THE FATE OF FERTILIZER-N APPLIED TO
A FLORIDA CITRUS SOIL
Daniel Apollo Hagillih
A DISSERTATION PRESENTED TO THE
GRADUATE COUNCIL OF THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
The author wishes to express his gratitude and appre-
ciation to Dr. R.C.J. Koo, the chairman of his supervisory
committee, for his patient guidance in the execution of
this research and the preparation of the manuscript.
Throughout the graduate training of the author, Dr. Koo
strived to imprint on the author an appreciation of the
scientific method in research.
The author is equally indebted to Dr. A.H. Krezdorn,
Professor Emeritus of Fruit Crops, who had been the Chair-
man of his supervisory committee. Even in retirement, Dr.
Krezdorn continued to serve on this committee and to will-
ingly give of his time, counsel and support. To Dr. W.S.
Castle, a member of his committee, the author owes a debt
of gratitude and appreciation for his valuable contribu-
tion in suggestions, counsel and involvement in all phases
of this study.
Sincere appreciation is extended to Dr. N. Gammon,
Jr., Professor Emeritus of Soil Chemistry, for his construc-
tive criticism and counsel particularly in the soil aspect
of the research. The author also wishes to thank Dr. D.A.
Graetz for his assistance in the analysis of soil samples.
The graduate training of the author was sponsored by
the Agricultural Research Corporation of the Sudan. Towards
the end of his training, the Department of Fruit Crops
offered the author a graduate assistantship. He grate-
fully acknowledges the financial support from both.
Finally, the author wishes to thank Ms. Janice L.
Cambridge for her companionship and encouragement through-
out the duration of this study.
TABLE OF CONTENTS
ACKNOWLEDGEMENTS . .
ABSTRACT . . .
INTRODUCTION . . .
LITERATURE REVIEW . .
Nitrogen Balance and Its Major Components
Considerations for Increasing the
Efficiency of N Usage by Citrus Trees
Analytical Procedures for N Balance
Components. . ... 10
Isotope Approach to N Utilization
by Citrus Trees . . 15
MATERIALS AND METHODS. . ... 18
Experimental Site . .. 18
Experimental Design . 18
Field Sampling Procedures . .. .24
Sample Preparation and Analysis . 27
Statistical Analysis. . 29
RESULTS AND DISCUSSION . 30
Experiments 1 and 2 ..... .. 30
Experiment 3. . .. 70
CONCLUSIONS. .... ... 87
LITERATURE CITED ..... . 91
BIOGRAPHICAL SKETCH. . .... .. 103
Abstract of Dissertation Presented to the
Graduate Council of the University of Florida in
Partial Fulfillment of the Requirements
for the Degree of Doctor of Philosophy
THE FATE OF FERTILIZER-N APPLIED TO
A FLORIDA CITRUS SOIL
Daniel Apollo Hagillih
Chairman: Dr. R.C.J. Koo
Major Department: Horticultural Science
The fate of fertilizer-N applied to Astatula fine sand
was studied by the destructive sampling of 1-year-old 'Pine-
apple' orange (Citrus sinensis [L.] Osbeck) trees on 'Alemow'
(C. macrophylla Wester), and through soil chemical analysis,
using the difference method. The fall-winter study involved
unfertilized control, isobutylidene diurea (IBDU) and NH4NO3
sources in a factorial combination with no irrigation, low
and high irrigation levels. Nitrogen source and irrigation
treatments lasted for 14 and 6 weeks, respectively. The
spring study involved only the 3 N sources and lasted for
6 weeks. Leaf and fruit sampling of bearing 'Pineapple'
orange trees, fertilized with the same N sources, and soil
chemical analysis of the unreplicated tree plots were also
used. Soil water content in all experiments was determined
by a neutron probe.
The results indicated that growth parameters of the
young trees, including percent trunk diameter increase, dry
weights of component parts and the entire tree, showed no
statistical differences attributed to N source, apparently
because of the short duration of the studies. There was,
however, a trend for IBDU and NH4NO3 fertilization to re-
sult in greater dry matter, particularly in the more succu-
lent aerial parts. This was apparent in the leaves of the
fertilized trees which were statistically greater than those
of unfertilized trees.
There was a highly significant relationship between
N source and N concentration and total N of tree parts,
and total N of the entire tree. In every case, the order
of N absorption was NH4NO3- > IBDU-fertilized > control
trees. This confirms that a soluble source such as NH4NO3
is more readily absorbed in the short-term than a controlled
release form of N. It is suggested that this trend may not
prevail in the long-term.
