­
creases the cost of obtaining the desired information.
The inte­
gration of these points to estimate seasonal activity is subject
to considerable error (Goh et al., 1978).
(3)
The tremendous natural variability in plant size, vigor, root
morphology, modulation and nodule effectiveness, compounded by the
natural variability in soil characteristics and spatial distribu­
tion of legume plants in a pasture, require that a large number of
subsamples be taken (Goh et al., 1978).
Indeed, the resulting
high coefficients of variation often prevent the detection of
12
significant treatment differences.
(4)
Acetylene reduction does not allow sensitive partitioning of
herbage N into soil- and atmosphere-derived N fractions.
(5)
This method normally involves repeated, destructive sampling of
plants from a field plot, eventually depleting the stand.
In summary, it appears that the acetylene reduction method is best
suited to preliminary screening of plant genotypes for N fixation poten­
tial (Seetin and Barnes, 1977), and for determining the qualitative
effects of controlled environment conditions on nitrogenase activity
(for example. Ball et al., 1979; Oral le and Heichel, 1978).
A
isotope dilution method was developed for field use in
quantifying seasonal N fixation rates and the proportion of legume N
derived from the soil (McAuliffe et al., 1958; Vail is et al., 1967).
It involves the application of a small amount of N compound, highly
enriched with the stable isotope
to the test plot soil at least
1 week before the plant growth cycle begins, to allow mixing with the
microbial N pool.
The degree of dilution of
the legume tissue by atmospheric
15
N-labelled soil N in
compared to the
enrichment
of a nonfixing reference crop, grown either in mixture or in adjacent
plots, is a measure of the proportion of legume N derived from fixa­
tion.
In theory, the product of that proportion and the amount of
legume N equals the total amount of N fixed (McAuliffe et al., 1958;
Fried and Middleboe, 1977).
Val lis et al. (1967) have modified the
technique slightly to measure within-season transfer of legume N to an
associated grass.
Several assumptions are made in the use of the dilution technique
(Edmeades and Goh, 1978; McAuliffe et al., 1958; Vallis et al., 1967):
(1)
The fixing and nonfixing crops take up soil
and
in the
same average ratio;
(2)
There is no N fixation occurring in the reference crop;
(3)
The nitrogenase enzyme system does not discriminate between
and
(4)
No transfer of N from the fixing to the nonfixing plants occurs
during the period of measurement;
(5)
The net NHg exchange rates between the atmosphere and foliage are
similar for the fixing and nonfixing plants.
Not all of these assumptions have been verified, especially the
fifth assumption.
Errors resulting from the first four assumptions
appear minor compared to levels of N fixation and soil N uptake encoun­
tered in tests (Vallis et al., 1967).
McAuliffe et al. (1958) were the first to use this dilution method
in a legume-grass mixture (the grass served as the reference crop) in
their study of the effect of N fertilizer on N fixation in Ladino
clover (Trifolium repens L.) and tall fescue (Festuca arundinacea,
Schreb.).
Vallis et al. (1967) refined the method for evaluating the
competition between legume and grass for soil N.
Vallis et al. (1977)
made use of the technique to determine the effect of botanical composi­
tion of several legume-grass mixtures on their N economies.
Edmeades
and Goh (1978) studied the effect of pasture age and botanical
14
composition on N fixation contribution.
This dilution technique is cur­
rently the basis for the University of Minnesota program of breeding
for enhanced N fixation in alfalfa (Heichel et al., 1981).
The advantages of the dilution method are that:
(1)
Since sampling occurs at the normal harvest time, the N fixation
rate is automatically integrated over time, environmental fluctua­
tions, and soil and spatial heterogeneity (Heichel et al., 1981).
(2)
It allows quantitative measurement of both the total amount of N
fixed and the proportion of soil-derived legume N, provided the
assumptions are valid.
This allows detailed investigation of the
N economy of pasture mixtures (Vallis et al., 1967).
(3)
Error and variability are relatively low, allowing significant
differences to be detected from smaller plots, requiring less
replication and fewer samples (Goh et al., 1978).
(4)
The method allows repeated nondestructive sampling of plants if
only the shoots are harvested, as is the practice with pasture mix­
tures, thus allowing the study to continue for several years.
(5)
The procedure can be carried out with minimal physical disturbance
of the soil-plant system.
The disadvantages are:
(1)
Cost of ^^N-labelled material is high, currently at $US 99.00/g of
actual
However, only trace amounts are normally needed in ex­
periments, and the cost is small relative to the labor cost in
sample preparation (Bremner, 1977).
(2)
Preparation of samples for
analysis is time-consuming, tedious.
15
and requires skill in avoiding losses and contamination (Bremner,
1977).
(3)
A mass spectrometer is generally required for isotope ratio analysis.
The mass spectrometer is
expensive to purchase and maintain, and
requires a skilled operator, thus they are not widely available
(Bremner, 1977).
In summary, this
15
N-dilution method is generally the preferred
but less common way to measure the contribution of N fixation to legumegrass mixtures (Burris, 1980; Goh et al., 1978; Haystead and Lowe,
1977; Heichel et al., 1981).
The discovery that the
15
N enrichment of total soil N is usually
slightly higher than that of atmospheric N (Cheng et al., 1964) has led
some workers (e.g., Amarger et al., 1979; Rennie et al., 1976) to propose
using differences in natural abundance of
in legume and grass com­
ponents as a measurement of symbiotic N fixation.
This form of
dilution is considered inadequate to serve as a method of quantifying N
fixation (Bremner, 1977), since
enrichment of total soil N varies
with soil type and depth, and that of mineralized N can vary signifi­
cantly with time (Bremner and Tabatabai, 1973).
The term, ^^N-dilution method, will henceforth be used in this dis­
sertation to signify the technique whereby the soil N pool is artificial­
ly enriched with ^^N-labelled mineral N compounds as used by Vallis et
al. (1967).
"16
Factors affecting ^ fixation in pastures
The amounts and proportions of legume N derived from N fixation
vary widely with species, environment and management by exerting their
effects on the Rhizobium bacteria, the nodulation process, and on N
fixation per se (Nutman, 1977).
Combined N
It has often been observed that applying N to
established legumes does not result in a response in yield or N uptake
(Whitehead, 1970).
Alios and Bartholomew (1955, 1959) applied varying
levels of combined N to six legume species in pot culture and observed
that the proportion of legume N from fixation decreased as N level in­
creased.
Shoot DM yields, however, increased linearly with N level.
At a low level of applied N, the proportion of N fixed in alfalfa,
sweet clover (Melilotus spp.) and Ladino clover, was slightly lower than
the check treatment, but the amount of N fixed was at a maximum.
suggests a stimulatory effect at a low N level.
bined N lowered the amounts of N fixea.
This
Higher levels of com­
Alios and Bartholomew (1959)
concluded that under N-limiting conditions, combined N can add to the
legume N uptake and stimulate N fixation, but under nonlimiting condi­
tions, combined N replaces rather than supplements N fixation.
phenomenon has been verified by using acetylene reduction and
This
15
N-
tracing techniques.
Nitrate has been shown to inhibit nodulation at high levels (Gibson
and Nutman, 1960; Dart and Mercer, 1965; Thornton and Ni col, 1938).
Nitrate sometimes stimulated nodulation at low to intermediate levels
(Dart and Mercer. 1965; Dart and Wildon, 1970; Oghoghorie and Pate,
17
1971).
This effect showed an interaction with Rhizobium strain in
alfalfa (Heichel and Vance, 1979).
Using the acetylene reduction assay, nitrogenase activity decreased
to 6% of controls in red clover (Trifolium pratense L.) exposed to 112
ppm NOg-N in pots (Hojjati et al., 1978) and to 25% in white clover
receiving 90 kg/ha of N (Moustafa et al., 1969).
Using ^^N-tracing,
similar decreases in N fixation rates were observed in peas by Oghoghorie
and Pate (1971), in alfalfa and Ladino clover (McAuliffe et al., 1958)
and in white clover (Walker et al., 1956).
Applying N to legume-grass mixtures may or may not increase DM
yield, but almost invariably results in a decrease in the legume com­
ponent (Walker et al., 1956; Whitehead, 1970).
Although combined N can
directly decrease the amount and proportion of N fixed by legumes
(McAuliffe et al., 1958), the deleterious effect of N fertilizer on per­
centage clover is
considered to be indirect by increasing the competi­
tion for light and nutrients from the associated grass (Carter and Scholl,
1962; Dilz and Mulder, 1962; Donald, 1963; Linehan and Lowe, 1960;
MacLeod, 1965; Mouat and Walker, 1959a; Robinson and Sprague, 1947).
Both the direct and indirect effects of combined N can decrease the con­
tribution of N fixation to the total N yield of a legume-grass mixture.
The advantages to plant vigor and DM yield of supplying low to
moderate levels of combined N have been shown in grain legumes (Dart and
Wildon, 1970; Harper, 1974; Oghoghorie and Pate, 1971), but little is
understood of the use of combined N by established legumes in legumegrass mixtures (Vallis, 1978).
18
Dry matter yield increases resulting from N fertilizer have been
reported in mixtures containing white clover (Linehan and Lowe, 1960;
Reid, 1970), alfalfa (Carter and Scholl, 1960; Chan and MacKenzie, 1971;
West et al., 1980), red clover (Carter and Scholl, 1962; West et al.,
1980), and birdsfoot trefoil (Lotus corniculatus L.) (West et al., 1980).
Woodhouse and Chamblee (1958) and Gibson (1977) suggested that the
availability of combined N during legume seedling establishment would
enhance seedling vigor and secondary root growth, eventually leading to
enhanced nodulation, N fixation and competitiveness in the mixture.
Depending on the legume species and degree of competition from the asso­
ciated grass. West et al. (1980) showed that legume yield and percentage
legume can actually be increased by moderate rates of N fertilizer at
establishment.
The tactical use of fertilizer N on established white
clover-grass stands under N stress was observed to improve DM yield
without decreasing the legume component (Field and Ball, 1978; Young,
1958).
Proportion of legume in the mixture
Although it would seem
reasonable that N fixation on an area basis would vary widely with the
proportion of legume in the sward, very little has been published on
how this variable affects the total amount and proportion of N fixed.
Field measurements of N fixation are often reported as an average across
a range of legume percentages, or the legume content is often not even
mentioned (Reed, 1981).
Peterson and Bendixen (1961) reported Ladino clover N yields of 0,
188, and 356 kg/ha at 0, 58, and 100% legume, respectively.
Although
19
N fixation was not measured in this study, it probably increased on a
land area basis with increase in percentage clover.
Since the percentage legume is affected by available soil or
fertilizer N, the two parameters can be confounded in their effects on
quantity and proportion of N fixed.
In pot experiments. Walker et al.
(1956) reported a decrease in the percentage white clover associated
with Italian ryegrass (Lolium multiflorum. Lam.) by applying ^^Nlabelled fertilizer, with a resulting decrease in the amount of N
fixed and in the proportion of clover N derived from the soil.
When exogenous N was not added (Vallis et al., 1967), the in­
clusion of Rhodes grass (Chioris gayana Kunth.) in association with
Townsville stylo (Stylosanthes humilis H.B.K.) decreased the amount of
N fixed per pot, but increased the proportion of legume N fixed as com­
pared to the legume grown alone.
Edmeades and Goh (1978) found the
proportion of N fixed in white clover to decrease from 88 to 82% as the
percentage clover increased from 20 to 46%.
Vallis et al. (1977) showed
that increasing the percentage legume in various combinations with
tropical grasses resulted in the legume accounting for a greater propor­
tion of the soil-derived herbage N.
In a California winter-range
environment, Phillips and Bennett (1978) detected declines in the pro­
portion of N fixed in Tri folium subterraneum L. as percentage legume
increased from 50% (grown with Bromus mollis L.) to 100%.
Summarizing
a series of N fixation studies in New Zealand pastures, Hoglund et al.
(1979) concluded that lower annual N fixation rates were associated with
the lower legume yields that were found in pastures low in percentage
20
legume and high in soil N availability.
Effect of temperature, light and water
Since N fixation is a
complex process involving the physiology and genetics
of both the legume
and the Rhizobium bacteria, any environmental factor affecting the vigor,
persistence, or function of either will affect the rate of N fixation
(Gibson, 1977).
The biological reduction of Ng to NHg is an energy-
intensive process depending on an adequate supply of carbohydrate and
reducing equivalents in effective nodules (Hardy and Havelka, 1976).
With photosynthetic assimilates as the source of this energy, environ­
mental factors inhibiting photosynthetic rate and assimilate transloca­
tion to the nodules can inhibit N fixation (Hardy and Havelka, 1976).
Temperature affects N fixation directly at the nodule site and
through its effect on plant growth and photosynthetic rate, in that N
fixation exhibits seasonal and diurnal variations corresponding to
fluctuations in soil and air temperatures (Eckart and Raguse, 1980).
Temperature optima generally fall between 20 and 35°C depending on the
species (Gibson, 1977).
Light intensity and duration affect modulation and N fixation rates
(Gibson, 1977).
McKee (1962) found that shading alfalfa, red clover
and birdsfoot trefoil (either artificially or by associated grasses)
reduced nodulation, with the effect more pronounced in birdsfoot
trefoil.
Day and Dart (1969) reported increased legume growth, N
fixation and nodule mass in seven legumes with higher light intensity.
In the field, the effects of light are often confounded with tempera­
ture (Gibson, 1976).
21
Water stress affects practically all physiological processes in
plants (Hsiao, 1973).
Soil moisture levels detrimental to plant growth
generally inhibit N fixation (Gibson, 1977).
Engin and Sprent (1973)
demonstrated direct inhibitory effects of water stress on N fixation
in white clover, with recovery depending on the degree and duration of
stress.
White clover growth and N fixation responded positively to
seasonal rainfall across various New Zealand pasture sites (Hoglund et
al., 1979).
Hoglund et al. (1979) also observed that with progressive
surface drying a greater proportion of N fixation activity occurred at
lower soil depths.
Defoliation
Defoliation of pasture plants removes leaf area
and arrests the flow of carbon assimilates from the shoot to the nodules.
More frequent and intensive defoliations result in DM yield reduction
and lower nodule number (Harris, 1978).
Cutting results in nodule loss
in white clover, with the effect enhanced at greater cutting frequencies
(Wilson, 1942).
Langille and Calder (1971) reported greater nodule loss in birdsfoot trefoil when cut to 2.5 cm rather than 7.5 cm, and when cut more
frequently.
Moustafa et al. (1969) detected rapid declines in acetylene
reduction rates in white clover when completely defoliated, with recovery
trends similar to regrowth patterns.
Defoliating alfalfa at early bloom
caused an 88% decline in nitrogenase activity, but nodules remained
intact, showing only a senescing region which was regenerated upon shoot
regrowth (Vance et al., 1979).
Cralle and Heichel (1978) reported that
complete shoot removal inhibited acetylene reduction more than
22
partial shoot removal in both alfalfa and birdsfoot trefoil.
It is generalized that defoliation inhibits N fixation by restrict­
ing the supply of carbon assimilates to the nodule (Lawn and Brun, 1974),
but decreased host plant demand for N is also proposed (Hoglund and
Brock, 1978).
Animal factors
The grazing animal influences plant growth by
selective grazing, defoliating, trampling, defecating, urinating,
camping, and redistributing nutrients (Watkin and Clements, 1978).
The
effects of simple defoliations on N fixation have been studied under con­
trolled conditions (Vance et al., 1979), but the effects of repeated de­
foliations at different intensities by grazing animals have not been
directly measured.
Brock et al. (1981) tested the effect of rotational
grazing versus continuous grazing on annual N fixation rates in white
clover-perennial ryegrass pastures in New Zealand.
Using acetylene
reduction, they estimated 10% higher N fixation rates in the continuously
grazed system, with the difference occurring during the dry season.
Lowering the herbage allowance per animal (i.e. higher grazing
pressure) results in greater frequency and severity of defoliation
(Brougham et al., 1978).
Thus, a higher grazing pressure could adversely
affect N fixation in pastures, but such a study has not been reported.
Increased grazing pressure can enhance or deplete the legume content of
a mixture, depending on the legume's growth habit (Brougham et al.,
1978), which can indirectly affect legume yield and N fixation.
Defecation on pasture plants renders them unpalatable due to odors,
thereby protecting them from defoliation (Marten and Donker, 1966).
23
Defecation and urination concentrate essential nutrients, notably com­
bined N, into small areas which can stimulate grass growth and reduce
N fixation (Watkin and Clements, 1978).
Applying urine-N to white
clover-ryegrass plots at the rates of 300 and 600 kg/ha decreased N
fixation rates (acetylene reduction) to 16 and 6% of the control,
respectively, and decreased the proportion of legume N from fixation
(Ball et al., 1979).
The direct effects of grazing habit and nutrient recycling on
pasture N fixation are poorly understood.
Ecophysioloqy of legume-grass mixtures
Competition
Using the strict definition of plant competition by
Donald (1963), where the combined demands of associated species for a
limited resource cannot be met, legume growth and N fixation, when asso­
ciated with grasses, are influenced largely by competition both above
ground and below ground (Hall, 1978).
McKee (1962) noted decreased nod-
ulation in alfalfa, red clover, and birdsfoot trefoil with increased
competition for light.
Grasses are generally considered to be more com­
petitive for soil N than legumes (Vallis, 1978).
However, white clover,
alfalfa (Simpson, 1976), and the tropical legumes Desmodium intortum
(Hall, 1978) and Macroptilium atropurpureum (Vallis et al., 1967)
have been reported to compete with the associated grass for soil N.
Hall (1974) described the interacting effects of competition for
soil N and K and percentage legume on herbage DM yield and legume N
yield.
Under a K-limiting situation over a range of percentage legume.
24
the legume and grass species showed no relative advantages in terms of
DM yield, K and P yield, with a modest relative N yield advantage
between 50 and 95% legume.
With K nonlimiting, DM, K, P and N yields
all showed dramatic relative yield advantages between 25 to 95% legume.
White clover is a poor competitor for P (Mouat and Walker, 1959a),
which Brougham et al. (1978) attribute to the shortness and infrequency
of root hairs, thus exploiting a smaller effective soil volume relative
to grasses.
Drake et al. (1951) and Mouat and Walker (1959b) attribute the
lower competitiveness of alfalfa and white clover for
to the higher
cation exchange capacity of legume roots, resulting in their being more
apt to take up divalent cations.
Thus, the associated grass is more
competitive at low soil K levels (Drake et al., 1951).
The competitive ability of legumes for available water depends on
legume species and the associated grass species (Chamblee, 1972).
Alfalfa has a deep root system which is effective in extracting water
from great soil depths during drought periods, while the growth rate of
the associated cool-season grass slows considerably (Chamblee, 1972).
Water stress at shallow depths can change the soil N uptake patterns of
both the legume and the grass, but there is very little information on
these interactions in legume-grass mixtures.
In contrast to alfalfa,
white clover has a shallow root system and does not compete well with
grasses during dry periods (Brougham et al., 1978).
Management factors that increase the competitiveness of a legume in
a mixture, such as frequent or severe defoliation (not necessarily both)
25
(Harris, 1978), P and K fertilization (Chamblee, 1972), and cutting
instead of grazing (Munro and Davies, 1974), can increase the amount of
N fixation but the effects on the proportion N fixed are unknown.
Complementation
Although there is considerable overlap of
regions of exploitation by legumes and grasses in mixtures with respect
to light utilization, P and K nutrition, and water uptake, there are
several reports (DanCik, 1976; McCloud and Mott, 1953; Roberts and
Olson, 1942)
of alfalfa-grass mixtures yielding more DM than either
component in monoculture.
Depending on the age of the stands, McCloud
and Mott (1953) showed increased DM yields of alfalfa mixtures contain­
ing smooth bromegrass, Kentucky bluegrass (Poa pratensis L.) and timothy
(Phleum pratense L.).
In contrast, Vartha (1973) reported similar
yields of pure legume and legume-grass stands.
The inclusion of a legume in a grass stand results in greater N
uptake in the latter compared to the grass grown alone (Vailis, 1978).
It is assumed that the source of this additional N is symbiotically
fixed N recently released from the legume (Whitehead, 1970).
The trans­
fer of fixed N to a grass in pastures can occur via (1) the release of
soluble N compounds from the living legume plant, (2) the decomposition
of senesced legume nodules, roots and shoot tissue and (3) the return
of ingested legume N as animal excreta (Vallis, 1978).
The exudation of soluble N compounds from live legume roots was
detected in greenhouse experiments, but a significant grass response
occurred only under long, cool days with low light intensity (Wyss and
Wilson, 1941), conditions which do not favor high N fixation rates
26
(Gibson, 1977),
exudation
Other pot experiments indicated trace amounts of N
from intact legume roots (Vallis et al., 1967; Whitney and
Kanehiro, 1967).
Direct transfer of exuded legume N to an associated
grass has not been shown experimentally in the field (Henzell, 1970).
Simpson (1976) correlated the root tissue senescence patterns of
three legumes with the N uptake response of orchardgrass grown in mixture
in field plots.
In the first year, white clover, which continually
sloughs off roots and nodules during the growing season, elicited the
greatest response in N uptake in orchardgrass.
In subsequent years,
subterranean clover and alfalfa, which do not release tissue N until the
end of the growing season, elicited greater responses.
Haystead and Marriott (1978, 1979) presented evidence of substan­
tial delay of the grass response to sloughed white clover nodules due
at least in part to temporary immobilization of released N by bacteria.
Indeed, Stewart and Chestnutt (1974) and Simpson (1976) observed that
the response in associated grass N levels to legumes correlated better
with the previous season's legume growth than that of the current season.
Using the ^^N-dilution technique, Vallis et al. (1967, 1977) and
Haystead and Lowe (1977) detected little or no N transfer from legume to
grass, although N uptake by the grass in mixture significantly exceeded
that of the grass grown alone.
These discrepancies illustrate the
limitations of current techniques of quantifying complex N transforma­
tions (Henzell, 1970).
Haystead and Lowe (1977) suggested that the
discrepancy may be explained by the possibility that grass growing with
legumes is in some way stimulated to take up more soil N.
27
Substantial legume N transfer may occur via consumption by animals
and return of excreta, since only 5 to 25% of ingested N is retained
(Whitehead, 1970).
Shaw (1966) reported 14 to 50% increases in the
grass N uptake in grazed versus clipped plots, although actual N
transfer was not demonstrated.
Since excreted N is distributed very
unevenly and is subject to large losses (Ball et al., 1979), N transfer
via animal excretion may be of questionable efficiency (Henzell, 1970;
Whitehead, 1970).
Amounts and proportions of ^ fixed
Quantities of N fixed reported for forage legumes vary widely with
species, age
and environmental conditions (Nutman, 1976).
Reports on
alfalfa range from 56 to 463 kg/ha-yr (aver. 199) and on clovers from
45 to 673 kg/ha-yr (aver, 183) (Nutman, 1976).
These ranges were
derived from experiments using widely different techniques, most of
which do not distinguish between soil-derived and atmosphere-derived N.
Using acetylene reduction in white clover-based pastures, Halliday
and Pate (1976) in Northern Ireland estimated 268 kg/ha-yr of N fixed
(55% clover), Haystead and Lowe (1977) in England estimated 100 kg/ha-yr
(40% clover), Hoglund and Brock (1978) reported 184 (35% clover) and
231 kg/ha-yr (68% clover) in New Zealand.
Hoglund et al. (1979) sum­
marized surveys of N fixation at nine New Zealand sites showing a range
of 107 to 392 kg/ha-yr of N fixed (aver. 184).
Using the ^^N-dilution technique, Edmeades and Goh (1978) reported
N fixation rates of 45 to 142 kg/ha-yr in New Zealand, depending on the
proportion of white clover and on stand age.
The proportion of fixed N
28
in the herbage was 81 to 93%.
In pure alfalfa stands Heichel et al.
(1981) detected 148 kg/ha-yr of N fixed during the seeding year in
Minnesota, of which 43% was from fixation.
In another trial of Heichel
et al. (1980), alfalfa, red clover and birdsfoot trefoil fixed 158,
119, and 106 kg/ha-yr, respectively, in the seeding year, and 170, 99,
and 75 kg/ha-yr, respectively, in the second year.
The corresponding
proportions of N fixed were 60, 60, and 40% in the seeding year, and
44, 35, and 30%, respectively, in the second year.
The proportion N fixed was found to vary widely with time of year
in alfalfa (Heichel et al., 1980, 1981; Elgin and Legg, 1977), red
clover and birdsfoot trefoil (Heichel et al., 1980), and with alfalfa
genotype (Elgin and Legg, 1977; Heichel et al., 1978).
Nitrogen and the Productivity and Quality of
Cool-Season Grasses
The effect of
on productivity
Cool-season perennial grasses generally require large amounts of N
fertilizer to attain maximum yields, although economic N input levels
may be less than that level giving maximum yield (Wedin, 1974).
Dry
matter productivity generally responds linearly to N fertilizer up to
levels between 200 to 300 kg/ha (Reed, 1981), with the response sometimes
being higher with split applications (Whitehead, 1970).
At N input rates
up to 200 kg/ha, the response in DM yield increase per unit N applied
varies from 15 to 35 kg (Reed, 1981); however, this efficiency ratio
steadily declines with higher N input levels.
Cool-season grass species differ in their yield response to N
29
fertilizer (George et al., 1973; Schmid et al., 1979) and these dif­
ferences are not consistent across environments (Schmid et al., 1979).
The periods of greatest response to N fertilizer in cool-season grasses
correspond to the seasons of highest growth rate (Wedin, 1974), presum­
ably because N demands are also highest then (Andrew
and Johansen,
1978).
In most species this occurs in spring and early summer (Wedin,
1974).
Tall fescue exhibits an additional period of higher growth rate
and response to N in the fall compared to smooth bromegrass, orchardgrass and reed canarygrass (Wedin et al., 1966).
Under irrigation in Oregon, reed canarygrass and orchardgrass had
more uniform seasonal yield distributions than tall fescue and perennial
ryegrass (Yungen et al., 1977).
At the highest N fertilizer rate
(315 kg/ha.yr), tall fescue yielded the most DM (15.5
by reed canarygrass (14.4
t/ha).
t/ha), followed
Reed canarygrass, however, had the
highest N concentration (2.8%) and N yield (490 kg/ha).
Tall fescue and reed canarygrass exhibited bimodal growth patterns
through the growing season in Minnesota (Schmid et al., 1979), with the
maxima in June and September, while orchardgrass yield was relatively
stable and smooth bromegrass declined steadily in yield after June.
The effect of ^ on^ forage quality of coolseason grasses
Application of N on cool-season grasses alters the relative amounts
of highly digestible constituents such as protein, nonprotein nitrogen,
starch, and sugars (Blaser, 1964; Decker et al., 1967), but does not
consistently result in changes in the forage digestibility or intake
30
potential (Decker et al., 1967; Krueger and Scholl, 1970; Schmid et al.,
1979).
Blaser (1964) reported that N fertilization increased the carry­
ing capacity and live weight gain per ha from pure grass pastures, but
output per animal was unaffected.
Various authors have detected little
or no effect of N fertilizer on forage digestibility over a wide range
of N rates (Blaser, 1964; Blaxter et al., 1971; Niehaus, 1971; Krueger
and Scholl, 1970; Schmid et al., 1979).
Although N fertilizer has been shown to decrease voluntary intake
by sheep on grass pasture (Reid et al., 1970), most studies detect no
effect of N on DM intake (Reed, 1981; Reid and Jung, 1965).
Nitrogen
applied to reed canarygrass can increase the total alkaloid content,
resulting in reduced palatability and sometimes intake (Marten et al.,
1974).
Nitrogen amendments almost invariably increase the crude protein
(N X 6.25) content of grasses up to levels equal to or exceeding those
of legumes (Whitehead, 1970) but crude protein percentage is not a
reliable indicator of overall forage quality (Marten and Hovin, 1980).
Fate of applied fertilizer N,
Upon application to a grassland, fertilizer N can undergo a variety
of processes that affect its availability to plants (Black, 1968; Heal,
1979; Henzell, 1970; Henzell and Ross, 1973; Woldendorp et al., 1966).
From the pool of soluble N, NH^ can be adsorbed onto the exchange complex
of clays, immobilized by clay fixation, immobilized by microflora and remineralized; N can volatilize as NHg, or be oxidized to NO3", during which
31
gaseous loss as nitrous oxide can occur (Bremner and Blackmer, 1978).
Nitrate can be leached below the zone of root uptake or denitrified to
volatile forms.
One commonly used index of the efficiency of N fertilizer use by
grasses is the measurement of the harvested N yield minus the N yield
of the nil-N treatment, expressed as a percentage of N applied.
Such
"apparent recoveries" seldom exceed 60% (Woldendorp et al., 1966).
15
The use of
N-tracing techniques often results in lower values than the
difference method (Allison, 1966; Hauck and Bremner, 1976).
Efforts to
account for the balance of applied N are limited by the complexity of N
transformations in the soil (Allison, 1955).
Allison (1955) and Walker et al. (1956) suggested that losses of N
from pastures via leaching and devitrification are small at low N input
levels, but that these losses are greatly enhanced as the level of N in­
put increases.
Dilz and Woldendorp (1960) could not account for 19 to
40% of ^^NOg" applied to sandy, clay, and peat soils.
It was assumed
that denitrification accounted for most of that loss, although denitrification was not measured.
Woldendorp et al. (1966) used lysimeters in the Netherlands con­
taining a fine, sandy soil to measure NOg" leaching under a Festuca
pratensis sward receiving 330 kg/ha-yr of N as ammonium nitrate.
One to
16% of the applied N was recovered in the leachate during the growing
season, and up to 32% was recovered during the cool, high rainfall
months of winter.
Kolenbrander (1969) estimated that 5% of the 180
kg/ha-yr of N applied to the average pasture in the Netherlands was lost
32
by leaching.
Few measurements of mineral N loss from pastures via surface runoff
have been reported.
Moe et al. (1967) detected a 15% loss from a 224
kg/ha N dressing in the runoff from a tall fescue pasture.
In southern
Iowa, Nelson (1973) measured runoff of N from orchardgrass and Kentucky
bluegrass plots on a 5% slope receiving 269 or 538 kg/ha.yr of N.
In a
dry year, losses were negligible, whereas in a wet year, losses of 2.4
and 4.1 kg/ha at the high fertilizer rate were detected from orchardgrass and Kentucky bluegrass, respectively.
Surface losses of mineral
N from grasslands are generally considered to be minor since the dense
foliage and rooting effectively reduce water runoff (Allison, 1955).
Denmead et al. (1974) measured losses of NHg by volatilization
attaining 0.26 kg/ha-day in late summer, but Denmead et al. (1976)
demonstrated the ability of the vegetative canopy to reabsorb some of
that NHg evolved from the soil.
In summary, losses of fertilizer N from grasslands via the indi­
vidual pathways of leaching, denitrification, surface runoff and
volatilization are generally low, but in combination may lead to sig­
nificant quantities.
The degree of loss may be aggravated by high N
fertilizer rates.
Reed canarygrass as a pasture species
Reed canarygrass is indigenous to most temperate regions of the
world, but is most widely cultivated in the Pacific Northwest, the north
central states of the U.S., and southern Canada (Marten and Heath, 1973).
33
Reed canarygrass is a cool-season grass that tolerates flooding as well
as drought.
It also forms a dense sod, is very winter-hardy and has
great persistence.
These characteristics make reed canarygrass an
effective species for productive pasture and hay crops as well as in
controlling soil erosion in grassed waterways and along pond banks
(Marten and Heath, 1973).
Dry matter yields of reed canarygrass are among the highest of the
cool-season grasses in the north-central region.
Schaller et al. (1972)
summarized clipping trials involving four grass species at several Iowa
locations.
Reed canarygrass yielded as high or higher than the other
species in six trials out of seven.
Reed canarygrass yielded more than
tall fescue, orchardgrass and smooth bromegrass in Minnesota (Marten and
Hovin, 1980) and yielded as much as smooth bromegrass and orchardgrass in
Wisconsin (Krueger and Scholl, 1970).
Marten and Donker (1968) reported identical average daily gains of
daily heifers (0.74 kg/day) grazing either reed canarygrass or smooth
bromegrass, but carrying capacity and animal gain per ha were 20%
higher on the former.
In Iowa (Wedin et al., 1970), beef steers gained
faster on reed canarygrass than on tall fescue in spring and summer, but
the reverse was noted during fall grazing.
In vitro dry matter digesti­
bility and DM yield of reed canarygrass decline rapidly in the fall,
which limits its use as a fall-stockpiled forage (Wedin et al., 1966;
Bryan et al., 1970).
Two factors which limit the quality of reed canarygrass in pastures
are (1) the presence of indole alkaloids, which depress weight gains
34
in steers and sheep (Marten et al., 1976), and (2) a high concentration
of cell wall constituents (cellulose + hemicellulose + lignin) relative
to legumes, which limits the rate of forage intake by grazing animals
(Marum et al., 1979).
The identification of genotypes lower in alka­
loids (Marten et al., 1981) and cell wall constituents (Marum et al.,
1979) offer potential for improving the quality and increasing the use
of reed canarygrass as a pasture species.
Animal Use of Pasture N
Animal productivity on pastures
Pastures and rangeland are found in practically all regions of the
world representing the extremes of temperature, moisture and soil type.
Pasture productivity potential is governed largely by temperature and
moisture conditions (Child and Byington, 1981), and soil type (Wedin
and Vetter, 1970) through their effects on plant growth rate, species
adaptability, and length of growing season.
Beef productivity from pastures has been reported as high as 800
kg/ha of liveweight gain in temperate Scotland (Blaxter, 1978), 1200
kg/ha with irrigation in Washington state (Van Keuren and Heinemann,
1958), and 2760 kg/ha in subtropical Australia (Blaxter, 1978).
These
have occurred under conditions of adequate moisture and long growing
seasons.
In the central U.S., some commonly reported liveweight gains per
ha are 506 kg on birdsfoot trefoil-grass in southwestern Wisconsin
(Paulson et al., 1977), 300 kg on Ladino clover-orchardgrass in northern
35
Michigan (Reid et al., 1975), and 561 kg on Ladino clover-orchardgrass
in Tennessee (Fribourg et al., 1979).
Pasture trials in Iowa have yielded liveweight gains per ha of 423
kg on birdsfoot trefoil-Kentucky bluegrass (Wedin et al., 1967), 432,
355
and 336 kg on birdsfoot trefoil-tall fescue, reed canarygrass,
and alfalfa-smooth bromegrass-orchardgrass, respectively (Wedin et al.,
1979).
Wedin and Vetter (1970) determined that liveweight gain from
pastures in Iowa was not always correlated with a soil's suitability to
produce corn.
Rohweder (1963) reported that pasture land was more
frequently found on the low productivity soil types in Iowa.
The rate of beef gain and milk production are usually lower on
pastures compared to those resulting from animals fed high grain rations.
This disparity is one factor limiting the usage of pasture for fattening
cattle and high producing milk cows.
The problem appears to be that
intake of digestible energy is too low to meet the animal needs for an
economic level of production (Conrad et al., 1978).
Utilization of herbage
j_n ruminant digestion
There are numerous reports of animal production rates being less than
expected when animals grazed forages considered high in quality (MacRae
and Ulyatt, 1974; Nicol and McLean, 1970).
Such low performance has
been attributed to inefficient N utilization in the animal (Minson,
1981).
In the rumen, ingested herbage protein is either degraded, result­
ing in NHg production, or passes on undegraded to the gastro-intestinal
36
tract.
Some of the NH^ released in the rumen is available for microbial
protein synthesis, but the rate of this process depends on the avail­
ability of readily fermentable carbohydrate to supply energy and carbon
skeletons (Kay, 1976).
Excess rumen NHg is absorbed into the blood­
stream, converted to urea in the liver, some of which is excreted in the
urine, resulting in poor N utilization by the animal.
Microbial protein
from the rumen and dietary (by-passed) protein that escapes rumen
degradation supply amino acids for digestion and absorption in the small
intestine (Kay, 1976).
High production of meat, milk and wool requires a high flux of
amino acids in the proper proportions to the sites of product protein
synthesis (Chalupa, 1975).
It is recognized that a portion of the
dietary protein must by-pass the rumen to meet the amino acid demand
for maximal production (Burroughs et al., 1975).
Different sources of
dietary N show various efficiencies of utilization by the animal.
These differences have been attributed to the energy content of the
ration, intake level, and the amount of protein by-pass from the rumen
(Beever and Thompson, 1977).
Under intensive management of temperate species, the crude protein
contents of high quality hay and young, succulent herbage range from 18
to 30% of DM, which is in excess of protein requirements of most
ruminants when that herbage is the sole feed (MacRae, 1976).
Sub-
optimal management (lack of N fertilization on grasses) and the use of
tropical grasses and crop residues yield forages with crude protein
levels of 4 to 12% (MacRae, 1976).
Minson (1976) summarized the trends
37
in herbage crude protein concentration as follows:
(1) legumes > grasses,
temperate forages > tropical forages, leaf blade > leaf sheath > stem,
(2) percentage crude protein and leaf to stem ratio decrease with advanc­
ing maturity, (3) crude protein concentration increases with N fertiliza­
tion, and (4) 80 to 90% of crude protein is actual protein.
The amino acid composition of leaf protein is very similar over a
wide range of species (Chibnall et al., 1963) and is not changed by
either the physiological state of the plant or fertilizer treatments
(Wilson and Tilley, 1965).
Although the balance of essential amino
acids in leaf protein is considered highly favorable (Boulter and
Derbyshire, 1978), the quality of ingested protein that is degraded in
the rumen has no effect on protein quality available post-ruminally
(Lewis and Mitchell, 1976).
The inefficient utilization of fresh herbage N appears to be asso­
ciated with a high degree of protein degradability in the rumen (MacRae,
1976) and a short residence time of herbage protein in the rumen (Hogan
and Weston, 1981).
Using perennial ryegrass in sheep feeding trials,
Beever et al. (1974) reported increases in the amounts of amino-N enter­
ing the small intestine and microbial production of 51 and 57%,
respectively, with dried compared to fresh herbage.
MacRae and Ulyatt
(1974) showed with cannulated sheep that about 50% of the herbage
digestible protein was released in the rumen and degraded by rumen
microbes.
Beever and Thompson (1977) listed some forage protein sources
with degrees of degradability of up to 85% in the rumen.
Evidence that
amino acid supply was a factor limiting milk production in cows grazing
38
high protein (28%) perennial ryegrass was demonstrated by Minson (1981)
when daily milk output increased 6% in cows receiving protected casein
versus unprotected casein.
Increasing the availability of fermentable carbohydrate in the
rumen can improve the utilization of N in high NHg-producing feeds by
stimulating microbial protein synthesis (MacRae, 1976).
Moore et al.
(1980) supplemented Coastal bermudagrass with glucose for sheep and
showed an increased ratio of digestible energy to digestible protein,
lower urinary N excretion, and higher N retention.
Eskeland et al.
(1973) increased N retention with intravenously supplied glucose,
acetate, propionate and butyrate.
Grenet and Demarquilly (1978) studied N utilization in sheep fed
alfalfa and four cool-season grass species harvested at various growth
stages and regrowths.
They concluded that high N retention and animal
growth rate were closely associated with high organic matter digesti­
bility, soluble carbohydrate levels and rate of rumen passage of the
forage sources.
Minson (1981) suggested that selecting for higher
soluble carbohydrate levels in plants could improve the capacity of a
forage protein source to supply amino acids to the small intestine.
Despite the reports of inefficient herbage N utilization, forages
have a great capacity to meet the protein requirements of ruminants.
Since forages supply 79% of the feed units consumed by all ruminants in
the U.S. (Wedin et al., 1980), a very large proportion of the protein
consumption must also be met by forages.
High protein alfalfa can meet
the protein requirements of young beef animals and can replace some of
39
the protein supplement from oilseed meals in beef-finishing rations'
(Matsushima, 1972).
The protein requirements for high rate of gain and
choice carcass grade in lambs can be met with alfalfa alone or alfalfaorchardgrass mixtures (Van Keuren and Marten, 1972).
As a protein
supplement for high producing dairy cows, alfalfa can replace some
(DePeters and Kesler, 1980; Marten and Donker, 1974) or all (Marten
and Donker, 1974) of the soybean meal.
Fate of excreted
jjx pastures
Grazing animals constitute the final step in the pastoral process
of converting soil and atmospheric, inorganic N into high quality,
edible protein.
There is, however, a resulting loss of N from the eco­
system, by excreting N in volatile forms and by accumulating N in muscle
or wool growth and milk production, which is removed for marketing
(Whitehead, 1970).
Lactating cows excrete 75% of their N intake, whereas
growing steers excrete 80 to 95%, depending on the N concentration in the
diet.
The more labile fraction of fecal N can be readily converted to
volatile NH^,depending on weather conditions, but most of it is slowly
decomposed and released in the soil by microorganisms and soil fauna
(Hoiter, 1979).
The proportion of excreted N in the urine increases as
the dietary N concentration increases above that required for optimum
growth (Azevedo and Stout, 1974).
Barrow and Lambourne (1962) observed
a range in proportion urinary N from 43 to 80% from sheep grazing herbage
containing from 0.8 to 4.0% N.
About 90% of urinary N occurs as urea
40
and amino-N forms in which deamination on the soil surface readily
occurs.
Low soil cation exchange capacity and hot, dry conditions
promote volatilization of urinary N as NHg.
Doak (1952) reported a 12%
loss of urinary N as NHg within 3 days after excretion.
Ball et al.
(1979) measured a 16% loss of NH^ from urine spots within 11 days after
excretion, as well as significant leaching of NOg".
The grazing and excretion patterns of animals tend to concentrate
the evenly distributed N from herbage into relatively small areas at
levels greatly exceeding the needs of the plants growing in those areas
(especially around water tanks, feeders and shade), leading to excessive
losses and inefficient recycling (Whitehead, 1970).
Ball and Keeney
(1981) proposed that the annual loss by leaching and volatilization of
several hundred kg per ha of N from urine patches in intensively managed
New Zealand pastures could put those systems in a negative N balance.
They did not, however, consider the potential of reabsorption of the
volatilized NHg in the herbage canopy as a check against excessive N
loss (Denmead et al., 1976).
The amount of N removed in animal products from the pasture eco­
system depends on the nature of the product and the productivity of the
pasture (Whitehead, 1970).
Dairy cows producing 6000 kg/ha-yr of milk
at 0.64% N would export 38 kg/ha-yr of N, beef cattle gaining 400
kg/ha-yr of liveweight at 2.4% N would remove 10 kg/ha-yr of N.
The N
export by meat animals is small relative to N input available from N
fertilizer or N fixation (Whitehead, 1970).
replace the export of meat N.
Precipitation N could
41
The effect of N-forage system on forage quality
and
utilization
In an effort to increase animal production per unit land area, N
fertilizer application is an effective tool in increasing DM production
and carrying capacity (Blaser, 1964).