Irrigation level did not statistically affect tree
growth and N content parameters, and soil N content in the
fall-winter study. The narrow range and short duration of
the irrigation treatments were probably not sufficient to
elicit treatment responses.
Nitrogen balance data showed that 66 and 30% of ap-
plied N from IBDU and NH4NO3, respectively, were retained
in the 0- to 60-cm soil profile. Twenty-five and 45% from
the same respective sources were accounted for in the young
trees. Total apparent N recovery in the soil-plant sys-
tem was 91 and 76% for IBDU- and NH4NO3-N, respectively.
The deficit was assumed to have been leached and/or lost to
There was no difference in fruit yield between IBDU-
and NH4NO3-fertilized bearing trees. Fertilized trees,
however, yielded 4 times as much as the control. Fruit
N concentration and removal and leaf N concentration were
in the order of NH4NO3- > IBDU-fertilized > unfertilized
tree. The absence of replication precluded the establish-
ment of a plausible N balance. Nevertheless, 40% of the
applied N was estimated to be contained in the fruits and
High N recoveries on experimental plots involving young,
nonbearing, and bearing trees suggest that substantial leach-
ing of N may not be a major route for N loss from citrus
groves established on deep, well-drained soils when reason-
able rates of fertilizer-N are applied over the root zone.
The potential for a single application of IBDU to
young citrus trees, which are normally fertilized several
times a year, needs further study.
Florida's sandy soils planted to citrus are character-
ized by low native fertility which necessitates the regu-
lar addition of fertilizer nutrients to maintain a com-
mercial level of production. Such an edaphic environment
is also conducive to the substantial leaching of mineral N
by rainfall and irrigation (114). Soil tests which measure
N carry-over from previous N additions are of little value
in these soils. As a result, the replacement of nutrients,
including N, lost from the soil is generally dependent up-
on an evaluation of the need of the crop.
The N requirement of commercial citrus cultivars is
based on many long-term investigations in which tree growth
and yield were related to fertilizer rate, source, time of
application and leaf N content. Nitrogen uptake by the
citrus tree is considered to be a small part of the total
N loss from the deep, sandy soils of central Florida (100).
The efficiency of tree N utilization, i.e., the propor-
tion of N used by the tree compared to that applied, has
been estimated to be 25-30%. As a result, there is public
and grower concern regarding the quantitative extent of the
leaching loss of N applied to citrus. Leached N represents
an irretrievable loss of a costly resource and may
contribute to the NO3 pollution of underground water and
to lake eutrophication.
No meaningful N balance studies, i.e., accounting
quantitatively for the fate of N applied to citrus soils
as a fertilizer, have been done in Florida as they have in
California (29, 68, 69). In order to enhance our under-
standing of N use in Florida soils planted to citrus, it
is essential that the fate of fertilizer-N be studied.
The primary objective of this research was to make a
quantitative accounting of fertilizer-N applied to bearing
and young trees on a deep, well-drained sandy soil of the
Florida ridge area, where much of the citrus in Florida is
grown. A second objective was to evaluate isobutylidene
diurea (IBDU) as a controlled-release N fertilizer source
Nitrogen Balance and Its Major Components
Important sources for the N in the soil solution in-
clude fertilizer-N additions and net mineralization. Other
sources also likely to add some N to the soil solution in-
clude rainfall, irrigation water, biological fixation, dif-
fusion of atmospheric ammonia into the soil surface, direct
absorption of atmospheric ammonia by plant leaves, and
transformations involved in the N cycle (80, 125). Major
sources of N removal from the soil solution are plant up-
take, leaching, net fixation in the soil and denitrifica-
tion. Smaller losses of N in most agricultural soils in-
clude gaseous losses through ammonia volatilization and
physical removal by wind and animals.
Attempts to account for the fate of applied N in the
soil-plant-atmosphere continuum have yielded some useful
information, but it is rare that a complete quantitative
recovery of applied N is obtained (2, 50, 97). Calculation
of a N balance for a field system is particularly diffi-
cult because of the many possible fates of N in soils (65),
and because of the physical difficulties encountered in
obtaining quantitative data for some of the major compo-
nents of a N balance such as leaching, plant uptake and
residual N (8, 97).