Applying N fertilizer to legume-
grass mixtures is sometimes recommended to make up the N deficit result­
ing from an insufficient legume component (Field and Ball, 1978; Wedin
et al., 1965).
Increasing the N input level in a fertilizer-grass system does not
consistently alter the digestibility of cool-season grasses (Blaser,
1964).
Increasing the N input level in a system based solely on symbi­
otic N fixation results from increasing the legume yield and the N fixa­
tion capacity of the legume component, which in turn are closely related
to the legume proportion in the sward (Hoglund and Brock, 1978; Vallis,
1978).
Since legumes generally exhibit higher digestibilities and in­
take potentials than grasses at similar stages of maturity (Thornton and
Minson, 1973), increasing the legume proportion could favorably alter
the quality of the herbage (Napitupulu and Smith, 1979).
In tissue
analyses of alfalfa-orchardgrass mixtures, Napitupulu and Smith (1979)
found that the percentage cell wall constituents (an indicator of intake
potential) increased from 39 to 59% as the percentage alfalfa decreased
from 100 to 0%.
changed.
In vitro dry matter digestibility, however, was un­
Davis and Greenwood (1972) observed that depressed animal per­
formance on clover-grass pastures in Australia was associated with de­
clines in the proportion of clover.
42
Chalupa et al. (1961) fed reed canarygrass hay that had received
0, 100, or 200 kg/ha of N, or unfertilized alfalfa hay to dairy
heifers.
Nitrogen retention in reed canarygrass-fed animals increased
from +4.4 to +14.8 g/day as the N fertilizer rate increased, and was
+38.0 in alfalfa.
Dry matter and N intake levels were not stated,
but the authors implied that they varied.
Increasing the N fertilizer
rate on orchardgrass from 33.5 to 100 kg/ha increased N intake and
urinary N excretion, but decreased N retention, even though organic
matter digestibility was slightly increased (Grenet and Demarquilly,
1977).
No reports were found in which direct comparisons of N utilization
were made between a legume-grass system at varying legume proportions
and a fertilized-grass system at varying N fertilizer input levels.
In vitro dry matter digestibility of an alfalfa-reed canarygrass mix­
ture was the same as that of reed canarygrass alone in the first cut­
ting, but was higher in the mixture in subsequent cuttings (Krueger
and Scholl, 1970).
Hubbard and Nicholson (1968) detected no differences
in average daily gain in wethers grazing reed canarygrass fertilized
with N at levels of 0 to 300 kg/ha and those grazing a Ladino cloversmooth bromegrass-orchardgrass mixture.
43
PART I.
THE CONTRIBUTION OF N FIXATION TO THE N
ECONOMY OF ALFALFA-GRASS PASTURES
44
INTRODUCTION
Greater understanding of the contribution of symbiotic N fixation
to the N economy of legume-grass mixtures is needed to most effectively
exploit this source of N nutrition in pasture production.
Environmental
effects on N fixation in forage legumes have been widely studied under
controlled conditions, but data on amounts and proportions of N fixed in
established pastures in temperate North America are lacking.
Annual rates
of N fixation in white clover- (Trifolium repens L.) based pastures in
New Zealand were estimated as high as 650 kg/ha (Sears et al., 1965).
Recent estimates of N input from symbiotic fixation in New Zealand aver­
aged 184 kg/ha-yr from nine locations (Hoglund et al., 1979). Phillips
and Bennett (1978) reported an N fixation input of 103 kg/ha-yr in a
50:50 mixture of subterranean clover (Trifolium subterraneum L.) and
Bromus mollis L., and 183 kg/ha-yr with 100% T. .subterraneum L. in a
California winter-range environment.
Seeding-year alfalfa fixed an aver­
age of 148 kg/ha-yr in Minnesota (Heichel et al., 1981).
15
Nitrogen-dilution methods of estimating symbiotic N fixation are
gaining in use in field experiments.
The acetylene reduction method
(Hardy et al., 1968) has been used more extensively because of its low
cost and simple procedure.
Goh et al. (1978) and Phillips and Bennett
(1978) compared the acetylene reduction technique to the use of ^^Ndilution in artificially enriched plots (as outlined by Fried and
Broeshart, 1975; McAuliffe et al., 1958) to estimate N fixation in clovergrass mixtures.
They concluded that the ^^N-dilution measurements have
45
the advantages of integrating N fixation over a range of environmental
fluctuations, plant growth stages and soil conditions.
In addition, the
15
N-dilution method allows the quantifying of soil-derived and atmospherederived fractions of herbage N, which the acetylene reduction method does
not.
It is preferred that a near-isogenic, nonnodulating (or a nodulating, but ineffective) legume be used as the nonfixing reference crop when
the ^^N-dilution technique is employed (Phillips and Bennett, 1978).
Nonfixing isolines are unavailable for most forage legumes.
In estab­
lished pasture mixtures, the accompanying grass provides the only practical
nonfixing control plants.
When using a nonlegume reference plant, two
principal assumptions are that the two species absorb labelled and un­
label led soil N in the same average ratio, and that no transfer of fixed
N from legume to grass occurs during the test period (McAuliffe et al.,
1958).
Heichel et al. (1981) provided evidence to support the first
1C
assumption when comparing two slightly different
N-tracer methods in
alfalfa (Medicago sativa L.) and reed canarygrass (Phalaris arundinacea
L.).
In support of the second assumption, Simpson (1976) did not detect
N release from alfalfa roots until the end of the growing season.
In using ^^N-dilution to assess the partitioning of soil and atmos­
pheric N in tropical legume-grass mixtures. Vallis et al. (1977) observed
that ca. 92% of the legume N was atmosphere-derived, and that the propor­
tion of soil-derived legume N increased curvilinearly with increases in
percentage legume composition.
Edmeades and Goh (1978) found the propor­
tion of N fixed in white clover to decrease from 88 to 82% as the
46
percentage clover increased from 20 to 46%; however, this trend was con­
founded with pasture age.
Percentage N fixed in subterranean clover
declined from ca. 94 to 88% as percentage clover increased from 50% to
100%, at medium and high plant densities (Phillips and Bennett, 1978).
Symbiotic fixation accounted for an average of 43% of the total N uptake
of alfalfa in pure stands (Heichel et al., 1981), but varied widely with
harvest date.
This study was designed (1) to determine the contribution of sym­
biotic N fixation to the N economy of herbage in an alfalfa-orchardgrass
(Dactylis qlomerata L.)-smooth bromegrass (Bromus inermis Leyss.) mix­
ture, and (2) to determine how the level of contribution is affected by
time of season, percentage legume composition and soil factors.
47
MATERIALS AND METHODS
Site Characteristics
This study was carried out on four experimental alfalfa-orchardgrass-smooth bromegrass pastures that were part of a 5-year grazing
trial at the McNay Memorial Research Center in Lucas County of southcentral Iowa.
The soil types were Shelby silty clay loam, Grundy
silty clay loam, and Haig clay loam (Aquic Argiudoll-Typic Argiaquoll).
The pH levels were 6.70 to 6.85 and organic matter content ranged from
3.3 to 3.9% among soil types.
Soil test levels of available P were low,
whereas those of K were high to very high.
The pastures were on 2 to 13%
slopes, of which two pastures faced south and two faced north.
The
pastures were established in the summer of 1974 by seeding 'Pioneer 520'
alfalfa, 'Napier' orchardgrass, and 'Baylor' smooth bromegrass (5.8, 2.7,
and 8.1 kg/ha, respectively) with 'Portal' oats (Avena sativa L.) as a
companion crop.
the mixture.
By 1979, smooth bromegrass comprised less than 5% of
The pastures were 1.32 ha in size and were divided into
three, 0.44-ha paddocks for rotational grazing.
Beef cattle grazing
trials were conducted from early May to late October in each year from
1976 to 1980.
Plot Establishment
In 1979 and 1980, four 1-m^ microplots were installed in paddock 2
of each of the four pastures.
Two microplots were randomly located
along each of the two electric fences bordering paddock 2, with loca­
tions being reassigned in 1980 (Appendix Figure A.l).
Cages were
48
secured over each microplot, around which electric fences were built to
prevent grazing.
On 30 April 1979 and 15 April 1980 a solution of
^^N-enriched ammonium sulfate (77.6 atom-percent in 1979 and 90.2 atompercent in 1980) was introduced into each microplot at the rate of 0.4
g/m of actual
N (a rate of N considered insufficient to inhibit
nodulation or N fixation).
In order to minimize soil disturbance, the
solution was injected with a modified syringe by discharging 5 ml/injec­
tion on a 5 X 10-cm grid through a 10-cm long needle with lateral
holes (see photographs in Figures A.2 to A.4).
Sample Preparation
The microplot herbage was harvested with a hand clipper at a height
of 3 cm on the following dates:
22 May, 4 July, 3 August, 10 September,
and 12 October in 1979, and 22 May, 23 June, 28 July, 13 September and
13 October in 1980.
grazing in paddock 2.
These were the same dates that the cattle began
Samples were frozen at -15°C until it was con­
venient to manually separate them into legume and nonlegume fractions.
Some samples contained small amounts of white clover, red clover
(Trifolium pratense L.). smooth bromegrass, downy bromegrass (Bromus
tectorum L.). Kentucky bluegrass (Poa pratensis L.) and tall fescue
(Festuca arundinacea Schreb.).
The latter two species predominated
in one or two plots each year.
The plant samples were forced-air dried
at 65°C to constant weight and ground to pass a 1-mm screen, with pre­
cautions taken to prevent cross-contamination.
Duplicate samples were
Kjeldahl-digested and ammonium was distilled into 0.05
IN
H2S0^, then
49
back-titrated with 0.05 ^ NaOH to determine total N concentration.
Unlabelled ammonium sulfate was distilled between samples to minimize
cross-contamination by
Each distillate was acidified to a pH of
no less than 2.8, reduced in volume to 3 to 5 ml without boiling,
and stored in a vial in preparation for mass spectrometric analysis
according to Bremner (1965a).
The ^^N-enrichment was analyzed in the
laboratory of Dr. C. M. Cho (Dept. of Soil Science, University of
Manitoba), using a single collector scanning method (coefficient of
variation of 0.6%).
Soil samples to a depth of 15 cm were taken on 7 April 1980 in all
the 1979 and 1980 microplots and analyzed for available P (Bray 1),
exchangeable K, organic matter (ISU Soil Testing Laboratory), total N
(Bremner, 1965b) and anaerobic mineralizable N (Waring and Bremner,
1964).
Additional samplings were made in 1980 on 13 May, 11 June, 17
July, 1 September, and 2 October (the approximate mid-points of each
growth period), and analyzed for NOg'-N and NH^^-N by steam distill­
ing soil extracts (Bremner, 1965c).
Calculations of Plant Measurements
In this paper, herbage refers to legume plus grass.
Percentage
legume composition for each harvest was defined as 100 (legume DM
yield/herbage DM yield).
Nitrogen yield for each plant fraction was
calculated as DM yield x N concentration.
Using the grass portion as
the nonfixing reference crop, the proportion of legume shoot N derived
from fixation was calculated according to the formula:
50
,.
H excess atom-» in legume , p^oponion legume N fixed
N excess atom-% in grass
(Fried and Middleboe, 1977), expressed as a percentage in this paper.
The amount of legume shoot N from fixation in kg/ha-yr was determined
as proportion legume N fixed x legume N yield, expressed as amount N
fixed for brevity.
The percentage of herbage N from fixation equaled
the amount N fixed/herbage N yield.
Soil N recovery in the herbage
(expressed as soil N uptake in kg/ha) was calculated as herbage N yield
minus amount N fixed.
Total annual yields were determined by summing
the harvests and annual percentages of legume composition of the sward,
legume N fixed, and herbage N fixed were calculated from weighted
averages of the five harvests.
Analyses of variance, correlations and regressions were carried
out by the Iowa State University Computation Center using the Statistical
Analysis Systems program.
51
RESULTS
Annual Means
The mean annual yields of DM, percentage legume, and N concentra­
tion are presented in Table I.l.
Herbage and legume DM yields tended
to be higher in 1980, although the wide ranges of values exhibited
among plots prevented significant differences.
Percentage legume
varied greatly among plots within years but the overall means were not
different.
Nitrogen concentration was unchanged between years in the
legume but was slightly higher (P < .10) in 1980 in the grass portion.
Herbage and legume N yields did not differ significantly between
years (Table 1.2).
Percentage legume N fixed varied among plots within
years with values all above 80%.
Since these values were confined to
a relatively narrow range, the amounts of N fixed per ha were related
to legume N yield.
The mean percentage of herbage N derived from fixa­
tion was unchanged between years.
Since new plot locations were
established in 1980, differences between years could have resulted
from a combination of differences in weather, soil and sward composition.
Seasonal Trends
Figures I.l to 1.6 illustrate the effects of year and harvest date
on various aspects of DM yield and N fixation.
In all six cases, there
was a significant (P < .01) harvest date effect within years and a
significant (P < .05) year x harvest date interaction.
A cool, moist
spring and early summer in 1979 stabilized DM yields at relatively high
levels during the first two growth periods (Figure I.l).
Herbage and
Table I.l.
DM yield, percentage legume, and N concentration of herbage by year; means of 16
plots with ranges
1979
C.V.b
Significance®
1980
NS
6668
4090-9980
22
NS
2234
440-4330
53
%
Total yield, kg/ha
6233
4300-8140
Legume yield, kg/ha
2151
490-4400
33
11-55
NS
33
11-56
41
Legume %
3.43
3.09-3.79
NS
3.43
3.06-3.64
5
Grass %
2.42
2.17-2.67
+
2.53
2.17-2.85
7
Percentage legume^
^Comparison between years; NS = nonsignificant, + = P < .10.
'^Coefficient of variation across both years.
^Weighted averages over five harvest dates.
Table 1.2.
N yields and N fixation by year; means of 16 plots with ranges
1979
Significance®
1980
c.v.b
Herbage N yield, kg/ha
174
99-225
NS
191
103-312
29
Legume N yield, kg/ha
75
16-160
NS
81
16-151
55
% Legume N fixed^
91.6
84-98
NS
90.5
80-95
5
N fixed, kg/ha
68
15-136
NS
72
15-122
52
% Herbage N fixed''
36
15-53
NS
36
14-59
35
^Comparison between years; NS = nonsignificant.
'^Coefficient of variation across both years.
''Weighted averages over five harvest dates.
54
legume DM yields in 1980 (Figures I.l and 1.2) showed a maximum at the
first harvest date, but the rate of regrowth in the second period
declined due to dry conditions.
Diminished DM yields in the
final harvest coincided with the onset of dormancy.
Seasonal trends
in herbage (Figure A.5) and legume N yield (not shown) followed the .
trends in DM yields.
(See Table A.l for weather data.)
Percentage legume composition fluctuated throughout the two seasons
(Figure 1.3).
Cool-season grasses are generally more dominant in
legume-grass mixtures in the spring than in subsequent growth periods.
Percentage legume was highest in the last harvest, when DM yields were
the lowest.
The proportion of legume N fixed varied slightly but significantly
(P < .01) among harvest dates (Figure 1.4). The divergent trends during the
final growth period are not explainable by obvious changes in growing
conditions.
Since the amount of legume N fixed per ha is the product of legume
N yield and the proportion N fixed, and since the proportion N fixed
was relatively stable through time, the amounts of N fixed varied in
patterns similar to changes in legume DM yield and legume N yield
(Figure 1.5),
Heichel et al. (1981) observed that the rate of N fixa­
tion is closely linked to growth rate in seeding-year alfalfa.
The proportion of herbage N derived from fixation followed seasonal
trends similar to those of percentage legume (Figure 1.6).
1979
2500
1980
2000
^
Herbage
DM yield 1500
kg/ha
1000
500
0
__i
May
I
I
I
I
1
June
July
Aug.
Sept.
Oct.
1—
Nov.
Harvest date
Figure I.l.
Herbage DM yield by year and harvest date; means of 16 plots ±. S.E.
— - -1979
1000
\
^00
kg/ha
^
\
r
400
\
Legume
DM yield
1
800
1980
\
\
^
^
4—-—
Y.'
200
0
I
May
1
1
June
July
I
Aug.
I
\ N.
'^ \
\ N.
^
\\r
\
1
.
Sept.
1
1
Oct.
Nov.
Harvest date
Figure 1.2. Legume DM yield by year and harvest date; means of 16 plots ±S.E.
1979
%
Legume
50
-
40
-
30
-
1980
y
y
y
-
10
-
0
^
—
-
r
-
•
May
- " X1
i
-r
y
20
J
June
... .1...
•
July
Aug.
Sept.
*
Oct.
'
Nov.
Harvest date
Figure 1.3. Percentage legume by year and harvest date; means of 16 plots iS.E.
- - - -1979
100
1980
95
90
% Legume
N fixed gg
80
75
0
I
May
!
June
!
July
!
Aug.
I
Sept.
I
Oct.
Harvest date
Figure 1.4. Percentage of legume N from fixation; means of 16 plots ±. S.E,
!
Nov.
1979
25
1980
20
N fixed
kg/ha
10
\
5
0
1
•
May
I
I
1
1
1
June
July
Aug.
Sept.
Oct.
1—
Nov.
Harvest date
Figure 1.5. Amount of legume shoot N fixed by year and harvest date; means of 16 plots ±.S.E.
50
40
30
% Herbage
N fixed 20
1979
10
0
1980
_i
May
I
I
1
1
1
June
July
Aug.
Sept.
Oct.
1—
Nov.
Harvest date
Figure 1.6. Percentage of total herbage N derived from fixation; means of 16 plots ±.S.E.
61
Components of DM and N Yield
Schematic representations of how the contributions of the legume
and grass DM components varied with the observed range of total herbage
DM yields are presented in Figures 1.7 and 1.8.
In both years, legume
DM yield increased linearly with herbage DM yield.
Grass DM yield
increased significantly only in 1980, although with lower slope and
correlation coefficient than legume DM yield.
Grass DM yield in 1979
was unchanged (see Table 1.3 for regression equations).
Changes in
legume DM yield accounted for 72 to 80% of the variability in herbage
DM yield.
It is apparent from the patterns in Figures 1.7 and 1.8 that
the percentage legume contribution to herbage DM yield increased as the
legume DM yield increased.
It appears that legume presence was a major
limiting factor to herbage DM yield within the range of yields observed
in these plots.
The pool of herbage N was partitioned into botanical components
and sources of N uptake (Figures 1.9 and I.10).
Again, changes in
legume N yield accounted for most of the change in herbage N yield in
both years.
Similarly, legume N constituted a progressively greater
proportion of herbage N as herbage N yield increased.
In contrast to
the DM yield pattern of 1979, grass N yield increased significantly
with herbage N yield in both years (see Table 1.3 for regression equa­
tions).
Part of the enhanced grass N yield can be explained by the
tendency that the N concentration in the grass was positively corre­
lated with herbage N yield (Table A.2).
As herbage N yield increased, the amounts of N uptake from both
Grass DM
DM yield
g
t/ha
CTï
ro
Legume DM
4 A
2 -
Herbage DM yield
t/ha
Figure 1.7. Components of DM yield in 1979; total of five harvests
10
Grass DM
8
DM yield
g
cr>
t/ha
CO
Legume DM
4
2
0
4
5
6
7
Herbage DM yield
8
9
10
t/ha
Figure 1.8. Components of DM yield in 1980; total of five harvests
r-y/
300
Grass N
250
200
Legume N
N uptake
N from
kg/ha
soil
100
N from
50
fixation
0
100
150
200
Herbage N uptake
250
300
kg/ha
Figure 1.9. Components of N uptake in 1979; total of five harvests
Grass N
Legume N
N from
soil
N from
fixation
Table 1.3.
Regressions of yield components on herbage DM and N yield; n = 16
y variable
Year
Regression
X = herbage DM yield
R
p
t
Legume DM yield
1979
1980
y = -2869 + 0.81x
y = -2253 + 0.69x
.72**
.80**
Grass DM yield
1979
1980
y = 2869 + 0.19x
y = 2253 + 0.31x
.13 NS®
.45**
X = herbage N yield
Legume N uptake
1979
1980
y = -67.5 + 0.82x
y = -50.9 + 0.69X
.87**
.88**
Grass N uptake
1979
1980
y = 67.5 + 0.18x
y = 50.9 + 0.31X
.25*
.59**
N from fixation
1979
1980
y = -53.1 + 0.69x
y = -33.8 + 0.55X
.86**
.79**
N from soil
1979
1980
y = 53.1 + 0.31X
y = 33.8 + 0.45x
.55**
.71**
®NS = nonsignificant.
*,**Significant at .05 and .01 levels, respectively.
67
soil and symbiotic fixation increased in both years, although at a
greater rate by fixation (compare slopes).
Percentage legume was highly
positively correlated with herbage N yield in both years (Table A.2).
Thus, it is notable that as these two variables increased, soil N uptake
was not replaced by fixed N.
Correlations Among Plant Factors
Correlations between various measurements of plant DM yield, N up­
take and N fixation as measured on a yearly basis are summarized in
Table 1.4.
The high magnitude of correlations of legume DM yield and
percentage legume vs. N fixation was due partly to the fact that the var­
iables were calculated with common measurements, and thus were not com­
pletely independent observations.
For example, the same legume DM
yield value was used in calculating percentage legume and amount N
fixed.
Nevertheless, the highly positive correlations among percentage
legume, legume DM yield, legume N yield (Tables 1.4 and A.2), amount
N fixed and percentage herbage N fixed demonstrate the overwhelming
effect that legume presence had in determining the contribution, of
symbiotic fixation to the sward's N econon\y.
This would not neces­
sarily have been the case had the percentage legume N fixed exhibited
a wider range.
The latter measurement did, however, vary negatively
with legume DM yield, amount N fixed, and percentage legume (when
atypical plots were deleted in 1980).
In contrast, Heichel et
al. (1981) noted a positive correlation between amount N fixed and
Table 1.4.
\1979
Correlations (r) between selected plant measurements; n=16 for both years
Legume
DM yield
Grass
DM yield
Grass
% N
Percent
legume
% Legume
N fixed
Amt. N
fixed
Soil N
uptake
.95**
-.73**
.99**
.44+
% Herbage
N fixed
1980^^^
Legume
DM yield
-.18
Grass
DM yield
.27
Grass
% N
.74**
.36
Percentage
legume
.92**
-.08
.81**
.06
-.42
.74**
.62**
-.06
-.17
-.43+
-.81**
.82**
.65**
.66**
-.60*
.95**
.19
.98**
-.70**
-.31
% Legume
N fixed
-.63**
-.79**
-.74**
Amount N
fixed
.99**
.17
.68**
.95**
-.53*
Soil N
uptake
.60*
.89**
.72**
.28
-.96**
.51*
% Herbage
N fixed
.82**
.45+
.97**
-.10
.88**
-.24
.77**
.90**
+,*,**Significant at .10, .05 and .01 levels, respectively.
-.62**
.44+
-.47+
.92**
.12
.08
69
percentage legume N fixed in seeding-year alfalfa grown in pure stands.
Relationships between legume presence and aspects of N fixation are
evaluated in more detail in the forthcoming regression analyses.
Soil N uptake in herbage was positively correlated with legume DM
yield and amount N fixed in both years, but of marginal significance in
1979.
The negative correlation between percentage legume N fixed and
soil N uptake was due to a higher proportion of legume N derived from
soil and a concomitant increase in grass N yield (Table A.2) as per­
centage legume N fixed decreased.
Grass N concentration was highly positively correlated with legume
DM yield, percentage legume and amount N fixed suggesting that the grass
N status was enhanced by the presence of legume.
Since grass DM yield
was unchanged in relation to legume DM yield and percentage legume,
enhanced grass N percentage was not due to the grass N being concen­
trated in less grass DM at the higher legume percentages.
Instead,
grass N yield increased as legume DM yield, percentage legume and amount N
fixed increased.
This commonly observed phenomenon strongly suggests N
transfer from legume to grass, but this experiment was not designed to
measure N flux along this pathway.
Correlations Between Plant and Soil Factors
Tables 1.5 and 1.6 summarize the correlations between various soil
and plant measurements.
testing significant.
In general, the correlations were low with few
In 1979, available P was negatively correlated
with percentage herbage N fixed and percentage legume, for which there
was no logical explanation.
In 1980, available P showed no such
Table 1.5.
Correlation coefficients from comparing soil and plant parameters, 1979; n=16
Available
P
Amt. N fixed
Exchangeable
K
Organic
matter
Total
N
Mineralizable
N
C:N
-.41
-.12
-.11
.06
.07
-.22
% Legume N fixed
.18
-.02
.06
-.01
.14
.09
% Herbage N fixed
-.53*
-.13
-.26
-.14
-.05
-.32
Total yield
-.03
-.02
.06
.28
.02
-.08
Legume yield
-.41
-.13
-.17
.01
.02
-.27
Total N yield
-.15
-.06
.04
.24
.06
-.10
Legume N yield
-.41
-.12
-.13
.05
.05
-.23
% Legume
-.52*
-.16
-.27
-.14
.04
-.36
Grass % N
-.11
-.11
-.09
-.01
.14
-.02
.37
.08
.30
.47+
.03
.16
-.11
-.21
.17
.42
Soil N uptake
Mineralizable N
+,*Significant at .10 and .05 levels, respectively.
-.002
Table 1.6.
Correlation coefficients from comparing soil and plant parameters, 1980; n=16
Organic
matter
Total
N
Mineralizable
N
P
K
Amount N fixed
-.10
-.12
.35
.10
-.24
.29
% Legume N fixed
-.08
-.04
-.40
-.19
-.25
-.29
% Herbage N fixed
-.14
-.15
.14
-.07
-.31
.16
Herbage Dm yield
.05
.01
.56*
.24
-.07
.43+
Legume DM yield
-.09
-.11
.38
.10
-.16
.31
Herbage N yield
.01
-.04
.52*
.18
-.09
.41
Legume N yield
-.10
-.11
.38
.14
-.19
.30
% Legume
-.12
-.16
.21
-.05
-.21
.22
Grass % N
.04
-.10
.35
-.21
.03
.44+
Soil N uptake
.14
.07
.57*
.23
.11
.44+
-.05
-.28
Mineralizable N
-.39
+,*Significant at .10 and .05 levels, respectively.
-.01
C:N
-.39
72
relationship.
Instead, organic matter was significantly related to
herbage DM yield, herbage N yield and soil N uptake, again, with no
obvious explanation.
The carbon (C) to N ratio followed similar trends
since C/N ratio and organic matter were highly correlated with each
other in this study.
Hoglund et al. (1979) noted positive correlations
between C/N ratio and amount N fixed and fixation efficiency (amount N
fixed/t of herbage DM).
Anaerobic N mineralization exhibited no relationship to either
soil N uptake or percentage legume N fixed, which was also observed by
Hoglund et al. (1979).
Field sampling and laboratory error may have
masked any trends, and (or) the technique is not an adequate indicator
of season-long N release and recovery in herbage under the highly vari­
able conditions of these pastures.
Correlations of amount N fixed, legume N yield, legume DM yield and
percentage legume vs. total soil N content were not expected to con­
tribute any information on the capacity of N fixation to increase the
total N level of these soils because the initial soil N levels of these
plots in 1975 are not known, and they are assumed to have been highly
variable.
Percentage Legume and N Fixation
Regression equations were determined to elucidate the relation­
ships between percentage legume and three aspects of N fixation:
(1) the proportion legume N fixed, (2) amount of N fixed, and (3) the
percentage of herbage N derived from fixation.
These regressions.
73
summarized in Table 1.7, were performed without certain data points
which deviated greatly from the observed trends.
The predominant
grass species in those deleted plots were shorter in stature than orchardgrass, which resulted in measurements of percentage legume that were not
in proportion to legume plant frequency.
Three notes of caution should be considered in the interpretation
of these regressions.
First, as mentioned in the correlation analysis,
the dependent and independent variables of some regressions resulted
from calculations based on common plant measurements.
This tends to
exaggerate correlations compared to those determined between completely
independent observations.
Secondly, the independent variable (percentage
legume) was not a treatment level imposed by the experimenter; rather,
it was merely observed.
cause and effect.
Consequently, these regressions do not establish
Thirdly, since percentage legume was not imposed as a
treatment level, there is error associated with the independent variable.
This results in the slopes being slightly underestimated.
Nevertheless,
such regressions describe the "response" patterns of N fixation to a
parameter that is known to influence N fixation in plant communities,
and the regressions are useful in making predictions.
Percentage legume N fixed
Percentage legume N fixed was negatively related to percentage
legume in both years (Table 1.7).
Figure I.11.
The corresponding data are plotted in
In 1980, the correlation was not as strong as in 1979.
The
modest level of correlation indicates that factors other than percentage
legume may also have direct or indirect effects on the proportion of
Table 1.7.
y
Regressions of aspects of N fixation on percentage legume (x variable) in the 2
years separated and combined
variable
% Legume N f i x e d
Amount N f i x e d , kg/ha
% Herbage N fixed
Regression
Year
Combined
y
y
y
1979*
1980b
Combi nedc
y
y
y
1979a
1980°
y
y
y
1979®
1980b
Combined^
—
97.6 - 0.197X
97.6 - 0.251X
=
97.3 - 0.214X
=
=
=
=
=
=
.53**
.43*
.44**
-18.3 + 2.71X
-24.1 + 2.99X
-20.1 + 2.81x
.96**
.94**
.95**
-2.58 + 1.58X - 0.0104x2
6.31 + 0.87X
0.69 + 1.33X - 0.0071x2
.99**
.92**
.96*
= 15, one outlier deleted in 1979 regressions.
^n = 14, two outliers deleted in 1980 regressions.
= 29, three outliers deleted in the combined-years regressions.
*,**Significant at the
R2
.05 and .01 levels, respectively.
- -•--1979
(J1
% Legume
N fixed
0
20
40
60
80
Percentage legume
Figure 1.11. Relationships between percentage legume N fixed and percentage legume
75
legume N derived from fixa* ion.
Total soil N availability throughout the
growing season is probably a main factor directly affecting the propor­
tion legume N fixed due to the propensity of combined N to replace
atmospheric N as a source of N to the legume plant.
As an index of soil
N availability integrated over time, soil N uptake in the herbage was
negatively correlated with percentage legume N fixed in both years (Table
1.4).
Soil N uptake and percentage legume were not correlated indicating
that the two factors acted independently in their effects on the propor­
tion legume N fixed.
As percentage legume increased, there was propor­
tionately more legume plant mass present to compete for available soil
N, resulting in a diminished percentage legume N fixed.
High soil N
availability in a particular location of the pasture, as implied by high
soil N uptake, may have resulted from a history of the high N input from
fixation associated with high percentage legume in that locality.
Soil
and NO^" levels were consistently low and not significantly
related to soil N uptake or proportion legume N fixed.
ranged from 0.2 to 2.0 ppm and
Nitrate levels
from 1.6 to 7.6 ppm.
A factor not measured in this experiment which might have explained
some of the variation in proportion legume N fixed was the pattern of
legume and grass plant frequency in the mixture.
Assuming that the in­
direct effect of percentage legume was mediated through its effect on
interspecific competition for soil N, a nonuniform species arrangement
could have resulted in a degree of competition that differed from that of
a uniform arrangement.
Other microenvironmental variables such as soil temperature and
77
moisture may have affected the percentage legume N fixed through their
effects on zones of rooting and soil N uptake.
Amount
fi xed
Similar response patterns between amount N fixed and percentage
legume were detected for the 2 years (Table 1.7 and Figure 1.12).
The
amount of N fixed per ha increased by 28 kg per 10 percentage-point in­
crease in legume within the range of legume percentages observed.
Below
the observed range, the curve would have necessarily approached zero.
If,
above the range, the proportion legume N fixed continued to decrease, and
if the rate of increase of legume N yield was either stable or lower,
then the curve would be expected to decrease in slope at points above 60%
legume.
Total soil N uptake by the herbage was positively correlated with
amount N fixed (Table 1.4) in both years.
It cannot be determined from
these data whether soil N availability enhanced N fixation per ha.
Amount
of N fixation in the legume shoot, as it was calculated in this experiment,
is directly determined by legume N yield and proportion of legume N from
fixation.
The proportion legume N fixed varied little among plots.
Therefore, legume N yield was the major determinant of the amount of shoot
N from fixation.
While legume N yield increased, percentage legume,
herbage N yield, and grass N yield (which made up the bulk of soil N up­
take) all increased (Table A.2).
Such intercorrelations probably explain
the parallelism between soil N uptake and percentage legume and their
relationship to amount N fixed.
140
120
100
N fixed GO
kg/ha
60
40
20
0
0
20
40
60
Percentage legume
Figure 1.12. Relationships between amount N fixed and perentage legume
80
79
Percentage herbage
fi xed
The proportion of herbage N from fixation showed a significant
quadratic fit in 1979 and a linear pattern in 1980'(Table 1.7, Figure
1.13).
A lack of observations above 43% legume in 1980 may have pre­
vented the expression of a quadratic pattern.
The combined-year
analysis rendered a significant quadratic function suggesting diminish­
ing contributions of symbiotic fixation from succeeding increments of
percentage legume.
This would be expected since at the high legume per­
centages fixed N constituted a progressively smaller proportion of
legume N.
It would have been more informative if legume percentages of 0 to
100% had been represented by these plots.
If so, the effects of high
legume percentages on the contribution of N fixation to the total
herbage N pool could have been ascertained.
In theory, the curves for
proportion legume N fixed and proportion herbage N fixed would converge
toward a common value as percentage legume approaches 100%, since at
that point all the herbage would be composed of legumes.
It is evident
from Figure 1.13 that the proportions of herbage N fixed for the 2 years
would deviate widely from each other when the curves are extrapolated
beyond 60% legume.
However, within the observed range of data, the two
cruves coincide closely.
60
50
40
% Herbage
N fixed
20
10
0
0
20
40
60
80
Percentage legume
Figure 1.13. Relationships between proportion total herbage N fixed and percentage legume
81
DISCUSSION AND CONCLUSIONS
Total yields of DM and N, percentage legume and aspects of N fix­
ation were all similar for the 2 years of this study.
Seasonal trends
within years differed, however.
Variations in total yields of herbage DM and N resulted mainly
from changes in total legume DM and N yields.
It is evident from the
yield component schemes in Figures 1.7 through I.10 and from the posi­
tive correlations of legume DM yield vs. percentage legume and amount of
legume shoot N fixed that legume DM yield in these pastures was the pre­
dominant factor controlling yields of herbage DM and N and the amount
of symbiotically fixed N in the herbage.
The contribution of N fixation
to the total herbage N economy was determined by a combination of per­
centage legume and proportion of legume N fixed, both of which were highly
correlated with legume DM yield.
Within the range of values observed, grass DM was not replaced by
legume DM as percentage legume increased; however, as percentage legume
approaches 100%, the grass presence would necessarily diminish to zero.
It would be desirable to know how the contribution of N fixation to the
sward N econorny would be affected at legume percentages higher than 60%,
because it is at those levels that forage quality and yield are maxi­
mized.
The trends would not necessarily follow from those observed since
the contributions from grass DM and grass N would be replaced by the
legume component as percentage legume approaches 100%.
Different results from those observed would be possible in other
legume and grass species combinations because of root and shoot growth
82
habits that contrast with those of alfalfa and orchardgrass.
Diverse
species combinations could manifest diverse patterns of competition for
soil N, depth of soil N and water uptake, and competition for light by
the shoot.
In the atypical plots, the grass component, compared to
orchardgrass, was short in stature.
Hence, small increases in alfalfa
DM yield resulted in relatively large increases in percentage legume
and percentage herbage N fixed.
This study was conducted during the fifth and sixth years after
pasture establishment.
The observed effects were the net result of many
interacting factors of soil, sward and environment.
Amounts and propor­
tions of N fixed and their relationships with percentage legume may have
been different shortly after pasture establishment.
Later, the cumula­
tive effects of grazing and residue return on soil N availability
would be manifested.
Data are lacking on how the contribution of N
fixation to a sward's N econony changes over the course of several
years.
Heichel et al. (1981) reported values of percentage legume N
fixed in seeding-year alfalfa that were substantially lower than those
observed in this trial.
Soils under long-term, undisturbed grass sod
have been noted to exhibit reduced rates of ammonification and nitrifi­
cation, compared to tilled soils (Soulides and Clark, 1958), which has
implications for soil N availability.
The distinction between soil-derived and atmosphere-derived N in
the legume shoot was made possible by the use of ^^N-tracing, presuming
that the relevant assumptions were justified.
This allowed an evalua­
tion of the contribution of symbiotic N fixation to the N econon\y of a
83
predominantly alfalfa-orchardgrass sward available for grazing.
The
calculated annual amount of N fixed per ha did not include the fixed
N that was partitioned to new root growth.
Yet that may be the por­
tion of symbiotically fixed N that contributes more to stimulating
grass N uptake than fixed N partitioned to the shoot.
It is assumed in
this study, though, that the amount of shoot N from fixation is an in­
dex of the total amount of N fixed. Measurement of the N content and
15
N enrichment of the legume and grass roots in this study in order to
estimate the total amount of N fixed would have been highly conjectural.
Complete recovery and species separation of root mass was impractical
and differences in degree of isotope equilibration with the established
root N pool of the legume vs. the grass were uncertain. In comparing
15
the N-enrichments of seeded vs. transplanted alfalfa, Heichel et al.
(1981) concluded that the assay of shoot N gave a similar estimate of
proportion legume N fixed to that of whole plant assay, but that the
assay of shoot N alone did not account for all the N fixed.
The following conclusions are made from this study:
(1)
The amount of symbiotically fixed N recovered in the legume shoot
varied with harvest date in a manner closely related to legume DM
yield.
(2)
The proportion of legume shoot N derived from fixation varied
slightly with harvest date and averaged ca. 91%.
(3)
The proportion of herbage N derived from fixation varied with
harvest date in a pattern similar to that of percentage legume.
(4)
Legume DM yield and percentage legume were the major factors
84
limiting total herbage DM and N yield and amount N fixed.
There is evidence that the annual proportion of legume N derived
from fixation was negatively affected by legume DM yield and per­
centage legume.
The annual proportion of herbage N derived from fixation is limited
by the percentage legume at levels up to ca. 50% legume, above
which percentage legume N fixed becomes an additional limiting
factor.
In these particular pastures, the amount of symbiotically derived N
in the herbage could be closely estimated by multiplying legume N
yield by 0.90.
Alfalfa-orchardgrass-smooth bromegrass pastures should be managed
for high legume yield and high percentage legume (above 50%) to
maximize the contribution of symbiotic N fixation to the herbage
N economy.
Maintaining at least 10% grass would be desirable to
efficiently utilize available soil N.
85
PART II.
THE NITROGEN USE EFFICIENCY OF ALFALFA-GRASS
AND REED CANARYGRASS PASTURES
86
INTRODUCTION
The productivity of pastures in the subhumid to humid regions of
the world is limited largely by the availability of nitrogen (N).
Applications of fertilizer N can increase the carrying capacity and prof­
itability of a grazing system (Martin and Berry, 1970), but the rising
cost
of N fertilizer has prompted concern about the efficiency of N
use in pastures.
Enhanced protein by-pass to the small intestine and
animal production rate as a result of feeding high-protein herbage
treated with formaldehyde (to inhibit ruminai protein degradation)
(Hemsley et al., 1970) has also stimulated interest in the efficiency
of N use by grazing animals.
Numerous trials have demonstrated enhanced forage quality of grass
pastures by including a legume in the sward.
Mixtures of tall fescue
(Festuca arundinacea Schreb.) or orchardgrass (Dactylis glomerata L.)
with Ladino clover (Trifolium repens L.) or alfalfa (Medicago sativa L.)
gave higher average daily gains (ADG), steer liveweight gains (LWG) per
ha and a higher degree of carcass finish than those grasses grazed in
pure stands and receiving 200 kg/ha-yr of N (Van Keuren and Heinemann,
1958).
Blaser et al. (1956) also reported enhanced daily weight gains
of steers grazing tall fescue or orchardgrass with Ladino clover vs. the
pure grass stands receiving 240 kg/ha-yr of N.
A mixture of orchard-
grass-Ladino clover exhibited greater ADG, feed conversion efficiency
(kg DM intake per kg of LWG) and daily intake per animal than Midland
bermudagrass (Cynodon dactyl on (L.) Pers.) receiving 448 kg/ha of N
(Fribourg et al., 1979).
Beef cow and calf daily gains were increased
87
by including Ladino clover with tall fescue compared to fescue alone
receiving N (Burns et al., 1973; Petritz et al., 1980),
Petritz et al.
(1980) also reported enhanced cow conception rates and economic return
in the legume-based system.
Reed canarygrass received 0, 150 or 300 kg/ha or was grown with
Ladino clover without N fertilizer and was grazed by sheep (Hubbard and
Nicholson, 1968).
Increasing the N fertilizer rate resulted in a higher
carrying capacity and LWG per ha, but no change in ADG or feed conver­
sion efficiency.
The LWG per ha and ADG from the legume-grass sytem
were similar to those of grass receiving 150 kg/ha of N, but the cost
was a, one-time $.50/ha for legume seed at establishment vs. an annual
cost of $42.00/ha for the N fertilizer.
In his review of the effects of N fertilizer on pasture productivity
and quality. Blaser (1964) concluded that augmenting the N fertilizer
level increased the carrying capacity and animal productivity per ha, but
usually did not affect the performance per animal.
Niehaus (1971) de­
tected no effect of N fertilizer at rates up to 600 kg/ha on in vitro
digestibility of reed canarygrass.
In vitro digestibility was slightly
depressed by N fertilizer in the first cutting of reed canarygrass, but
unaffected in subsequent cuttings (Krueger and Scholl, 1970).
In con­
trast, Chalupa et al. (1961) reported that the digestible energy content
of reed canarygrass fertilized with 200 kg/ha of N increased from 2,478
to 2,694 cal/g, while that of pure alfalfa was 2,346 cal/g.
The problems in comparing the N economies of legume-grass and Nfertilized grass pastures were reviewed by Henzell (1970).
He stated
88
that the evaluation of the efficiency of N use for vegetative and animal
production is complicated by the lack of accurate techniques for measur­
ing the flow rates of N along many of the pathways of the N cycle and
the fact that grazing animals harvest their feed and excrete their
wastes over an extended period and area.
Estimates of energy avail­
ability and intake by animals when calculated in reverse from observed
rates of gain using feeding standards were used by Hubbard and Nicholson
(1968) and Mott et al. (1970) to compare pasture treatments.
That method,
based on the net energy system (National Research Council, 1976), was
used in this experiment to estimate DM and N intake by steers to com­
pare relative differences between treatments.
The objectives of this experiment were (1) to compare, at two graz­
ing pressures, the plant and animal use efficiency of pasture N of two
N-forage systems:
a legume-based system consisting of alfalfa-orchard-
grass-smooth bromegrass (Bromus inermis Leyss.) receiving no N fertil­
izer vs. reed canarygrass receiving 180 kg/ha-yr of N; and (2) to evalu­
ate the effects of N-forage system and grazing pressure on total soil N
accretion in the surface 7.5 cm.