Leaching of N
Leaching has been defined as the movement of solutes
from one soil zone to another by percolated water (107).
Leaching of N is important because it represents the loss
of a costly resource and because it may contribute to en-
Few measurements have been reported giving accurate
data concerning nutrient loss from agricultural lands by
leaching (34). Most of the mineral N lost by leaching in
citrus orchards is in the NO3 form (42, 51). The extent
of NO3 loss from the soil-plant system depends on the fer-
tilizer material, soil type, plant species, rainfall and
management practices. High infiltration rates and low
water storage capacities of sandy soils make them espe-
cially subject to NO3 leaching. Other factors which in-
fluence the magnitude of NO3 leaching include evaporation,
nutrient concentration, depth of rooting, N reserves, min-
eralization of the soil, nitrification of the NH4-N and
soil temperature (113).
In California, the proportion of N applied to citrus
leached annually as NO3 ranged from 45 (8) to 90% (51).
These high NO3 leaching losses, however, occurred under
conditions of excessive irrigation on well-drained soils
atypical of many irrigated areas in arid zones. In a deep
sandy Florida soil, split application of 200 kg/ha of N
annually to a mature citrus grove did not result in NO0
leaching even when the rainfall was high (35). Studies in-
volving a single annual application of N to evaluate the
magnitude of NO3 leaching loss from citrus plantings in
Florida have not been done.
Plant Cycling of N
Information on the N content of an entire citrus tree
is meager. The size of the mature tree makes the destruc-
tive biomass sampling a difficult and costly task (111).
In forest tree studies, where there is a commercial inter-
est in the wood, biomass sampling procedures for the tree
are more highly developed (9, 66, 91, 126). Also, with
citrus, leaf N content is considered an adequate index of
the N status of the tree. The commercial adoption of leaf
analysis as a guide to fertilization of citrus has report-
edly reduced N usage by about 50% in California (31).
A few studies on total citrus tree N content have been
reported, but very often these studies were lacking in ade-
quate replication (5, 63, 100, 121). In Florida, it was
estimated that a 15-year-old orange tree absorbed only 25-
30% of applied N (100). Of this amount, 33% is returned
to the soil during the year through shedding of old leaves,
flowers and young fruits, 50% is removed in the crop of
fruit, and about 15% goes into permanent structures in the
form of new twigs and trunk enlargement. Data from a
single 19-year-old grapefruit (Citrus paradisi Macf.) tree
in Florida (5) indicated that the N content of component
tree parts on a % dry weight basis in descending order was
leaves, immature fruit, twigs less than 1.3 cm diameter,
fibrous roots, large roots, limbs up to 1.3 cm diameter
A series of studies to determine the N content of
'Valencia' orange trees have been reported from California
(19,20, 121). One study showed that a 10-year-old tree
contained 200 g of N in the leaves (19), while another
study indicated that nearly one-half of the N content of
the tree was in the leaves (20). Wallace et al. (121)
reported that a 14-year-old 'Valencia' orange tree contained
673 g of N. The authors did not, however, indicate how
much N was applied so that it is difficult to relate this
figure to trees of similar age elsewhere. Apparently, a
significant amount of N added to the soil is removed by
citrus fruits (59, 131). Zidan and Wallace (131) reported
that a mature 'Washington' navel orange fruit with a fresh
weight of 192 g contained 394 mg of N while a 'Valencia'
orange fruit of 165 g contained 327 mg.
Residual Soil N
A part of the N balance which is the most difficult to
account for because of spatial variability is the amount of
N remaining in the soil at any given sampling time.
Sampling and analysis of the soil material for NO3N (1, 3,
51, 61, 81, 82) and/or total N (36, 84, 102) is commonly
used to measure soil N content but large samples are usu-
ally required to achieve an acceptable degree of accuracy.
The variation in residual soil N content, both within and
among experimental units, has been reported to decrease
with increasing depth in the soil profile (79). Also, the
variability of virgin land has been reported to be less for
N than similar areas of cultivated land (7).