These particular pasture systems were
compared because they were established, had a known grazing history,
and are both recommended as high quality, productive and persistent
systems for the north-central region of the U.S.
89
MATERIALS AND METHODS
This study was carried out on the same experimental pastures as
those described in Part I.
The experimental design was a 2 x 2 random­
ized complete block factorial with two replications.
of two N-forage systems:
One factor consisted
a legume-based system with alfalfa-orchardgrass-
smooth bromegrass receiving no N fertilizer, and reed canarygrass receiv­
ing 180 kg/ha of N annually.
The N dressings were split equally between
mid-April and late-July applications.
Each system was grazed at two
grazing pressures for which the intended herbage allowances were 11 and
8 kg of forage dry matter (DM) per animal-day at the beginning of each
grazing period, for medium and heavy grazing pressures, respectively.
Soil Analysis
Soil samples were taken in the north and south halves of each of
the eight pastures in 1975 at three depths:
and 15 to 30 cm.
On 24 April 1979, the entire area of each pasture was
sampled at three depths:
cores/sample).
0 to 7.5 cm, 7.5 to 15 cm,
0 to 3 cm, 3 to 15 cm, and 15 to 30 cm (30
On 1 November
1979, the middle paddock of each pasture
was divided into three 0.15-ha sampling areas and sampled only at 0 to 3
cm, again 30 cores/sample.
On 21 July
1981, the pastures were resampled
as in 1975, but only at 0 to 7.5 cm (35 cores/sample), in order to compare
the total N levels before and after 5 years of grazing.
All soil samples
were analyzed for total Kjeldahl N (Bremner, 1965b) without pretreatment
to include NO^N, since NOg" levels were consistently negligible.
Analyses
90
of organic carbon (C) (assuming negligible free carbonate-C in these
soils) were made on the 1979 samples using dry combustion in a LECO
carbon analyzer (Tabatabai and Bremner, 1970).
The 1975 and 1981
samples were air-dried and ground to pass a 60-mesh screen with vege­
tative debris and dung sifted out.
The 1979 samples, including vege­
tative debris and dung, were air-dried and ground to pass an 80-mesh
screen.
Vegetative Analysis
Forage availability and DM yields were estimated by mowing three
2
4-m random strips in the middle paddock of each pasture as the cattle
were moved to that paddock during each rotation cycle.
Subsamples
were taken for DM determination and were ground through a l-rrân
screen and analyzed for total Kjeldahl N (Bremner, 1965b).
Sub-
samples were also taken and preserved by freezing until it was conveni­
ent to manually separate them into legume, grass, and dead residue
fractions.
These samples were analyzed for total Kjeldahl N and in
vitro dry matter digestibility (IVDMD) (Marten and Barnes, 1979).
Grass
fractions from only the reed canarygrass pastures were treated with
salicylic acid prior to Kjeldahl digestion to reduce NG^" to
(Bremner, 1965b).
Other fractions had negligible NOg" levels.
At each
harvest, visual estimates of percentage legume and grass composition in
the alfalfa-grass pastures were made from quadrat readings in 10 random
locations in each of three subdivisions of the middle paddock.
The
amount of herbage N derived from symbiotic fixation was estimated by
91
multiplying the legume N yield by the proportion of legume N derived
from fixation.
The latter value was determined for each pasture as the
mean percentage N fixed observed in caged microplots using the ^^Ndilution method (see Materials and Methods, Part I).
The DM yield and
N yield responses to levels of N fixation (using the amount of shoot N
fixed as an index of total amount N fixed) were determined by regres­
sion analysis of the data from the randomly harvested strips.
Since the
reed canarygrass pastures received a uniform input of N at 180 kg/ha,
DM yield and N yield responses to fertilizer N were evaluated by setting
2
out four 4-m plots in each of the four pastures in 1980. These plots
received 0, 60, 120, and 180 kg N/ha as ammonium nitrate.
The IVDMDs
were determined for the first two harvests of 1980 only.
Animal Analysis
The pastures were grazed by beef steers of various breeds in 170 to
180-day grazing seasons, being weighed every 28 days, and spending 10
to 14 days in each paddock.
Stocking rates were adjusted upon entering
each new paddock by using "put-and-take" grazers to control grazing
pressure.
Average daily gains (ADG) in kg/animal-day were calculated
from the gains of the tester animals, of which six were on the pastures
until the end of July, and three remained for the rest of the season.
Total liveweight gains on an area basis (LWG in kg/ha) were calculated
as the product of the ADG of the testers and the number of animals car­
ried per ha, including testers and grazers.
The amount of DM intake
theoretically necessary to attain the observed ADGs was calculated on
92
an animal-day basis and on an area basis using tables based on the net
energy system (National Research Council, 1976).
Nitrogen intake was
determined as DM intake x percentage N of herbage corrected for dead
plant residue.
Nitrogen gain in animal tissue was estimated from pub­
lished body-composition tables (Fox et al., 1977).
Nitrogen excretion
was calculated as the difference between N intake and N gain.
Differ­
ences between N-forage systems and grazing pressures were tested by
single-degree-of-freedom comparisons in analysis of variance.
93
RESULTS
Total Availability of Forage DM and N
The total amounts of herbage DM and N available for grazing through­
out the grazing season and herbage N concentration are presented in Table
II.1,
Only the values corrected for dead plant residue are used in the
Table II.1.
Means and ranges of DM and N yields in kg/ha, and N con­
centration of herbage in alfalfa-grass and reed canarygrass pastures; means are of two replications and two
grazing pressures
Alfalfa-•grass
Reed canarvqrass
1979
1980
1979
1980
Harvested DM®
6228
4569-7687
5480
3656-7259
8186
7990-8639
7753
7424-7815
Corrected DM®
5528
3819-6987
4839
3268-6301
7219
6981-7586
6638
6143-6816
Harvested N®
169
106-231
150
86-211
249
222-268
241
228-248
Corrected N®
162
100-221
140
82-196
233
207-248
221
212-232
2.93
2.62-3.16
2.89
2.51-3.11
3.23
2.97-3.46
3.33
3.11-3.52
Corrected
^Within-year differences between N-forage system are significant
(P<.05).
subsequent plant and animal response analyses.
Dry matter and N yields
in reed canarygrass were consistently higher and less variable than in
alfalfa-grass.
Two reasons for these trends were:
(1) reed
94
canarygrass received a higher level of N input (180 kg/ha-yr) compared
to alfalfa-grass (aver. 70 kg/ha-yr from fixation);
and
(2) alfalfa-grass was a complex mixture exhibiting highly variable
frequencies and spatial distributions of legume plants among the four
pastures.
It was demonstrated in Part I that total herbage yield was
closely related to legume yield.
Vegetative Responses to N Input
Regression equations describing the responses in DM and N yield to
the different forms of N input (N fixation vs. N fertilizer) are presented
in Table II.2.
Table II.2.
Unfortunately, data are unavailable for the reed
Regressions of DM and N yield on N input level (independ­
ent variable)
N source
Year
Regression
^y«x
-y variable = DM yieldN fixation^
1979
1980
y = 3524 + 36.9x
y = 2793 + 43.Ix
.76**
,78**
694
633
N fertilizer
1980
y = 3381 + 13.6x
.71**
617
—y variable = N yield
N fixation'
N fertilizer
1979
1980
y = 77.9 + 1.54x
y = 64.4 + 1.60x
.86**
.82**
20.5
20.9
1980
y = 94.2 + 0.67x
.85**
20.0
®n = 12.
^n = 16.
**Significant regression at P<.01.
95
canarygrass yield responses for 1979.
In comparing results of 1980 only,
the slope of the DM yield response in alfalfa-grass was significantly
higher (P<.01) than that of reed canarygrass.
A 1 kg/ha increase in N
input resulted in a DM yield increase that was three-fold higher in the
alfalfa-grass mixture than in reed canarygrass.
A similar pattern existed in the N yield response, in that the slope
was significantly greater (P<.01) in alfalfa-grass.
It is noteworthy that
the slope was greater than one for alfalfa-grass and less than one for
reed canarygrass.
From the ^^N-microplot data on the composition of N
yield (Figures 1.9 and I.10), it was shown that as the amount of harvested
N from fixation increased, the amount of soil N uptake in the herbage also
increased.
The slope of 0.67 in the reed canarygrass N yield response im­
plies an apparent N recovery of 67%.
In 1980, two microplots in the reed
canarygrass were delineated and received 180 kg/ha of N as aimonium sul­
fate enriched with
fertilizer N recovery.
at 2 atom-%.
Herbage
enrichment indicated 54%
Immobilization of fertilizer N in organic matter,
leaching and gaseous losses, and partitioning of N to the roots explain
the failure to recover all the N applied.
By contrast, about 75% or more
of the N fixed in established alfalfa is translocated to the shoot (G. H.
Heichel, University of Minnesota, personal communication, 1981).
In
addition, enhanced levels of soil N uptake were associated with increased
amounts of N fixation (see Figures 1.9 and I.10 for data on N yield
composition in the
microplots).
The IVDMD and N percentages of the alfalfa fraction were consis­
tently greater than those of the orchardgrass-bromegrass fraction in
96
the legume-based system (Table II.3).
The IVDMD of the alfalfa-grass
mixture would be expected to increase as the percentage legume increased.
The IVDMD of reed canarygrass was modestly enhanced by N fertilization
in the first harvest but unchanged in the second harvest.
Nitrogen fer­
tilization enhanced the herbage N concentration in the first, third
and fourth harvests of reed canarygrass (Table A.3 ).
Table II.3.
The IVDMD and N percentage
of alfalfa and orchardgrassbromegrass fractions and reed canarygrass at four levels
of N fertilizer, in first two harvests of 1980
% IVDMD
Harvest 1
Alfalfa
Orchardgrass
Bromegrass
Reed canarygrass
0
60
120
180
74.0
67.3
*
54.0
57.0
56.9
58.4
*
% N
Harvest 2
67.4
57.4
*
57.5
56.8
58.2
58.7
NS
Harvest 1
Harvest 2
3.18
2.10
3.02
2.22
**
**
2.41
2.96
3.32
3.69
3.00
3.19
3.13
3.38
NS
**
*,**Significantly different at .05 and .01, respectively.
Animal Responses
Only the main effects of N-forage system and grazing pressure are
reported here because no interactions were detected.
The actual grazing
pressures that were calculated ex post facto are listed in Table II.4.
Herbage allowance was difficult to control Tn orië'alfalfa-grass pastûrë'bf
medium grazing pressure.
Herbage DM production was the lowest in tha.t
Herbage allowance, weight gain and number of animals carried by N-forage system and
grazing pressure treatments; n=4
Herbage
allowance
1979
N-forage system
Alfalfa-grass
Reed canarygrass
Grazing pressure
Medi um
Heavy
1980
Animals
carried
ADG
X
1979
1980
LWG
X
1979 1980
1
1
Treatment
-animal-days/ha-
CL
Table II.4.
X
-kg/animal-day—
—kg/animal-
9.3
10.4
NS*
7.9
8.8
NS
8.6
9.6
NS
.524
.551
NS
.477
.442
NS
.500
.496
NS
712
838
691
891
702
863
**
**
**
10.8
8.9
NS
9.5
7.2
NS
10.1
8.0
NS
.576
.499
NS
.517
.402
.546
.451
714
834
700
882
707
858
**
**
**
*
*
®NS = nonsignificant.
*,**Significant at .05 and .01 levels, respectively.
1979
1980
kg/ha-
X
——
371
458
NS
322
392
345
425
*
*
412
417
NS
358
357
NS
385
387
NS
98
pasture and the number of tester animals constituted the lower limit
of stocking rate.
The result was a herbage allowance that was more
similar to that of the pastures under heavy grazing pressure.
When
omitting that pasture from the actual grazing pressure mean, the mean
herbage allowance of the two N-forage systems would be similar and
the herbage allowance of the medium treatment would be 11.7 and 10.4
kg/animal-day in 1979 and 1980, respectively.
That pasture was re­
tained as a medium grazing pressure treatment in the subsequent mean
comparisons because replication was too low to sacrifice degrees of
freedom.
The ADGs, numbers of animals carried, and LWGs per ha on these
pastures are presented in Table II.4.
Average daily gain was not
significantly different between the N-forage systems, but was higher
(although barely nonsignificant in 1979) under medium grazing pressure,
since forage availability per animal-day was higher under medium pres­
sure.
The number of animals carried throughout the grazing season
was greater in the reed canarygrass sytem because of its higher forage
yield, and was greater in the heavy grazing pressure.
Liveweight gain
per ha was also higher in reed canarygrass, since it had a higher carry
ing capacity than alfalfa-grass, but was unaffected by grazing pressure
treatment.
Estimated DM intake was calculated on area and animal-day bases
(Table II.5).
Intake per ha was higher in reed canarygrass because
of its greater herbage Dm production and higher average stocking rate.
Intake per ha tended to be greater in the heavy grazing pressure.
99
Table II.5.
Estimated DM intake and efficiency of conversion to liveweight gain for N-forage system and grazing pressure
treatments; n=4
ZZ:
N-forage ' system
Alfalfa-grass
Reed canarygrass
1979
1980
4968
6067
kg/ha
4755 4862
7262 6665
*
Grazing pressure
Medium
Heavy
5295
5740
NS
X
**
**
5665
6354
+
5480
6047
*
1979
1980
_
X
--kg/animal-day—
6.98 6.96
7.28 8.16
NS^
*
-.sasT
1979 1980
X
—---kg/ikg
--
6.97 13.5 14.9 14.2
7.72 13.3 18.7 16.0
*
NS
**
*
7.39 8.03 7,71 12.9 15,9 14.4
6.87 7.10 6.98 13.9 17.7 15.8
NS
*
*
NS
+
*
^NS = nonsignificant.
+,*,**Significant at .10, .05 and .01 levels, respectively.
presumably because of more thorough utilization of available forage.
Intake per animal-day was slightly greater in reed canarygrass but tor no
obvious reason.
A heavy grazing pressure resulted in a slightly lower
DM intake per animal-day which would normally result from a diminished
herbage allowance.
The conversion efficiency of herbage DM into animal
gain was unaffected by N-forage system in 1979 but was greater (i.e. lower
value) in alfalfa-grass.
As a manifestation of enhanced forage quality,
greater DM conversion efficiency would be expected to increase with
increases in IVDMD.
The IVDMD values (Table II.3) were, on the average,
higher in alfalfa-grass than reed canarygrass.
Estimates of N gain and N intake are presented in Table II.6.
Since
these values were calculated from LWG and DM intake, N gain and N intake
follow trends practically identical to those of LWG and DM intake
Table II.6.
Nitrogen gain in animal tissue and N intake by N-forage system and grazing pressure
treatments; n=4
Treatment
^
1979
N-forage system
Alfalfa-grass
Reed canarygrass
Grazing pressure
Medium
Heavy
1980
X
1979
1980
X
~kg/animal- day—
1979
9.7 10.4
11.2
13.9 11.8 12.8
*
NS* *
.016
.017
NS
.014
.013
NS
.015
.015
NS
144
195
12.4 10.7 11.6
12.6 10.8 11.7
NS
NS
NS
.017
.015
NS
.015
.012
.016
.014
168
172
NS
*
*
———— —
*
^NS = nonsignificant.
+,*,**Significant at .10, .05 and .01 levels, respectively.
1980
X
-kg/ha-
-— —
—
136
242
140
219
**
**
180
199
NS
174
185
+
1979
1980
.202
.235
NS
.198
.273
.200
.254
**
**
.232
.205
NS
.250
.220
+
.241
.213
X
—kg/animal -day-
*
101
(Tables II.4 and II.5).
There was a proportionately greater increase
in N gain and N intake in reed canarygrass over alfalfa-grass than in
LW6 and DM intake because of the higher N concentration in the reed
canarygrass herbage.
Two indices of the animal use efficiency of herbage N are presented
in Table II.7.
Table II.7.
Since the two N-forage systems were at different levels
The animal conversion efficiency of N into LWG and tissue
N gain by N-forage system and grazing pressure treatments;
n=4
Treatment
^ intake/LWG
1979
N-forage system
Alfalfa-grass
Reed canarygrass
Grazing pressure
Medium
Heavy
1980
N intake/N gain
X
kg/kg -
1979
1980
I
— - - kg/kg—
.430
.430
NS^
.425
.622
**
.405
.526
*
12.8
14.2
NS
14.1
20.7
**
13.4
17.4
*
.404
.411
NS
.499
.547
NS
.452
.479
NS
13.4
13.6
NS
16.6
18.1
NS
15.0
15.9
NS
®NS = nonsignificant.
*,**Significant at .05 and .01 levels, respectively.
of N input, these indices relate animal response to N intake, rather
than to N input level, as was done in the comparison of plant response.
Comparing the amounts of N intake associated with 1 kg of LWG,
it appears that alfalfa-grass was more efficient (lower value).
Likewise, the N intake resulting in 1 kg of tissue N gain was lower
in alfalfa-grass and, again, more efficient than in reed canarygrass.
102
The proportion of consumed N retained in the animals (the reciprocal of
the N intake per N gain ratio) was higher in alfalfa-grass.
Grazing
pressure had no clear, consistent effect on the N use efficiency.
It is likely that the failure of the N-forage system means to test
significantly different in 1979 when the means were significantly dif­
ferent in 1980 and in the combined-year analysis was due to one anomalous
reed canarygrass pasture that exhibited a relatively low ADG even though
herbage allowance was medium.
With so few degrees of freedom for error
(df=3) in the analysis of variance F-tests, one anomalous observation
had a large effect on the variance.
Return of N to the Soil
The amount of N excreted per animal-day was greater in reed canarygrass than in alfalfa-grass (Table II.8) because the former system
exhibited enhanced N intake per animal-day and the amounts of N gain per
animal-day were equal.
Nitrogen excreted per animal-day was greater in
the medium grazing pressure than in the heavy, again, because N intake
per animal-day was greater in the former.
Nitrogen excreted per ha was
also greater in reed canarygrass because N intake per ha was greater and
differences between N gain per ha were small relative to the N flux
through the animals.
The heavy grazing pressure tended to exhibit a
modest increase in N excretion per ha over the medium pressure.
There was a greater annual rate of N return to the soil surface per
unit area as vegetative residue in reed canarygrass than in alfalfa-grass
(Table II.8).
This resulted from the higher DM productivity and N
Table II.8.
Amounts of N returned to the soil as vegetative residue and animal excretion; n=4
Animal N excretion
1979
N-forage system
Alfalfa-grass
Reed canarygrass
Grazing pressure
Mediurn
Heavy
1980
X
1979
--kg/animal -day—
.184
.260
.185
.239
133
181
**
**
*
.215
.190
NS
.235
.208
.225
.199
155
159
NS
*
1980
X
1979
—kg/ha-
.186
.219
NSD
+
residual^N
1980
Total N return®
X
1979 1980
-kg/ha-
126
230
130
206
**
**
169
188
NS
162
174
+
37.4
62.9
*
31.2
52.1
*
—kg/ha34.3
57.5
*
52.0 46.0 49.0
48.3 37.3 42.8
NS
NS
NS
®Sum of excreted N and residual N.
^NS = nonsignificant.
+,*,**Significant at .10, .05 and .01 levels, respectively.
X
170
244
157
282
164
263
**
**
**
207
207
NS
215
225
NS
211
216
NS
104
concentration of reed canarygrass.
Other factors contributing to the
greater residue return in reed canarygrass were the characteristics of
this species (1) to produce an abundance of herbage in May and June which
is inefficiently used by the animals because of trampling, and (2) to form
a coarse, dense stubble at ca. 15 cm, below which the cattle seldom
graze when in a dense sward.
Vegetative residue return was less under heavy than under medium
grazing pressure, which is what one would expect, but statistical pre­
cision was inadequate to render a significant test.
The total annual
amount of N return (residue N + excreted N) was significantly higher in
reed canarygrass but was unaffected by grazing pressure.
Nitrogen intake
per ha was slightly higher in the heavy pressure, but animal N gain
(i.e. N removal as animal tissue) per ha was unchanged by grazing pres­
sure (Table II.6), resulting in greater N excretion per ha in the heavy
pressure.
It is evident that approximately the same amount of herbage N
was returned to the soil regardless of the proportion of herbage N
diverted through the grazing animal.
Comparison of total N levels in the soil before and after 5 years
of grazing (Table II.9) indicates that N accumulated in the surface 7.5
cm of the reed canarygrass pasture at a greater annual rate than in the
alfalfa-grass pastures.
This was not unexpected since reed canarygrass
was at a higher annual N input level than alfalfa-grass (180 vs. ca. 70
kg/ha-yr).
The N input level of the two systems appeared to exert its
effect on soil N accumulation through its influence on total annual N re­
turn, since the ratio of alfalfa-grass to reed canarygrass N return
105
(164/263 = 0.62) and the'alfalfa-grass/reed canarygrass ratio involving
soil N accumulation (.045/.074 = 0.61) were essentially the same.
Table II.9.
Accretion in total soil N from 1975 to 1981 at 0 to 7.5 cm
Treatment
Total N
in 1981
Total N
in 1975
c
Amount of
increase
%rage points
I
Alfalfa-grass
Medium^
Heavyb
X
.206
.218
.252
.262
.046
.044
.045
Reed canarygrass
Medi umb
Heavya
X
.183
.202
.258
.276
.075
.074
.074
^n=4.
^^=2; soil samples from one alfalfa-grass pasture and one reed
canarygrass pasture lacking from 1975.