In a Florida study involving an orchard of mature
citrus trees, the soil nutrient content was the least vari-
able in samples collected below the tree dripline as com-
pared to other sampling locations (11). Nevertheless, it
was demonstrated that a large number of samples was re-
quired to reduce the variation regardless of the sampling
Considerations for Increasing the
Efficiency of N Usage by Citrus trees
Optimizing N use by citrus trees appears to involve
the development of management programs which prevent the
excessive loss of NO3 by leaching without sacrificing fruit
yield, size, and quality (29). It has been suggested that
optimum management could be achieved by applying only the
amount of N needed to sustain high fruit production, with
leaf analysis being used as a guide to determine the proper
rate (32, 90). Differences in citrus cultivar N require-
ments and soil types, however, dictate an approach in which
timing, rate, placement, N source and water regime must
be considered in the efficient use of fertilizer-N.
One approach to increasing the efficiency of N use
has been to split the annual application of N (17, 35, 88).
This is supported by a Florida experiment conducted under
high rainfall conditions (35). In terms of citrus fruit
productivity, however, other studies (17, 88) indicated
that a single application of fertilizer-N applied during
the drier portion of the year, in winter, gave the same
yield response compared to 2 applications per year.
Another approach is the foliar application of N in
the form of low-biuret urea. Studies from California (29,
30) indicated that foliar application of N was associated
with lower NO3 leaching potential than was soil applica-
tion. In general, foliar-applied N, under the low rainfall
conditions of California, was found to be as effective as
soil-applied N for fruit production (29, 30). This approach
may not be suitable in Florida where frequent rains may
wash away the N applied to the foliage. Efficient N use
may require the application of nominal rates to the soil,
supplementing this with foliar-applied N.
It has been suggested that in future studies, split
soil application of N be a treatment along with foliar-soil
combinations (54). A major limitation in foliar application
of N, however, is that it is energy intensive especially
if frequent applications are necessary.
Concept of Controlled Release N Fertilizers and IBDU
The concept of the controlled release of N to increase
fertilizer-N efficiency has been amply reviewed (62, 74,
76, 77, 85, 106). The basic approaches are the development
of compounds of limited solubility, use of coated granules,
and formulation of ammoniacal fertilizers with nitrifica-
tion inhibitors. The objectives of such fertilizer-N forms
are to minimize the immobilization, leaching and volatili-
zation losses of soil-applied N, and to release an adequate
amount of available N to satisfy the crop requirement. How-
ever, because of higher costs, these fertilizers are being
used primarily in turfgrass, ornamentals, home grounds,
nursery fields, and for other specialty situations.
Isobutylidene diurea is a condensation product of urea
and iso-butyraldehyde patented in Japan as a slow-release
N source. Hamamoto (41) described its preparation and
physical properties in greater detail. His review of the
studies conducted in Japan showed that the large granule of
IBDU was effective in preventing high N leaching loss when
applied to paddy soils. By using IBDU with green tea
(Camellia sinensis [L.] Kuntze) plants Hamamoto (41) re-
ported that one application was enough to get average yields
and thus save labor. The conversion of IBDU to plant avail-
able form appears to result from the dissolution of IBDU
granules. The rate of dissolution is strongly influenced
by particle size (41). Fine sized IBDU granules release
larger amounts of N than coarse granules. This was ob-
served in an incubation study (48), for turfgrasses (116,
117), and for flue-cured tobacco (Nicotiana tabacum L.)
(71). Hughes (48) confirmed Hamamoto's (41) findings that
IBDU-N release was greater at low pH values and high tem-
peratures. In evaluating IBDU for 'Merion' Kentucky blue
grass (Poa pratensis L.), Moberg et al. (72) indicated that
a uniform growth response was possible from 2 applications
of IBDU as compared to 3 of UREX (a urea-paraffin product).
Studies in citrus or other tree crops with IBDU have not
as yet been reported.
Analytical Procedures for
N Balance Components
Leached N Sampling and Analysis
Soil N losses from leaching have generally been stud-
ied in lysimeter experiments or in controlled field drain-
age plots by measuring the N content of the effluent water
(43). In citrus, lysimeter studies have largely been lim-
ited to young trees in the greenhouse. Such studies can
provide useful basic information, but they are limited in
scope because of the difficulty of extrapolating the data
to orchard conditions (39). Moreover, a major shortcoming
of lysimeters is the retention of more moisture than
corresponds with field capacity, resulting in the danger of
anaerobiosis and denitrification (23).
In mature citrus orchards, several approaches have
been made for estimating the leaching of fertilizer-N.