106
DISCUSSION AND CONCLUSIONS
It was demonstrated that the alfalfa-grass responses to N input from
fixation in terms of DM and N yield were greater than those of reed
canarygrass.
It should be noted that the N fertilizer plots in reed
canarygrass were on areas that had received 180 kg/ha of N during each
of the previous 5 years.
The build-up of soil N over that time may have
resulted in diminished plant growth responses to N fertilizer applica­
tions applied in later years.
Most published regression slopes for DM
yield response in cool-season grasses are in the range of 15 to 35 kg
DM per kg of N (Reed, 1981).
The more efficient use of herbage N by the animals in alfalfa-grass
was manifested in the observation that a higher proportion of alfalfagrass N intake was retained for liveweight gain.
Possible explanations
for this difference are that (1) the higher digestibility of alfalfagrass (as estimated by IVDMD) allowed a better energy-protein balance in
the course of rumen digestion of the forage, which may have prevented
excessive loss of ammonia from the rumen to the urine; (2) the N concen­
tration of reed canarygrass was higher than in alfalfa-grass, and since
protein was apparently less limiting than energy, there was excessive N
in reed canarygrass (NOg levels were negligible in reed canarygrass); and
(3) indole alkaloids in reed canarygrass (not tested in this study) may
have been above critical levels for causing metabolic disorders and re­
duced feed conversion efficiency.
At a mean daily rate of gain of 0.50 kg and at a mean body weight
of 287 kg in the 2 years of this trial, the steers required a daily
107
energy intake of 5.37 and 1.95 Meal for maintenance and gain, respective­
ly, for a total of 7.32 Meal.
The estimated DM intake was lower in
alfalfa-grass (Table II.5) even though ADG of the two pasture systems
were the same (Table II.4).
This resulted from higher concentrations
of net energy in alfalfa-grass for both maintenance and gain, hence
the higher feed conversion efficiency of alfalfa-qrass as compareîT to"
reed canarygrass.
the enhanced IVDMD values observed in alfalfa-grass
support the thesis that energy availability was greater in that system
than in reed canarygrass.
The mean daily requirement of total (crude) protein for the steers
was 0.61 kg/animal (National Research Council, 1976) or 0.10 kg of N.
Estimated N intake per animal-day (Table II.6) exceeded the requirement
in both N-forage systems, but exhibited a greater excess in reed canary­
grass than in alfalfa-grass.
factors:
This resulted from a combination of two
(1) the greater amount of DM intake required for 1 kg of LWG
in reed canarygrass necessarily caused more N intake; and (2) reed canary­
grass herbage contained a higher concentration of N.
The greater excre­
tion of N per animal-day in reed canarygrass suggests a higher degree of
N wastage than in alfalfa-grass.
The variety of reed canarygrass grown in these pastures, 'Rise',
often exhibits concentrations of tryptomine-carboline alkaloids that
have been shown to inhibit DM intake and ADG, and are associated with
higher incidence of diarrhea in sheep and cattle (Marten, 1973).
Environ­
mental factors which promote higher concentration of these alkaloids are
high N fertilizer input, high temperature and moisture stress (Marten,
108
1973), conditions which existed at times during the trial.
A low-alkaloid
reed canarygrass variety may have rendered different results.
If the animal response to N input level rather than to N intake was
used as the basis for comparing the two pasture systems (e.g. kg LWG
per kg N input either by fertilizer or fixation), the result would show
a large bias in favor of alfalfa-grass because its N input was so much
lower.
Simple ratios of productivity per unit of resource input normally
decline as the input level increases because that resource becomes less
limiting at higher levels.
Consequently, the question remains:
What'
would happen to the animal use efficiency of herbage N if the two systems
were at equal N input levels?
Increasing the N input level in alfalfa-
grass would necessitate a higher percentage legume composition.
Using
the combined-year equation relating percentage legume and amount N fixed
from Part I (Table 1.7; y = -20.1 + 2.81 x) and assuming that 90% of the
actual amount of N fixed is translocated to the shoot, one can estimate
that an alfalfa-grass mixture of 65% legume would be required to"
fix 180 kg/ha.
This would also result in a higher digestibili"ty ^
of the forage since the IVDMD of alfalfa exceeded that of the grass frac­
tion (Table II.3).
Napitupulu and Smith (1979) demonstrated that increas­
ing the percentage alfalfa in an alfalfa-orchardgrass mixture decreased
the concentration of cell wall constituents, suggesting a higher intake
potential.
The effects of increased legume percentage would improve
forage quality and rate of animal gain, which could lead to enhanced N
use efficiency.
However, a greater proportion of the diet would be com­
posed of alfalfa leaf protein, which is rapidly released and degraded to
109
NHg in the rumen and therefore subject to loss.
If the N input level of alfalfa-grass were held constant at 70 to
80 kg/ha and reed canarygrass received that much fertilizer N, the
digestibility of reed canarygrass probably would not be significantly
affected, but the N concentration would be lower.
The effect would be
less N intake per animal-day with the same energy availability (given
no change in herbage allowance), possibly resulting in a more efficient
conversion of herbage N to LWG and tissue N gain, as long as the
metabolizable protein needs of the animal were met.
a lower cost per unit of gain.
The result would be
The economic return per ha would also
be lower because of a lower carrying capacity.
Consequently, the
advisability of reducing the N rate to 80 kg/ha-yr would depend on the
relative costs of N and land.
Estimates of DM and N intake were based on the net energy feeding
standards of the National Research Council (1976).
The net energy values
were derived from feeding trials of confined animals, which probably re­
sulted in energy requirements for maintenance that are lower than those
for grazing animals.
Consequently, the absolute amounts estimated may
not be accurate, but large treatment differences should have been
detectable.
The higher efficiency of N use in plant growth and animal LWG in
alfalfa-grass cannot be attributed solely to the presence of the legume
since the grass species were not the same in the two systems.
In conclusion:
(1)
The alfalfa-grass system supported a higher level of DM and N yield
110
per unit of N input than the N-fertilized reed canarygrass system.
(2)
The alfalfa-grass system supported a higher level of beef produc­
tivity, greater N retention in animal tissue and less N excretion
per unit of N consumed than did the reed canarygrass system.
(3)
The higher rate of soil N accumulation in reed canarygrass,resulted
from a higher N input level into that system and also from the fact
that a higher proportion of herbage N was returned to the soil on
an area basis as vegetative residue and animal excretion in reed
canarygrass than in alfalfa-grass.
Ill
GENERAL DISCUSSION AND CONCLUSIONS
Legume-based pastures depending on symbiotic N fixation as a N
source (containing alfalfa, orchardgrass and smooth bromegrass) and Nfertilized reed canarygrass pastures were evaluated for their efficiency
of N use in vegetative and animal growth.
Two studies are reported in
which (1) the contribution of N fixation to the N econoniy of the alfalfagrass mixture was determined (Part I), and (2) the DM and N yield
responses to levels of N input and the relationships between animal gain
and N intake of the two pasture systems were compared (Part II).
The evidence presented in Part I indicates that N availability was
limiting herbage DM and N yield in alfalfa-grass, since these variables
were directly proportional to percentage legume and amount N fixed.
Indeed, the grass in the pasture lowest in legume percentage was chlorotic, lacking in vigorous growth and low in N concentration (soil pH was
6.3) in comparison to the grass in the other legume-based pastures.
The
DM yield of the legume-based pasture highest in legume percentage was
nearly as high as that of the reed canarygrass pastures.
The amount of
shoot N fixed in that alfalfa-grass pasture averaged 105 kg/ha-yr.
Napitupulu and Smith (1979) concluded that alfalfa-orchardgrass mixtures
should ideally be composed of 80% alfalfa to exploit the qualities of
low fiber, high protein and high mineral contents of that species, while
maintaining enough grass to minimize bloat, reduce soil erosion and weed
encroachment and hasten drying time for hay-making.
The results from
Part I support that conclusion in order to maximize the contribution of
112
N fixation to the N economy of the sward.
The seasonal trends in DM and N yield in reed canarygrass are pre­
sented in Figures A.6 and A.7.
The bimodal patterns are due to the tim­
ing of the split applications of N fertilizer.
The relative increase in
DM yield due to the second N application was not as great as the relative
increase in N yield, which resulted in N concentrations as high as 4%.
Dry matter accumulation rates during the hot, dry summer periods are
usually depressed in cool-season grasses such as reed canarygrass.
The
resulting concentrating effect on the herbage N content probably leads
to less efficient N use by the animals during that period compared to
other periods.
Since total N uptake after the second fertilizer appli­
cation was less than that of earlier in the season, more unabsorbed
fertilizer N contributed to the build-up of soil N reserves during the
second half of the season.
Some of that N was undoubtedly available in
subsequent years.
The question remains whether the N fertilizer rate in the second
application could be decreased while maintaining an economic level of
pasture productivity.
Mott et al. (1970) demonstrated that when N
fertilization of a tropical grass pasture in Brazil ceased after applying
200 kg/ha-yr for 8 years, the yield of total digestible nutrients de­
creased from 254% of the nil-N control during the final year of N input
to 252, 204, 151 and 130% during the succeeding 4 years, respectively.
Mott et al. (1970) concluded that after the soil N level of a grass
pasture has been built up, a lower annual input level would suffice to
maintain productivity because of N fertilizer carryover and recycling.
113a
A possible way to improve the efficiency of long-term N fertilizer use
in reed canarygrass would be to apply relatively high rates of N in the
first 2 to 4 years after pasture establishment, after which the second
N application is reduced or omitted with the animals being moved to a
productive legume-grass pasture.
More complete knowledge of the amounts
of N fertilizer carry-over and the net mineralization rates of soil N
under long-term reed canarygrass is needed in order to predict the
advisability of such an adjustment in fertilizer practice.
The build-up of soil N in a fertilized grass pasture is not neces­
sarily an inefficient use of N if the soil was initially infertile and
the land is to be used in rotation with other crops.
Indeed, it may be
more economical to follow such a practice if the price of N fertilizer
continues to increase.
An attempt was made to quantify some of the pools and pathways of
the N cycle in these pastures.
Nitrogen cycle schemes are presented in
Figures 1 and 2 for each N-forage system at the two grazing pressures.
The estimates are means of 2 years.
Nitrogen input from precipitation was determined by steam distill­
ing (Bremner, 1965c) rain samples collected at the pasture site.
cipitation N was equally divided between NH^^ and NOg" forms.
Pre­
The
amount of N input, 3 kg/ha-yr, was insufficient to replace the N removed
in animal gain.
Estimates of the total amount of N in the surface 30 cm
of soil (including roots) were based on an observed bulk density of 1.08
at 0 to 3 cm and an assumed bulk density of 1.47 at 3 to 30 cm.
of N by runoff, volatilization and leaching were not measured.
Losses
The
113b
11^ Product removal
Animal
I
134
123
Medium
grazing pressure
Symbiotic
fixation
70
Soil organic
N 7730
Legume
112
Available
soil N
Precipitation
Product removal
Animal
146
Symbiotic
fixation
70
1
Soil organic
N 8170
Heavy
grazing pressure
Grass
Available
soil N
Precipitation
Figure 1.
Annual flow rates of N in alfalfa-grass pastures at
two grazing pressures (kg/ha)» observed at McNay Memorial
Research Center, Lucas Co., Iowa
113c
Product removal
Animal
.212
200|
Medium
grazing pressure
Soil organic
N 8160
Available
soil N
18^
Fertilization
^Precipitation
Product removal
Animal
225
212
Y
Heavy
grazing pressure
Soil organic
N 8710
Grass
\
Available
soil N
180
3^
Fertilization
Figure 2.
Precipitation
Annual flow rates of N in reed canarygrass pastures at
two grazing pressures (kg/ha), observed at McNay Memorial
Research Center, Lucas Co., Iowa
114
other data are derived from the results of Parts I and II.
and grass pools do not include roots.
The legume
It was shown in Part II that the
difference in total soil N between the grazing pressure treatments
existed before the grazing trial began.
Two measurements that are lacking which would enhance our under­
standing of N use in these pastures are (1) the proportion of N fixed
not harvested in the legume shoot, since N losses from legumes are known
to occur by volatilization from foliage (Lemon and Van Houtte, 1980)
and root loss; and (2) the proportion of reed canarygrass N derived
from the N fertilizer within a given year.
In conclusion, the alfalfa-grass system was apparently more effi­
cient in the use of N for vegetative and animal production than the Nfertilized reed canarygrass system.
The annual cost of N fertilizer
was $90.00/ha for the reed canarygrass compared to zero for alfalfagrass.
Managing alfalfa-grass mixtures for high percentage legume com­
position can result in a more efficient and economical use of N in
pastures utilized by growing beef animals.
115
LITERATURE CITED
Allison, F. E. 1955. The enigma of soil nitrogen balance sheets.
Agron. 7:213-250.
Allison, F. E. 1966. The fate of nitrogen applied to soils.
Agron. 18:219-258.
Adv.
Adv.
Alios, H. F., and W. V. Bartholomew. 1955. Effect of available nitrogen
on symbiotic fixation. Soil Sci. Soc. Amer. Proc. 19:182-184.
Alios, H. F., and W. V. Bartholomew. 1959. Replacement of symbiotic
fixation with available N. Soil Sci. 87:61-67.
Amarger, N., A. Mariotti, F. Mariotti, J. C. Dunn, C. Bourguignon, and
B. Lagacherie. 1979. Estimate of symbiotically fixed N in field
grown soybeans using variations in
abundance. Plant Soil 52:
269-280,
Andrew, C. S., and C. Johansen. 1978. Differences between pasture
species in their requirements for nitrogen and phosphorus, p. 111127. 2n J. R. Wilson (ed.) Plant relations in pastures. CSIRO,
Melbourne.
Azevedo, J., and P. R. Stout. 1974. Farm-animal manures: An overview
of their role in the agricultural environment. University of Cal­
ifornia, College of Agriculture, Davis, California.
Ball, R., and D. R. Keeney. 1981. Nitrogen losses from urine-affected
areas of a New Zealand pasture under contrasting seasonal condi­
tions. XIV Int. Grassl. Cong., Lexington, Ky. Summaries of Papers,
p. 152.
Ball, R., D. R. Keeney, P. W. Theobald, and P. Nes. 1979. Nitrogen
balance in urine-affected areas of a New Zealand pasture. Agron.
J. 71:309-314.
Barrow, N. J., and L. J. Lambourne. 1962. Partition of excreted nitrogen,
sulphur and phosphorus between the feces and urine of sheep being f e d
pasture. Aust. J. Agric. Res. 13:461-471.
Beever, D. E., and D. J. Thompson. 1977. The potential of protected pro­
teins in ruminant nutrition, p. 66-82. Li W. Haresign and D. Lewis
(eds.) Recent advances in animal nutrition. Butterworths, London.
Beever, D. E., S. B. Cammell, and A. Wallace. 1974. The digestion of
fresh, frozen and dried perennial ryegrass. Proc. Nutr. Soc.
33:73k.
116
Black, C. A. 1968. Soil-plant relationships.
Inc., New York.
John Wiley and Sons,
Blaser, R. E. 1964. Symposium on forage utilization: Effects of fer­
tility levels and stage of maturity on forage nutritive value. J.
Anim. Sci. 23:246-253.
Blaser, R. E., R. C. Hammer, Jr., H. T. Bryant, C. M. Kincaid, W. H.
Skrdla, T. H. Taylor, and W. L. Griffeth. 1956. The value of
forage species and mixtures for fattening steers. Agron. J. 48:
508-511.
Blaxter, K. L. 1978. Limits to animal production.
Agric. Sci. 44:97-102.
J. Aust. Inst.
Blaxter, K. L., F. W. Wainman, P. J. S. Dewey, J. Davidson, H. Denerley,
and J. B. Gunn. 1971. The effects of nitrogenous fertilizer on
the nutritive value of artificially dried grass. J. Agric. Sci.,
Camb. 76:307-319.
Boulter, D., and E. Derbyshire. 1978. The general properties, class­
ification and distribution of plant proteins, p. 3-24. In_G. Norton
(ed.) Plant proteins. Butterworths, London.
Bremner, J. M. 1965a. Isotope-ratio analysis of nitrogen in nitrogen15 tracer investigations. Iji C. A. Black (ed.) Methods of soil
analysis. Part II. Agronomy 9:1256-1287. Am. Soc. Agron., Madison.
Bremner, J. M. 1965b.
of soil analysis.
Madison, Wis.
Total nitrogen. %n C. A. Black (ed.) Methods
Part II. Agronomy 9:1149-1178. Am. Soc. Agron.,
Bremner, J. M. 1965c. Inorganic forms of nitrogen. In C. A. Black (ed.)
Methods of soil analysis. Part II. Agronomy 19:1779-1237. Am.
Soc. Agron., Madison, Wis.
Bremner, J. M. 1977. Use of nitrogen-tracer techniques for research on
nitrogen fixation, p. 334-352. In^ A. Ayanaba and P. J. Dart (eds.)
Biological nitrogen fixation in farming systems of the tropics.
John Wiley and Sons, Inc., New York.
Bremner, J. M., and A. M. Blackmer. 1978. Nitrous oxide: Emission
from soils during nitrification of fertilizer nitrogen. Science
199:295-296.
Bremner, J. M., and M. A. Tabatabai.
soils and soil-derived nitrate.
1973. Nitrogen-15 enrichment of
J. Environ. Qual. 2:363-365.
117
Brock, J. L., J. H. Hoglund, and R. H. Fletcher. 1981. Effects of
grazing management on seasonal variation in nitrogen fixation. XIV
Int. Grassl. Cong., Lexington, KY. Summaries of Papers, p. 150.
Brougham, R. W., P. R. Ball, and W. M. Williams. 1978. The ecology
and management of white clover-based pastures, p. 309-324. Iji
J. R. Wilson (ed.) Plant relations in pastures. CSIRO, Melbourne.
Bryan, W. B., W. F. Wedin, and R. L. Vetter. 1970. Evaluation of
reed canarygrass and tall fescue as spring-summer and fall-saved
pasture. Agron. J. 62:75-80.
Burns, J. C., L. Goode, H. D. Gross, and A. C. Linnerud. 1973. Cow
and calf gains on Ladino clover-tall fescue and tall fescue,
grazed alone and with Coastal bermudagrass. Agron. J. 65:877-880.
Burris, R. H. 1974. Methodology,
biology of nitrogen fixation.
p. 9-33. %n A. Quispel (ed.) The
North-Holland Publ. Co., Amsterdam.
Burris, R. H. 1975. Nitrogen fixation, p. 887-908. Li J. Bonner and
J. E. Varner (eds.) Plant biochemistry. Academic Press, New York.
Burris, R. H. 1980. The global nitrogen budget—science or seance?
p. 7-27. Iji W. E. Newton and W. H. Orme-Johnson (eds.) Nitrogen
fixation. Volume I. University Park Press, Baltimore.
Burroughs, W., D. K. Nelson, and D. R. Mertens. 1975. Protein physi­
ology and its application in the lactating cow: The metabolizable
protein feeding standard. J. Anim. Sci. 41:933-944.
Carter, L. P., and J. M. Scholl. 1962. Effectiveness of inorganic N
as a replacement for legumes grown in association with forage
grasses. I. Dry matter production and botanical yield. Agron. ,
J. 54:161-163.
CAST.
1980. Foods from animals: Quantity, quality and safety. Council
for Agricultural Science and Technology, Report No. 82. Ames, lA.
Chalupa, W. 1975. Amino acid nutrition of growing cattle, p. 175-194.
2n Tracer studies on nonprotein nitrogen for ruminants. Vol. II.
FAO/IAEA, Vienna.
Chalupa, W. V., J. L. Cason, and B. R. Baumgardt. 1961. Nutritive
value of reed canarygrass as hay when grown with various nitrogen
levels. J. Dairy Sci. 44:874-878.
Chamblee, D. S. 1972. Relationships with other species in a mixture.
%n C. H. Hanson (ed.) Alfalfa science and technology. Agronomy
15:211-228. Amer. Soc. Agron., Madison, Wis.
118
Chan, Wing-To, and A. F. MacKenzie. 1971. Effects of shading and
nitrogen on growth of alfalfa-grass pastures. Agron. J. 63:667-669.
Cheng, H. H., J. M. Bremner, and A. P. Edwards. 1964. Variations of
nitrogen-15 abundance in soils. Science 146:1574-1575.
Chestnutt, D. M. B. 1972. The effects of white clover and applied
nitrogen on the nitrogen content of various grass/clover mixtures.
J. Br. Grassl. Soc. 27:211-216.
Chibnall, A. C., M. W. Reese, and J. W. H. Lugg. 1963. The amino acid
composition of leaf proteins. J. Sci. Food Agric. 14:234-239.
Child, R. D., and E. K. Byington, (eds.). 1981. Potential of the world's
forages for ruminant animal production. Second Edition. Winrock
Int. Livestock Res. Training Cen., Morrilton, Arkansas.
Conrad, H. R., R. W. Van Keuren, and J. W. Hibbs. 1978. Utilization
of alfalfa protein in ruminants, p. 11-14. Iji Proc. 8th Ann.
Alfalfa Symp. Univ. of Minn., St. Paul.
Cralle, H. T., and G. H. Heichel. 1978. Nitrogen fixation and regrowth
of alfalfa and birdsfoot trefoil following harvest. Agron. Abstr.
70:72.
DanEik, J. 1976. Effect of sowing rate, date and method on lucerne
yield. (In Polish). Rostlinnd Vjrroba 49:63-72.
Dart, P. J., and F. V. Mercer. 1965. The effect of growth temperature,
level of ammonium nitrate, and light intensity on the growth and
nodulation of cowpea (Vigna sinensis Endl. ex Hassk.). Aust. J.