Wander (123) analyzed water samples from the lower edges
of citrus plantings located on slopes in Florida. Be-
cause the actual amount of water used by mature citrus
trees and the exact rainfall in the local area was not
known, however, it was difficult to translate such water
analyses into figures representing actual amounts of N lost
Various systems of drain lines and tiles have been
devised in which data obtained by effluent water analysis
and water discharge rates are used to calculate the amount
of N leached. The amount of NO3-N in the drainage from
citrus groves has been determined by this procedure in
California (26, 60) and Florida (16, 18). However, the
reliability of this approach for monitoring subsurface
NO3-N losses has been questioned (108).
Several investigators have used porous ceramic cups to
collect soil solution samples from citrus groves (10, 26,
35, 60). This procedure was found to be useful for esti-
mating the potential for NO pollution in a Florida citrus
grove (35). The procedure also gave a reasonable estimate
of the amount of NO3-N leached from a millet (Pennisetum
typhoides L.) plot (37). However, this procedure has been
criticized because of the difficulty of providing enough
water samples for chemical analysis and the requirement
for a suction device. A new vacuum extractor was developed
to surmount these difficulties but its usefulness has been
limited by operational problems (27).
A new approach to calculating the leaching of NO3 from
the root zone of crops is based on a comparison of the be-
havior of NO and Cl ions in the drainage or percolating
waters that have moved below the root zone (82, 83).
Changes in the ratio of these ions were used to calculate
the amount of NO3 leached. In unsaturated soils where the
determination of NO3 and Cl concentrations in the percola-
ting water is difficult these concentrations were deter-
mined in a saturation extract. In both cases NO3 and Cl
ratios were found to be satisfactory in estimating NO3
leached below the root zone of crops.
Residual Soil N Sampling and Analysis
Soil samples from citrus orchards are obtained using
various sampling techniques. For cores greater than 6 m
in length, a power-driven 45-cm diameter auger is often
used (82). When samples are taken to only 6 m, a truck-
mounted hydraulically powered sampler is frequently pre-
ferred. Most sampling in citrus orchards is conducted to
shallower depths, generally not exceeding 4 or 5 m. In
such cases, conventional sampling soil tubes or augers are
Several analytical procedures have been adopted to
determine residual soil N. For total N determination, 2
methods have gained general acceptance: the Kjeldahl
method which is essentially a wet-oxidation procedure, and
the Dumas method which is fundamentally a dry-oxidation pro-
cedure (12). Residual soil N is also determined by analyz-
ing the soil for exchangeable NH4 and NO3 ions using steam
distillation procedures (13). Recent trends in quantita-
ting residual soil N, however, have involved the determina-
tion of NO3 (67) and NH (4) ions in saturation extracts
using specific ion activity electrodes.
Soil Water Measurement
Nitrogen is absorbed by plant roots from the soil solu-
tion. The soil water content is important, therefore, not
only because of its function as a solvent but also because
its mass movement is directly related to the leaching of the
Soil water content is measured by several techniques,
each with advantages and disadvantages. One technique used
extensively in loam- to clay-textured soils is based upon
the tensiometer (92, 93, 95, 105). The information fur-
nished by tensiometers tends to be qualitative. Tensio-
meters are generally used for determining timing and dura-
tion of irrigation and have not been found to be particu-
larly useful for sandy soils (52, 56).
Electrical-resistance blocks are not extensively used
on sandy soils (64, 105) because gypsum blocks operate most
reliably in the drier portion of the soil water content
spectrum. In the sandy Florida soils, the available water
is relatively small, and soil water should be maintained
above 65% of field capacity. Under these conditions gyp-
sum blocks are not particularly suitable.
For direct quantitative measurement of soil water by
volume, 2 methods are used. The gravimetric procedure in-
volves the calculation of a volumetric water content from
the weight of the water in a soil sample and the soil bulk
density (46, 64, 94). Rapid, large-scale soil water con-
tent sampling by this technique is laborious and time-
The neutron scattering method is a faster and equally
accurate method as compared to the gravimetric procedure
(14, 15, 40, 104). The method has been found to be useful
in studying patterns of soil water depletion in Florida
citrus groves (53).
Plant Tissue Sampling and Analysis
The leaf is the most commonly sampled tissue to deter-
mine the N status of a citrus tree (31, 32, 99, 101). As
a result, techniques in leaf sampling and analytical pro-
cedures have been highly developed. The entire mature tree,
however, is rarely sampled to determine its total N content.
One approach for determining the N content of the tree is
to uproot the entire tree. The tree is then fractionated
into various component parts and the N content of sample
parts determined after oven-drying and grinding. This
approach has been tried in Florida (5) and in California
(19, 20). Another approach has involved making a. periodic
weight and chemical determination of all the blossoms,
leaves and fruits that fall from citrus trees. Wallace
et al. (121) used this approach to obtain data which were
then combined with similar information for harvested fruits,
new leaves and twigs, to obtain a measure of the nutrient
content of the tree.
Isotope Approach to N
Utilization by Citrus Trees
The use of isotopes in tree nutrition research was
discussed by Walker (118) and Hauck (44). The advantages
of this technique include the direct measurement of trans-
port velocities and separation of newly absorbed nutrients
from those already present in the various components of the
Nitrogen occurs in nature in 2 stable isotopes, N14 and
N5. Only recently has it been possible, as the result of
advances in cryogenic techniques whereby nitric oxide gas
is liquefied and then distilled, to separate the 2 isotopes
with relative ease (21). Nitrogen15 fertilizer formula-
tions are those which have been enriched with excess N15
while Nl -depleted fertilizers, also known as N14 ferti-
lizers, are those from which much N5 has been removed.
Studies using N15 as a tracer are commonplace in agro-
nomic crops because the plants are relatively shallow-
rooted (6, 49, 50, 73, 112, 127, 128, 130). Deeper-rooted
forest and fruit trees do not lend themselves to N balance
studies using N15 (66, 75). The high cost of the tracer
has largely restricted its use to laboratory and greenhouse
studies with young trees and strongly affected the size of
the experiment and the experimental design. Furthermore,
the procedure requires the use of expensive instrumentation
to which many laboratories may not have access. Neverthe-
less, such studies have been reported for young apple
(Malus sylvestris Mill.) trees (24, 38, 45) and nonbearing
prune (Prunus domestic L.) trees (124).
Greenhouse studies with citrus using N15 were reported
by Wallace (119, 120) and Wallace et al. (122). Wallace
(119) found that growth complications and variability in
plant material resulted in conventional methods being in-
ferior to the isotopic technique in studying the influence
of temperature on N absorption and translocation. His qual-
itative investigation (120) showed that NO3-N was 2 to 5
times as readily absorbed from soil as was NH4-N by 8-week-
old cuttings of 'Eureka' lemon (C. limon L.). Nitrogen15
appeared in the leaves of a 3-year-old tree 4-7 days after
application to the tree (122).
After this initial interest, no studies in citrus
with N15 have been reported until recently. Kubota and
coworkers (57, 58) investigated the behavior of N supplied
in early spring and early summer on 9-year-old 'Satsuma'
mandarin (C. unshiu Marcovitch) trees. In the spring ex-
periment, N5 was detected in rootlets and leaves 3 and 7
days, respectively, after the application. Seventy-five per-
cent of absorbed N15 was found in the aerial parts of the
tree. In the summer study, N15 was detected in rootlets
and spring leaves the next day after application, indica-
ting a faster absorption of N in summer. Ninety-two percent
of the N15 was distributed in the above ground parts. A
review of N5 studies on citrus in Japan was recently
presented by Yuda (129).
Both N-- depleted and N- enriched fertilizer materials
have been reported to give accurate and precise measurement
of tracer N in plant tissue (28, 103). No studies with
N- depleted fertilizers, however, appear to have been re-
ported with citrus.
MATERIALS AND METHODS
Experiments for this study were conducted on an Asta-
tula fine sand site 24 km from Lake Alfred, Florida. The
soil is a Typic Quartzipsamment of the Entisols (47).
Typically, it is well-drained with a low water and miner-
al nutrient retaining capacity.
Two factors, N source and irrigation, were examined
in this experiment. Levels of the N source factor in-
cluded no N control, IBDU-N, and NH4NO3-N, while those of
the irrigation factor were no irrigation, low and high
irrigations. It was desired to study these treatments in
a factorial arrangement; however, it was not possible to
physically establish the experiment in this manner because
of the limitation imposed by the irrigation treatment.
Therefore, the 1-year-old 'Pineapple' orange (Citrus sin-
ensis [L.] Osbeck) trees on 'Alemow' (C. macrophylla Wester)
rootstock were spaced at 3 x 3 m in 2 rows on July 13, 1978,
using a modified split plot design (Fig. 1). The field
layout consisted of 3 replications with irrigation level
as the main plot treatment. The size of each main plot
was determined by 6-m lengths of perforated sprinkler pipe
used to apply the irrigation treatments. Each main plot,
therefore, had 4 trees, 2 on each side of the pipe. Each
single tree plot received one level of the N treatment but
4 replications of each N treatment were randomly dis-
tributed over the 3 irrigation main plots. Thirty-six trees
were used in the experiment.