Agric. Res. 16:321-345.
Dart, P. J., and D. C. Wildon. 1970. Nodulation and nitrogen fixation
by Vigna sinensis and Vicia atropurpurea: The inf1uence of con­
centration, form, and site of application of combined nitrogen.
Aust. J. Agric. Res. 21:45-46.
Davis, H. L., and E. A. N. Greenwood. 1972. Liveweight and wool growth
of sheep in response to the quantity and botanical composition of
annual pasture induced by nitrogen fertilizer and stocking treat­
ments. Aust. J. Agric. Res. 23:1101-1111.
Day, J. M., and P. J. Dart. 1969. Light intensity, nodulation and
nitrogen fixation. Rothamstad Exp. Sta. Rep. 1696:105. Harpenden,
UK.
119
Decker, A. M., G. A. Jung, J. B. Washko, D. D. Wolf, and M. J. Wright.
1967. Management and productivity of perennial grasses in the
northeast: Reed canarygrass. West Virginia Univ. Agric. Exp. Sta.
Bull. 550T.
Denmead, 0. T., J. R. Simpson, and J. R. Freney. 1974. Ammonia flux
into the atmosphere from a grazed pasture. Science 185:609-610.
Denmead, 0. T., J. R. Freney, and J. R. Simpson. 1976. A closed
ammonia cycle within a plant canopy. Soil Biol. Biochem. 8:161-164.
DePeters, E. J., and E. M. Kesler. 1980. Partial replacement of con­
centrate by high quality hay for cows in early lactation. J.
Dairy Sci. 63:936-944.
Dilworth, M. J. 1966. Acetylene reduction by nitrogen-fixing prepara­
tions from Clostridium pasturianum. Biochim. Biophys. Acta 127:
285-294.
Dilz, K., and E. G. Mulder. 1962. Effect of associated growth on yield
and nitrogen content of legume and grass plants. Plant Soil 16:
229-237.
Dilz, K.. and J. W. Woldendorp. 1960. Distribution and nitrogen balance
of T5N-labelled nitrate applied on grass swards. Proc. VIII Int.
Grassl. Cong. 8:150-152.
Doak, B. W. 1952. Some chemical changes in the nitrogenous constitu­
ents of urine when voided on pasture. J. Agric. Sci., Camb. 42:
162-171.
Donald, C. M. 1963. Competition among crop and pasture plants.
Agron, 15:1-118.
Drake, M., J. Vengris, and W. C. Colby. 1951.
of plant roots. Soil Sci. 72:139-147.
Adv.
Cation exchange capacity
Eckart, J. F., and C. A. Raguse. 1980. Effects of diurnal variation in
light and temperature on the acetylene reduction (nitrogen fixation)
of subterranean clover. Agron. J. 72:519-523.
Edmeades, D. C., and K. M. Goh. 1978. Symbiotic nitrogen fixation in a
sequence of pastures of increasing age measured by a '®N-dilution
technique. N. Z. J. Agric. Res. 21:623-628.
Elgin, J. H. Jr., and J. 0. Legg. 1977. Measuring N fixation in seven
alfalfa cultivars. Proc. 2nd Eastern Forage Improv. Conf. Penn­
sylvania State Univ.
120
Engin, M., and J. I. Sprent. 1973. Effects of water stress on growth
and nitrogen-fixing activity of Tri folium repens. New Phvtol. 72:
117-126.
Eskeland, B., W. H. Pfander, and R. L. Preston. 1973. Utilization of
volatile fatty acids and glucose for protein deposition in lambs.
Br. J. Nutr. 29:347-355.
Field, T. R., and R. Ball. 1978.
Agron. Soc. N. Z. 8:129-133.
Tactical use of fertilizer N.
Proc.
Fitzhugh, H. A., H. J. Hodgson, 0. J. Scoville, T. D. Nguyen, and T. C.
Byerly. 1978. The role of ruminants in support of man. Winrock
Int. Livestock Res. Training Cen. Morrilton, AK.
Fox, D. G., R. G. Crickenberger, and W. G. Bergen. 1977. A net protein
system for predicting protein requirements and feed protein values
for growing and finishing cattle. Michigan Agric. Exp. Sta. Res.
Rep. 328. p. 163-176.
Fried, M., and H. Broeshart. 1975. An independent measurement of the
amount of nitrogen fixed by a legume crop. PI. Soil 43:707-711.
Fried, M., and V. Middleboe. 1977. Measurement of the amount of
nitrogen fixed by a legume crop. Plant and Soil 47:713-715.
Fribourg, H. A., J. B. McLaren, K. M. Barth, J. M. Bryan, and J. T.
Connell. 1979. Productivity and quality of bermudagrass and
orchardgrass-Ladino clover pastures for beef steers. Agron. J.
71:315-320.
George, J. R., C. L. Rhykerd, C. H. Noller, J. E. Dillon, and J. C. Burns.
1973. Effect of N fertilization on dry matter yield, total-N, N
recovery, and nitrate-N concentration of three cool-season forage
grass species. Agron. J. 65:211-216.
Gibson, A. H. 1976. Recovery and compensation by nodulated legumes to
environmental stress, p. 355-403. %n P. S. Nutman (ed.) Symbi­
otic nitrogen fixation in plants. Cambridge Univ. Press, Cambridge.
Gibson, A. H. 1977. The influence of the environment and managerial
practices on the legume-rhizobium symbiosis, p. 393-450. In^
R. W. F. Hardy and A. H. Gibson (eds.) A treatise on dinitrogen
fixation. John Wiley and Sons, Interscience, New York.
Gibson, A. H., and P. S. Nutman. 1960. Studies on the physiology of
nodule formation. VII. A reappraisal of the effect of preplanting. Annals of Bot. N.S. 24:420-433.
121
Goh, K. M., D. C. Edmeades, and B. W. Robinson. 1978. Field measure­
ments of symbiotic nitrogen fixation in an established pasture using
acetylene reduction and a
method. Soil Biol. Biochem. 10:13-20.
Grenet, E., and C. Demarquilly. 1977. Utilization of fresh herbage
nitrogen by growing sheep: Influence of herbage growth stage,
herbage species, nitrogen fertilization and barley supplementation.
(In French) Ann. Zootech. 26:481-501.
Hall, R. L. 1974. Analysis of the nature of interference between plants
of different species. II. Nutrient relations in a Nandi Setaria
and Greenleaf Desmodium association with particular reference to
potassium. Aust. J. Agric. Res. 25:749-756.
Hall, R. L. 1978. The analysis and significance of competitive and
non-competitive interference between species, p. 163-174. In.
J. R. Wilson (ed.) Plant relations in pastures. CSIRO, Melbourne.
Halliday, J., and J. S. Pate. 1976. The acetylene reduction assay as
a means of studying nitrogen fixation in white clover under sward
and laboratory conditions. J. Br. Grassl. Soc. 31:29-35.
Hanawalt, R. B. 1969. Environmental factors influencing the sorption
of atmospheric ammonia by soils. Soil Sci. Soc. Am. Proc. 33:
231-234.
Hardy, R. W. F., and U. D. Havelka. 1976. Photosynthate as a major
factor limiting N fixation by field-grown legumes with emphasis on
soybeans, p. 421-439. jji P. S. Nutman (ed.) Symbiotic nitrogen
fixation in plants. Cambridge University Press, Cambridge.
Hardy, R. W. F., and R. D. Holsten. 1978. Methods for measurement of
dinitrogen fixation, p. 451-486. In^ R. W. F. Hardy and A. H.
Gibson (eds.) A treatise on dinitrogen fixation. Section IV:
Agronomy and ecology. John Wiley and Sons, Interscience, New York.
Hardy, R. W. F., R. D. Holsten, E. K. Jackson, and R. D. Burns. 1968.
The acetylene-ethylene assay for N2 fixation: Laboratory and field
evaluation. Plant Physiol. 43:1185-1207.
Harper, J. E. 1974. Soil and symbiotic nitrogen requirements for optimum
soybean production. Crop Sci. 14:255-260.
Harris, W. 1978. Defoliation as a determinant of the growth, per­
sistence and composition of pasture, p. 67-85.
J. R. Wilson
(ed.) Plant relations in pastures. CSIRO, Melbourne.
Hauck, R. D., and J. M. Bremner. 1976. Use of tracers for soil and
fertilizer nitrogen research. Adv. Agron. 28:219-266.
122
Haystead, A., and A. G. Lowe. 1977. Nitrogen fixation by white clover
in hill pasture. J. Br. Grassl. Soc. 32:57-63.
Haystead, A., and C. Marriott. 1978. Fixation and transfer of nitrogen
in a white-clover-grass sward under hill conditions. Ann. Applied
Biol. 88:453-457.
Haystead, A., and C. Marriott. 1979. Transfer of legume nitrogen to
associated grass. Soil Biol. Biochem. 11:99-104.
Heal, 0. W. 1979. Consequences of non-assimilation of nitrogen by
plants, p. 625-636.
E. J. Hewitt and C. V. Cutting (eds.)
Nitrogen assimilation of plants. Academic Press, London.
Heath, S. B. 1976. The efficiency of protein production by crop plants,
p. 147-158. %n A. N. Duckman, J. G. W. Jones, and E. H. Roberts
(eds.) Food production and consumption: The efficiency of human
food chains and nutrient cycles. North-Holland Publ. Co., Oxford,
U.K.
Heichel, G. H. 1978. Stabilizing agricultural energy needs: Role of
forages, rotations, and nitrogen fixation. J. Soil Water Cons.
33:279-282.
Heichel, G. H., and C. P. Vance. 1979. Nitrate-N and Rhizobium strain
roles in alfalfa seedling nodulation and growth. Crop Sci. 19:
512-518.
Heichel, G. H., D. K. Barnes, and C. P. Vance. 1978. Genetic variation
in dinitrogen fixation of alfalfa clones and populations. Agron.
Abstracts 70:76.
Heichel, G. H., D. K. Barnes, and C. P. Vance. 1980. Comparative
nitrogen fixation of alfalfa, birdsfoot trefoil, and red clover.
Agron. Abstracts 72:82.
Heichel, G. H., D. K. Barnes, and C. P. Vance. 1981. Nitrogen fixation
of alfalfa in the seeding year. Crop Sci. 21:330-335.
Hemsley, J. A., J. P. Hogan, and R. H. Weston. 1970. Protection of
forage protein from ruminai degradation. Proc. XI Int. Grassl.
Cong. 11:703-706.
Henzell, E. F. 1970. Problems in comparing the N economies of legume
based and N fertilized pasture systems. Proc. XI Int. Grassl.
Cong. 11:A112-A120.
Henzell, E. F. 1981. Contribution of forages to worldwide food produc­
tion: Now and in the future. XIV Int. Grassl. Cong., Lexington,
KY. Summaries of Papers, p. 4.
123
Henzell, E. F., and P. J. Ross. 1973. The nitrogen cycle of pasture eco­
systems. p. 227-246. 2n G. W. Butler and R. W. Bailey (eds.)
Chemistry and biochemistry of herbage. Vol. 2. Academic Press,
New York.
Hodgson, H. J. 1976.
26:625-630.
Forages, ruminant livestock, and food.
Bioscience
Hodgson, H. J. 1981. Forages, food, and energy, p. 14-20. In Rep.
27th Alfalfa Improvement Conf., Madison, Wis., 1980. North Central
Region of the SEA, USDA, Peoria, IL.
Hogan, J. P., and R. H. Weston. 1981. Laboratory methods for protein
evaluation, p. 75-87. In^ J. L. Wheeler and R. D. Mochrie (eds.)
Forage evaluation: Concepts and techniques. AFGC/CSIRO, Lexington,
KY/Melbourne, Australia.
Hoglund, J. H., and J. L. Brock. 1978. Regulation of nitrogen fixation
in a grazed pasture. N. Z. J. Agric. Res. 21:73-82.
Hoglund, J. H., J. R. Crush, J. L. Brock, R. Ball, and R. A. Carran.
1979. Nitrogen fixation in pasture. XII. General discussion.
N. Z. J. Agric. 7:45-51.
Hojjati, S. M., W. C. Templeton, Jr., and T. H. Taylor. 1978. Nitrogen
fertilization in establishing forage legumes. Agron. J. 70:429-433.
Holter, P. 1979. Effect of dung-beetles (Aphodius spp.) and earthworms
on the disappearance of cattle dung. Oikos 32:393-402.
Hsiao, T. C. 1973. Plant responses to water stress.
Phys. 24:519-570.
Ann. Rev. Plant
Hubbard, W. A., and H. H. Nicholson. 1968. Reed canarygrass vs. grasslegume mixtures under irrigation as pasture for sheep. 0. Range
Management 21:171-174.
Hutton, J. B. 1970. Crops or grass for efficient low-cost livestock
production. Proc. XI Int. Grassl. Cong. 11:A78-A94.
Jones, M. B., and R. G. Woodmansee. 1979. Geochemical cycling in annual
grassland ecosystems. The Bot. Rev. 45:111-144.
Kay, M. 1976. Meeting the energy and protein requirements of the grow­
ing animal, p. 225-269. %n H. Swan and W. H. Broster (eds.)
Principles of cattle production. Buttersworth, London.
Kolenbrander, 6. J. 1969. Nitrate content and nitrogen loss in drainwater. Neth. J. Agric. Res. 17:246-255.
124
Krueger, C. R., and J. M. Scholl. 1970. Performance of smooth bromegrass, orchardgrass, and reed canarygrass grown at five N levels
and with alfalfa. Univ. Wis. Coll. Agric. Life Sci. Res. Div. Res.
Rep. 69.
Langille, J. E., and F. W. Calder. 1971. Effect of harvesting prac­
tices on foliage and root development, digestibility, cold hardi­
ness, and modulation of birdsfoot trefoil. Can. J. Plant Sci.
51:499-504.
Lawn, R. J., and W, A. Brun. 1974. Symbiotic nitrogen fixation in soy­
beans. I. Effect of photosynthetic source-sink manipulations.
Crop Sci. 14:11-16.
Lemon, E., and R. Van Houtte. 1980.
surface. Agron. J. 72:876-883.
Ammonia exchange at the land
Lewis, D., and R. M. Mitchell. 1976. Amino acid requirements of rumi­
nants. p. 417-424. In_ D. J. A. Cole (ed.) Protein metabolism
and nutrition. Buttersworth, London.
Linehan, P. A., and J. Lowe. 1960. Yielding capacity and grass/clover
ratio of herbage swards as influenced by fertilizer treatments.
Proc. VIII Int. Grassl. Cong. 8:133-137.
Lockyer, D. R., and D. W. Cowling. 1977. Non-symbiotic nitrogen fixa­
tion in some soils of England and Wales. J. Br. Grassl. Soc.
32:7-11.
MacLeod, L. D. 1965. Effect of nitrogen and potassium on the yield,
botanical composition and competition for nutrients in three alfalfa
grass associations. Agron. J. 57:129-134.
MacRae, J. C. 1976. Utilization of the protein of green forage by rum­
inants at pasture, p. 93-98.
T. M. Sutherland, J. R. McWilliams
and R. A. Leng (eds.) From plant to animal protein. (Reviews in
Rural Science II). Univ. New England Publ. Unit, Armadale,
Australia.
MacRae, J. C., and Ulyatt. 1974. Quantitative digestion of fresh
herbage by sheep. II. Sites of digestion of some nitrogenous con­
stituents. J. Agric. Sci., Camb. 82:309-319.
Marten, G. C. 1973. Alkaloids in reed canarygrass. p. 15-31. hi
A. G. Matches (ed.) Anti-quality components of forages. Spec.
Pub. No. 4. Crop Sci. Soc. Am., Madison, Wis.
125
Marten, G. C., and R. F. Barnes. 1979. Prediction of energy digesti­
bility of forages with in vitro rumen fermentation and fungal
enzyme systems, p. 67-71. Jn^ Proc. Int. Workshop on Standardiza­
tion Anal. Methodology Feeds. IDRC-134e, Ottawa, Canada.
Marten, G. C., and J, D. Donker. 1966. Animal excrement as a factor
influencing acceptability of grazed forage. Proc. X Int. Grassl.
Cong. 10:359-363.
Marten, 6. C., and J. D. Donker. 1968. Determinants of pasture value
of Phalaris arundinaceae L. vs. Bromus inermis Leyss. Aqron. J.
60:703-705.
Marten, G. C., and J. D. Donker. 1974. High quality alfalfa will elim­
inate or reduce feed need for costly protein supplements in dairy
rations. Proc. 4th Annual Alfalfa Symp., Madison, Wis. p. 51-57.
Marten, G. C., and M. E. Heath. 1973. Reed canarygrass. p. 263-276.
In M. E. Heath, D. S. Metcalfe, and R. F. Barnes (eds.) Forages.
Third Ed. Iowa St. Univ. Press, Ames, Iowa.
Marten, G. C., and A. W. Hovin. 1980. Harvest schedule, persistence,
yield and quality interactions among four perennial grasses. Agron.
J. 72:378-387.
Marten, G. C., A. B. Simons, and J. R. Frelich. 1974. Alkaloids of
reed canarygrass as influenced by nutrient supply. Agron. J.
66:363-368.
Marten, G. C., R. M. Jordan, and A. W. Hovin. 1976. Biological signifi­
cance of reed canarygrass alkaloids and associated palatability
variation to grazing sheep and cattle. Agron. J. 68:909-914.
Marten, G. C., R. M.. Jordan, and A. W. Hovin. 1981. Improved lamb
performance associated with breeding for alkaloid reduction in
reed canarygrass. Crop. Sci. 21:295-298.
Martin, W. E., and L. J. Berry. 1970. Use of nitrogenous fertilizers on
California rangeland. Proc. XI Int. Grassl. Cong. 11:817-822.
Marum, P., A. W. Hovin, G. C. Marten, and J. S. Shenk. 1979. Genetic
variability for cell wall constituents and associated quality
traits in reed canarygrass. Crop Sci. 19:355-360.
Matsushima, J. K. 1972. Role in feedlot feeding. Iji C. H. Hanson
(ed.) Alfalfa science and technology. Agronomy 15:631-640.
Am. Soc. Agron., Madison, Wis.
126
McAuliffe, C., D. S. Chamblee, H. Uribi-Arango, and W. W. Woodhouse,
Jr. 1958. Influence of inorganic nitrogen on nitrogen fixation
by legumes as revealed by N'S. Agron. J. 50:334-337.
McCloud, D. E., and G. 0. Mott. 1953. Influence of association upon
the forage yield of legume-grass mixtures. Agron. J. 45:61-65.
McKee, G. W. 1962.
Effects of shading and plant competition on
seedling growth and nodulation in birdsfoot trefoil. Pennsylvania
Agric. Exp. Stn. Bull. 688.
Minson, D. J. 1976. Nutritional significance of protein in temperate
and tropical pastures, p. 27-30. Iji T. M. Sutherland, J. R.
McWilliam, and R. A. Leng (eds.) From plant to animal protein
(Reviews in Rural Science II). Univ. of New England Publ. Unit,
Armadale, Australia.
Minson, D. J. 1981. Characteristics of the nutritionally ideal grazing
plant, p. 543-550. Iji J. L. Wheeler and R. D. Mochrie (eds.)
Forage evaluation: Concepts and techniques. AFGC/CSIRO, Lexington,
KY/Melbourne, Australia.
Moe, P. 6., J. V. Mannering, and C. B. Johnson. 1967. Loss of fertil­
izer nitrogen in surface runoff water. Soil Sci. 104:389-394.
Moore, C. K., H. E. Amos, J. J. Evans, R. S. Lowry, and D. Burdick.
1980. Nitrogen balance, abomasal crude protein and amino acids in
wethers fed formaldehyde-treated coastal bermudagrass and infused
with methionine, glucose or monensin. J. Anim. Sci. 50:1145-1159.
Mott, 6. 0. 1960. Grazing pressure and the measurement of pasture pro­
duction. Proc. VIII Intl. Grass!. Cong. 8:606-611.
Mott, G. 0. 1974. Nutrient recycling in pastures, p. 323-362. In
C. A. Mays (ed.) Forage fertilization. ASA, CSSA, SSSA, MaiTson,
Wis.
Mott, G. 0., L- R. Quinn, and W. V. A. Bisschoff. 1970. The retention of
nitrogen in a soil-plant-animal system in Guinea grass (Panicum
maximum) pastures in Brazil. Proc. XI Int. Grass!. Cong. 11:414-416.
Mouat, M. C. H., and T. W. Walker. 1959a, Competition for nutrients
between grass and white clover. I. Effect of grass species and
nitrogen supply. Plant Soil 11:30-40.
Mouat, M. C. H., and T. W. Walker. 1959b. Competition for nutrients,
between grass and white clover. II. Effect of root cation exchange
capacity and rate of emergence of associated species. Plant and
Soil 11:41-52.
127
Moustafa, E., R. Ball, and T. R. 0. Field. 1969. The use of acetylene
reduction to study the effect of nitrogen fertilizer and defolia­
tion on nitrogen fixation by field-grown white clover. N. Z. J.
Agric. Res. 12:691-696.
Mulder, E. G. 1952. Fertilizer vs. legume nitrogen for grasslands.
Proc. VI Int. Grassl. Cong. 6:740-748.
Munro, J. M. M., and D. A. Davies. 1974. Potential pasture production
in the uplands of Wales. 5. The nitrogen contribution of white
clover. J. Brit. Grassl. Soc. 29:213-223.
Napitupulu, J. A., and D. Smith.
1979. Changes in herbage chemical
composition due to proportion of species in alfalfa-orchardgrass
mixtures. Comm. Soil Sci. Plant Anal. 10:565-577.