Fertilizer-N from IBDU and NH4NO3 sources at the rate
of 201.6 g/tree was hand broadcast in a 3 x 3 area around
each tree on September 28, 1978, 11 weeks after the trees
were planted. Trees in control plots were not fertilized.
Irrigation treatment was not started until November 27, 1978,
after rainfall had tapered off. Soil water content was
used to schedule irrigation. Approximately 0.75 cm of ir-
rigation water was applied whenever the average soil water
content fell below 4% by volume in the 0- to 60-cm soil
profile of the high irrigation plots. The high irrigation
plots were irrigated twice as often as the low irrigation
plots which also received 0.75 cm of irrigation water on
each application. A rain gauge at the experimental site
and an irrigation flow-meter were used to monitor water
supply. When the flow-meter malfunctioned 1500-ml cans
were placed in the plots for collection and measurement of
the amount of irrigation water delivered.
The experiment was terminated on January 9, 1979, be-
cause of freeze damage. Fertilizer-N and the first
irrigation had been applied 14 and 6 weeks, respectively,
prior to the termination of the experiment. The trees un-
der low irrigation had received a total of 2.20 cm of ir-
rigation water in 3 irrigations while those under high ir-
rigation had received 4.95 cm in 6 irrigations. For the
entire period of the experiment, the nonirrigated plots
had received 20.17 cm of rainfall; the low and high irriga-
tion plots had received a combined total of 22.37 and 25.12
cm of rainfall and irrigation, respectively.
This experiment was initiated to clarify certain re-
sults obtained in Experiment 1. The total N content of
NH4NO3-fertilized trees was 42% higher than that of the
IBDU-fertilized trees. Moveover, the complicating effect
of the freeze damage on N absorption was not known. It
was decided to determine if the same trend in N absorption
would prevail during the warm spring weather.
Nine unused trees planted at the same time as those
in Experiment 1 had been left unfertilized from the sum-
mer of 1978 through March, 1979. The 3 replications of
single tree plots were arranged in a randomized complete
block design, having as treatments unfertilized control,
IBDU and NH4NO3 as N sources. No irrigation treatment was
applied, however. Natural precipitation and routine
irrigation of the adjacent mature citrus grove provided
the only water supply.
Data from Experiment 1 indicated that NH4NO3-ferti-
lized trees had higher N content than IBDU-fertilized
trees. The effect of the freeze damage on N absorption
from the 2 N sources was not known. It was necessary
therefore to establish if the same trend would be manifest-
ed in warm weather. Hence, the trees in Experiment 2 were
deliberately grouped on the basis of their trunk diameter;
IBDU was applied to the largest trees, and NH4NO3 to the
smallest trees. Control trees were intermediate. The
same rate of fertilizer-N in Experiment 1 was hand broad-
cast in a 3 x 3 m area around the tree on March 23, 1979.
The experiment was terminated on May 8, 1979, 6 weeks after
Three 16-year-old 'Pineapple' orange trees in an
existing rate and fertilizer-N source study were used for
this experiment. The trees had been fertilized with the
same rate of NH4NO3 since April 5, 1973,until September 23,
1977. In the current experiment,one of the trees received
a total of 707.2 g of N per year, in 3 equal applications
on March 2, 1978, June 11, 1978, and October 25, 1978, as
IBDU. Another tree received the same rate of N as NH4NO3
on the same dates. The fertilizer was applied by hand
broadcasting around the tree. Other nutrients were
adequately supplied to the fertilized trees. The third
tree which did not receive any fertilizer-N at the speci-
fied dates was used in this experiment as the unfertilized
In order to study N movement in uncropped land 1 cir-
cular fallow plot 6 m in diameter, which corresponded to
the tree canopy diameter, was fertilized with IBDU on
October 25, 1978. Only one-third of the total annual ap-
plication in the tree plots was used. Another similar
plot was fertilized with the same rate of N as NH4NO3. A
third plot was left unfertilized as a control. The 6 un-
replicated plots in the experiment were subjected to the
routine management of the adjacent grove. The experiment
was terminated on February 17, 1979, 17 weeks after the
fall application of fertilizer-N, at which time the plots
had received 25.90 cm of rainfall and irrigation since
October 25, 1978. Evapotranspiration for the duration of
the study was estimated from climatological data (55).