National Research Council. 1976. Nutrient requirements of domestic
animals. No. 4. Nutrient requirements of beef cattle. National
Academy of Sciences--National Research Council, Wash., D.C.
Nicol, A. M., and J. W. McLean. 1970. The influence of pasture species
and exogenous oestrogen treatment on the live-weight gain and car­
cass composition of lambs. N. Z. J. Agric. Res. 13:385-394.
Niehaus, M. N. 1971. Effect of fertilizer on yield, crude protein con­
tent, and in vitro dry matter disappearance in Phalaris arundinacea
L. Agron. J. 63:793-794.
Nelson, C. 0. 1973. An analysis of runoff from nitrogen-fertilized
grassland. Unpublished M.S. Thesis. Library, Iowa State Univ.,
Ames, Iowa.
Nutman, P. S. 1976. IBP field experiments on nitrogen fixation by nodu­
lated legumes, p. 211-238. Iji P. S. Nutman (ed.) Symbiotic
nitrogen fixation by plants. Cambridge Univ. Press, London.
Oghoghorie, C. G. 0., and J. S. Pate. 1971. The nitrate stress syndrome
of the nodulated field pea (Pisum arvense L.). Techniques for
measurement and evaluation in physiological terms. Plant Soil
Special Volume: 185-202.
Paulson, W. H., J. M. Scholl, H. W. Ream, R. J. Vatthauer, and A. W.
Young. 1977. More beef from improved pastures of the drift!ess
area of the Upper Mississippi Valley. Univ. Wis. Coll. Agric. Life
Sci. Res. Div. Res. Rep. R2893.
Peterson, M. L., and L. E. Bendixen. 1961. Plant competition in rela­
tionship to nitrogen econon\y. Agron. J. 53:45-49.
128
Petritz, D. C., V. L. Lechtenburg, and W. H. Smith. 1980. Performance
and economic returns of beef cows and calves grazing grass-legume
herbage. Agron. J. 72:581-584.
Phillips, D. A., and J. P. Bennett. 1978. Measuring symbiotic nitrogen
fixation in rangeland plots of Trifolium subterraneum L. and
Bromus mollis L. Agron. J. 70:671-674.
Reed, K. F. M. 1981. A review of legume-based vs. nitrogen-fertilized
pasture systems for sheep and beef cattle, p. 401-417.
L.
Wheeler and R. D. Mochrie (eds.) Forage evaluation: Concepts and
techniques. AFGC/CSIRO, Lexington, KY/Melbourne, Australia.
Reid, D. 1970. The effects of a wide range of nitrogen application
rates on the yields from a perennial ryegrass sward with and with­
out white clover. J. Agric. Sci., Camb. 74:227-240.
Reid, D. J., D. J. Hunt, and H. D. Ritchie. 1975. A 3-year evaluation
of seven different forage mixtures by measuring animal gain on
pasture at the U. P. Research and Demonstration Center, p. 132139. Michigan Agric. Exp. Sta. Res. Rep. 288.
Reid, R. L., and G. A. Jung. 1965. Factors affecting the intake and
palatability of forages for sheep. Proc. IX Int. Grassl. Cong.
9(1):863-869.
Reid, R. L., A. J. Post, and G. A. Jung. 1970. The nutritive evaluation
of orchardgrass pasture under different levels of N-fertilization.
Proc. N. Z. Grassl. Assoc. 32:87-95.
Rennie, D. A., E. A. Paul, and L. E. Johns. 1976. Natural nitrogen-15
abundance of soils and plant samples. Can. J. Soil Sci. 56:43-50.
Roberts, J. L., and F. R. Olson. 1942. Interrelationships of legumes
and grasses grown in association. J. Am. Soc. Agron. 34:695-701.
Robinson, R. R., and V. G. Sprague. 1947. The clover populations and
yields of a Kentucky bluegrass sod as affected by N fertilization,
clipping treatments and irrigation. Agron. J. 39:107-116.
Rohweder, D. A. 1963. The nature and productivity of southern Iowa
pastures as affected by soil type, slope, erosion, fertility and
botanical composition. Unpublished Ph.D. Dissertation. Library,
Iowa State Univ., Ames, Iowa.
Schaller, F. W., W. F. Wedin, and I. T. Carlson. 1972. Reed canarygrass. Iowa State Univ. Coop. Ext. Serv. Pm-538. 4 p.
129
Schmid, A. R., L. J. Elling, A. W. Hovin, W. E. Lueschen, W. W. Nelson,
D. L. Rabas, and D. D. Warner. 1979. Pasture species clipping
trials in Minnesota. Minnesota Agric. Exp. Stn. Tech. Bull. 317.
Schubert, K. R., and H. J. Evans. 1976. Hydrogen evolution: A major
factor affecting the efficiency of nitrogen fixation in nodulated
symbionts. Proc. Natl. Acad. Sci. USA 73:1207-1211.
Sears, P. D., V. C. Goodall, R. H. Jackman, and 6. S. Robinson. 1965.
Pasture growth and soil fertility. VIII. The influence of grasses,
white clover, fertilizers, and the return of herbage clippings on
pasture production of an impoverished soil. N. Z. J. Agric. Res.
8:270-283.
Seetin, M. W., and D. K. Barnes. 1977. Variation among alfalfa geno­
types for rate of acetylene reduction. Crop Sci. 17:783-787.
Shaw, P. G., J. S. Brockman, and K. M. Wolton. 1966. The effect of
cutting and grazing on the response of grass/white-clover swards
to fertilizer nitrogen. Proc. X Int. Grassl. Cong. 10:240-244.
Simpson, J. R. 1976. Transfer of nitrogen from three pasture legumes
under periodic defoliation in a field environment. Aust. J. Exp.
Agric. An. Husb. 16:863-870.
Sinclair, A. G. 1973. Non-destructive acetylene reduction assay of
nitrogen fixation applied to white clover plants growing in soil.
N. Z. J. Agric. Res. 16:263-270.
Sinclair, A. G., R. B. Hannagan, P. Johnstone, and A. K. Hardacre. 1978.
Evaluation of non-destructive acetylene reduction assay of N fixa­
tion for pasture legumes grown in pots. N. Z. J. Exp. Agric. 6:
65-68.
Slesser, M. 1973. Energy subsidy as a criterion in food policy plan­
ning. J. Sci. Fd. Agric. 24:1193-1207.
Smith, G. R., H. L. Peterson, and W. E. Knight. 1981. Effect of plant
genotype on dinitrogen (C2H2) fixation efficiency in Tri folium
incarnatum L. XIV Int. Grassl. Cong., Lexington, KY. Summaries
of Papers, p. 149.
Soulides, D. A., and F. E. Clark. 1958. Nitrification in grassland
soils. Soil Sci. Soc. Amer. Proc. 22:308-311.
Stewart, D. A., and D. M. B. Chestnutt. 1974. Some mathematical aspects
of the clover, grass, and fertilizer nitrogen relationships. J.
Agric. Sci., Camb. 82:331-338.
130
Tabatabai, M. A., and J. M. Bremner. 1970. Use of the Leco automatic
70-second carbon analyzer for total carbon analysis of soils. Soil
Sci. Soc. Amer. Proc. 34:608-610.
Tabatabai, M. A., and J. M. Laflen. 1976. Nutrient content of precipi­
tation over Iowa. Water, Air, and Soil Pollution 6:361-373.
Templeton, W. C., Jr. 1976. Legume nitrogen versus fertilizer nitrogen
for cool-season grasses, p. 35-54. In^ C. S. Hoveland, W. E.
Knight, and G. C. Marten (eds.) Biological N fixation in foragelivestock systems. Spec. Publ. 28. Am. Soc. Agron., Madison, Wis.
Thomas, G. W. 1981. Resource allocation for animal grassland systems.
XIV Int. Grassl. Cong., Lexington, KY. Summaries of Papers, p. 9.
Thornton, H. G., and H. Nicol. 1938. Reduction of nodule numbers and
growth produced by the addition of sodium nitrate to lucerne in
sand culture. J. Agric. Sci., Camb. 26:173-188.
Thornton, R. F., and D. J. Minson. 1973. The relationship between
apparent retention time in the rumen, voluntary intake and apparent
digestibility of legume and grass diets in sheep. Aust. J. Agric.
Res. 24:889-898.
Vallis, I. 1978. Nitrogen relationships in grass/legume mixtures.
p. 190-201. jn John R. Wilson (ed.) Plant relations in pastures.
CSIRO, Melbourne.
Vallis, I., K. P. Haydock, P. J. Ross, and E. F. Henzell. 1967. Isotopic studies on the uptake of nitrogen by pasture plants. III.
The uptake of small additions of 15N-labelled fertilizer by Rhodes
grass and Townsville lucerne. Aust. J. Agric. Res. 18:865-877.
Vallis, I., E. F. Henzell, and T. R. Evans. 1977. Uptake of soil
nitrcjen by legumes in mixed swards. Aust. J. Agric. Res. 28:
413-425.
Van Keuren, R. W., and W. W. Heinemann. 1958. A comparison of legumegrass mixtures and grass under irrigation as pastures for yearling
steers. Agron. J. 50:85-88.
Van Keuren, R. W., and G. C. Marten. 1972. Pasture production and
utilization. j[n C. H. Hanson (ed.) Alfalfa science and technology.
Agronomy 15:641-658. Am. Soc. Agron., Madison.
Van Soest, P. J. 1965. Symposium on factors influencing the voluntary
intake of herbage by ruminants: Voluntary intake in relation to
chemical composition and digestibility. J. Anim. Sci. 24:834-843.
131
Vance, C. P., G. H, Heichel, D. K. Barnes, J. W. Bryan, and L. E.
Johnson. 1979. Nitrogen fixation, nodule development, and vege­
tative regrowth of alfalfa (Medicago sativa L.) following harvest.
Plant Physiol. 64:1-8.
Vartha, E. W. 1973. Performance of lucerne-grass pastures on Wakanui
silt loam. N. Z. J. Exp. Agric. 1:29-34.
Walker, T. W. 1956. Nitrogen and herbage production.
Grassl. Cong. 7:157-165.
Proc. VII Int.
Walker, T. W., A. F. R. Adams, and H. D. Orchiston. 1956. Fate of
labelled nitrate and ammonium nitrogen when applied to grass and
clover grown separately and together. Soil Sci. 81:339-351.
Waring, S. A., and J. M. Bremner. 1964. Ammonium production in soil
under waterlogged conditions as an index of nitrogen availability.
Nature 201:951-952.
Watkin, B. R., and R. J. Clements. 1978. The effects of grazing animals
on pastures, p. 273-289, Iji J. R. Wilson (ed.) Plant relations
in pastures. CSIRO, Melbourne.
Wedin, W. F. 1974. Fertilization of cool-season grasses, p. 95-119.
jhi D. Mays (ed.) Forage fertilization. Am. Soc. Agron., Madison,
Wis.
Wedin, W. F., and R. L. Vetter. 1970. Pasture for beef production in
the western Corn Belt—U.S.A. Proc. XI Int. Grassl. Cong. 11:
842-845.
Wedin, W. F., J. D. Donker, and G. C. Marten. 1965. An evaluation of
nitrogen fertilizer in legume-grass and all-grass pasture. Agron.
J. 57:185-188.
Wedin, W. F., I. T. Carlson, and R. L. Vetter. 1966. Studies on nutri­
tive value of fall-saved forage, using rumen fermentation and chem­
ical analyses, Proc. X Int. Grassl. Cong, 10:424-428.
Wedin, W. F., R. L. Vetter, J. M. Scholl, and W. R. Woods. 1967. An
evaluation of birdsfoot trefoil (Lotus corniculatus L.) in pasture
improvement. Agron. J. 59:525.
Wedin, W. F,, R. L. Vetter, and I, T. Carlson. 1970. The potential
of tall grasses as autumn saved forages under heavy nitrogen fer­
tilization and intensive grazing management. Proc. N. Z. Grassl.
Assoc. 32:160-167.
132
Wedin, W. F., H. J. Hodgson, and N. L. Jacobson. 1975. Utilizing plant
and animal resources in producing human food. J. Anim. Sci. 41;
667-686.
Wedin, W. F., C. P. West, J. R. Russell, and R. L. Vetter. 1979.
Progress Report: Efficiency of grazing study—McNay Memorial Re­
search Center. Iowa State Univ. Agric. Home Econ. Exp. Stn. A. S.
Leaflet R295.
Wedin, W. F., H. J. Hodgson, J. E. Oldfield, and K. J. Frey. 1980. Feed
production, p. 153-191. In^W. G. Pond, R. A. Merkel, L. D.
McGilliard, and V. J. Rhodes (eds.) Animal agriculture: Research
to meet human needs in the 21st century. Westview Press, Boulder,
Co.
West, C. P., N. P. Martin, and 6. C. Marten. 1980. Nitrogen and
Rhizobium effects on establishment of legumes via strip tillage.
Agron. J. 72:620-624.
Whitehead, D. C. 1970. The role of nitrogen in grassland productivity.
Commonwealth Agric. Bureau Bull. 48. Hurley, Berkshire, U.K.
Whitney, A. S., and Y. Kanehiro. 1967. Pathways of nitrogen transfer
in some tropical legume-grass associations. Agron. J. 59:585-588.
Wilson, J. K. 1942. The loss of nodules from legume roots and its sig­
nificance. J. Am. Soc. Agron. 34:460-471.
Wilson, R, F., and J. M. A. Tilley. 1965. Amino acid composition of
lucerne and of lucerne and grass protein preparations. J. Sci.
Food Agric. 16:173-178.
Woldendorp, J. W., K. Dilz, and G. J. Kolenbrander. 1966. The fate of
fertilizer nitrogen on permanent grassland soils, p. 53-68. In^
P. F. J. van Burg and G. H. Arnold (eds.) Nitrogen and grassland.
Proc. First Gen. Meeting Euro. Grassl. Fed., Wageningen, Neth.
Woodhouse, W. W., Jr., and D. S. Chamblee. 1958. Nitrogen in forage
production. North Carolina Agric. Exp. Stn. Bull. 383.
Hyss, 0., and P. W. Wilson. 1941. Factors influencing excretion of
nitrogen by legumes. Soil Sci. 52:15-23.
Young, D. 0. B. 1958. A study of the influence of nitrogen on the root
weight and modulation of white clover in a mixed sward. J. Br.
Grassl. Soc. 13:106-114.
Yungen, J. A., T. L. Jackson, and W. S. McGuire. 1977. Seasonal
responses of perennial forage grasses to nitrogen applications.
Oregon Agric. Exp. Stn. Bull. 626.
133
ACKNOWLEDGMENTS
My deep gratitude is expressed to Dr. Walter F. Wedin for his sound
guidance and support during the course of my doctoral program.
Appreciation is extended to the members of my doctoral committee
for their suggestions and critical review of this manuscript.
Special gratitude is accorded to Patricia Patrick for her skillful
and enthusiastic assistance, and to all the other members of the forage
crew for their help.
My thanks are extended to the Department of Agronomy and to the Iowa
State University Agriculture and Home Economics Experiment Station for
their financial support during my program.
I wish to express my deepest appreciation to tuy wife, Jane, for her
patience, understanding, and encouragement during the progress of n\y
doctoral program.
134
APPENDIX
330'-
HEP I
BFT-TF
médium
P-2
P-3
AllF-SB-OG
RC
heavy
heavy
BFT-TF
heavy
Vs"
P-6
AllF-SB-dG
medium
RC
medium
426'
<>
"31
-3-t—2-f^lP-8
REP II
AL^-SB-OG'
medium!
P-9
BFT-TF
RC
heavy
heavy
13=-ClP-10
P-ll
P-12
RC
medium'
ÎBFT-TF^
AL F-SB-OG
medium
iheavy I
W
Ol
Each pasture is divided with electric fence
(
) into 3 one-acre paddocks. Vegetative
sampling was done in paddock 2.
P = Pasture followed by number
BFT-TF = Birdsfoot trefoil-Tall fescue
ALF-SB-OG = Alfafla-smooth brotnegrass-orchard-
Reserve pasture
N
grass
Pump
house
Scales &
corral
= Reed canarygrass
heavy = grazing pressure
,,
I= approximate locations of plots used for 'N-tracing.
access roaTT
^ Locations of waterers
Figure A.l.
Field plan of experimental pastures at McNay Memorial Research Center,
Lucas Co., Iowa
136
Figure A.2.
Modified syringe needle for injecting ^^N-labelled
ammonium sulfate solution into alfalfa-grass microplots
137
Figure A.3.
Injection of ^^N-labelled ammonium sulfate solution
into soil of alfalfa-grass microplots
138
Figure A.4.
Caged microplot in alfalfa-grass pasture
139
Table A.l.
Precipitation and temperature data for McNay Memorial
Research Center in 1979 and 1980
Precipitation^
riUM Lll
1979
1980
—- — — —
Temperature^
Normal^
————————
1979
1980
————————
- - - - - -
Jan.
4.5
4.5
2.9
-13.0
Feb.
1.7
1.9
2.7
March
10.4
4.6
Apri 1
5.2
3.8
May
9.4
2.8
June
14.5
July
Normal^
-5.4
-9.6
-4.1
-5.3
6.0
3.1
1.8
2.7
9.0 .
8.6
10.7
10.7
10.2
16.4
16.6
16.4
26.7
13.8
21.2
21.3
21.3
6.4
7.9
9.7
23.2
26.4
23.7
Aug.
5.3
17.2
9.7
23.6
25.3
22.8
Sept.
2.5
7.5
10.4
18.8
19.5
18.1
Oct.
10.2
4.7
6.2
12.7
10.1
12.8
Nov.
3.1
3.9
3.5
4.8
7.6
Dec.
0.8
1.2
5.9
3.3
0.3
-1.9
-2.6
-2.7
Precipitation for Jan., Feb., March, Nov. and Dec. taken from
Chariton, lA, weather station. April to Oct. data recorded at the
pasture site.
^All temperature data are from Chariton, lA, weather station.
^Eighty-seven-year average at Chariton, lA, weather station.
"^Eighty-six-year average at Chariton, lA, weather station.
140
Table A.2.
Correlations (r) between selected plant measurements in
1979 and 1980; n=16
yield
yield
N
'T helge
fixed uptake
legume
-1979-
Herbage
.85**
Dm yield
.36
.80**
.68**
-.73**
.85**
.83**
.62**
-.16
.83**
.94**
-.75**
.99**
.46+
.90**
-.35
.15
.94** -.15
-.78**
.93**
.74**
.73**
Legume
N yield
.99**
Grass N
yield
.13
.93**
.42
Herbage
N yield
.92**
.19
.88**
-.10
.79**
1980
Herbage
Dm yield
.89**
.67**
.74**
.67**
-.85**
.84**
.88**
.52*
Legume
N yield
.99**
.27
.73**
.91**
-.63**
.99**
.60*
.82**
Grass N
yield
.49*
.94**
.66**
.16
-.92**
.39
.99**
Herbage
N yield
.93**
.58*
.81**
.74**
-.83**
.89**
.84**
-.03
.59*
+,*,**Significant at .10, .05 and .01 levels, respectively.
100
80
N yield
kg/ha
40
20
0
May
June
July
Aug.
Sept.
Oct.
Nov.
Harvest date
Figure A.5. Harvested-yield of N in alflafa-grass'^^N microplots by year.and harvest ;
date; means of 16 plots ± S.E.
142
Table A.3.
Percentage N in reed canarygrass plots at four rates of N
fertilizer by harvest date; means of four replications
Harvest
N rate
1
2
-kg/ha-
-
%
3
4
-
0
2.41
3.00
3.28
3.93
60
2.96
3.19
3.88
3.90
120
3.32
3.13
3.94
4.21
180
3.69
3.38
4.00
4.55
**
+
**
**
+,** Significant at .10 and .01 levels, respectively.
143
Table A.4.
Grazing
pasture
Percentage organic carbon, percentage N and C/N ratio
of pasture soils at three depths; samples taken in the
second of three paddocks on 24 April 1979
q_3
Alfalfa-grass
3.15 cm 15-30 cm
-
Medium
Heavy
X
2.90
3.18
3.04
---% C
2.03
2.19
2.11
-
Medium
Heavy
X
.278
.300
.289
Reed canar.ygrass
0-3 cm
3-15 cm 15-30 cm
.192
.203
.198
1.56
1.78
1.67
3.15
3.68
3.42
2.00
2.26
2.13
1.64
1.82
1.73
.312
.321
.317
.198
.210
.204
.166
.180
.174
10.0
11.4
10.7
10.1
10.7
10.4
9.9
10.1
10.0
-—% N
.156
.164
.160
C/N
Medium
Heavy
X
10.4
10.6
10.5
10.6
10.8
10.7
9.9
10.8
10.4
- —1979
5000
1980
4000
DM yield
kg/ha
3000
-
•
1--
2000
1000
0
-
^
•
May
1
June
, .. _ 1
July
...
. , .1.
.
Ar].
,
Sept.
'
1
1
Oct.
Nov.
Harvest date
Figure A.6.
HarvestedjDM yield of reed canarygrass receiving 180 kg/ha-yr of N by year
and harvest date; means of 12 plots ± S.E.
100
- - --1979
_
1980
K
80
T
-
/^
N yield
\ \
-
kg/ha
40
Vs.
20
\
\
n
•
May
•
June
1
July
1
1
Aug.
Sept.
1
Oct.
I L
Nov.
Harvest date
Figure A.7._ Harvested yield of N in reed canarygrass receiving 180 kg/ha^yr of N by
year and harvest date; means of 12 plots ±S.E.