Field Sampling Procedures
Soil Water Sampling
Changes in soil water content were measured with a
Nuclear-Chicago neutron probe (P-19) and a scaler (53).
One aluminum access tube for the probe was installed about
45 cm from every other tree in Experiments 1 and 2. In
Experiment 3, one tube was installed just inside the drip-
line of each tree and at an equivalent distance from the
center of each fallow plot.
One-minute readings were taken at depth increments
of 15 cm, beginning at 15 cm. The readings were converted
to % water content by volume using a calibration curve
developed from previous sampling in the experimental area
(53). The water content of the surface 15 cm was deter-
mined gravimetrically. A composite of duplicate 2.5 x
15 cm cores of soil per plot in the 3 experiments was col-
lected and its gravimetric water content determined by
over-drying to constant weight at 1050C. The water content
was transformed to a volume percentage by multiplying the
% weight by the average soil bulk density (1.54 g/cm3).
Water content in the soil profile was calculated by multi-
plying the average volume ratio by the depth of the entire
profile and results expressed as cm of water.
A composite of duplicate 2.5-cm diameter soil core
samples was taken within 30 cm from the tree from 0-7 cm
and 7-15 cm, and thereafter at 15-cm increments down to
60 cm in Experiments 1 and 2. Soil samples were similarly
taken at the tree canopy dripline down to 120 cm in Experi-
ment 3. In Experiment 1, samples were taken when cumula-
tive rainfall exceeded 1.25 cm. After the initiation of
the irrigation treatment, samples were taken only at the
termination of the experiment. In Experiment 2, samples
were taken after 2.28, 9.14 and 17.87 cm of cumulative
rainfall and irrigation. Samples were taken following more
than 4.87 cm of rainfall and/or irrigation in Experiment 3.
In order to prevent free movement of water from the
soil surface through the sample holes, each hole was filled
with white sand, obtained from elsewhere, and marked to
avoid future sampling in that immediate vicinity.
Plant Tissue Sampling
Tree growth measurements in Experiments 1 and 2 were
obtained by measuring the diameter of the trunks and the
destructive sampling of the tree. Tree trunks were painted
at a point 15 cm above the bud union and the initial and
final trunk diameter recorded. Trees were excavated by
digging a trench 60 cm from the tree and recovering the
root system as much as practical. Each tree was fraction-
ated in the laboratory into 7 components: fibrous and
lateral roots, taproot, rootstock trunk, scion trunk,
crotch branches, green twigs and leaves. The root system
was rinsed in running tap water and the fresh and oven-dry
weight (700C) of each component obtained.
In the mature tree experiment, fruits were harvested
on February 17, 1979, and the total fresh weight recorded.
One 30-fruit sample from each tree was ground in a com-
muniting machine, and 2 subsamples oven-dried at 700C for
dry weight determination and chemical analysis.
The number of leaves on each mature tree was esti-
mated by the following procedure. Leaves in a 30 x 30 cm
frame from 4 sides of the tree were stripped and the
average number in the frame determined. The leaves were
oven-dried for 48 hours at 700C and the total dry weight
determined. The total dry weight of the sampled leaves
was used to calculate the average dry weight of a leaf.
Tree height and tree width in 2 directions (north-south and
east-west) were also measured. The canopy surface area was
then calculated from the formula:
S = 27W/3h2[(W/16+h2) 15-(W/4)3
where h = height and W = width of the tree (98). The
number of leaves on a tree was then estimated from a know-
ledge of the canopy surface area.
Sample Preparation and Analysis
All soil samples were air-dried and sieved through a
2 mm screen. Total soil N was determined by a semi-micro
Kjeldahl procedure using a salicylic-sulfuric acid mixture
(12). Soil samples from the 4 replications in Experiment
1 were combined for the determination of exchangeable NH4
and NO, by a steam-distillation procedure using Devarda
alloy and MgO (13). Samples were not sufficient for the
determination of these ions for the second sampling date
in Experiment 1 and for all samples in Experiments 2 and
3. Total N was expressed as % N on a dry weight basis and
NH4-N and NO3-N as ppm. When a quantitative account for
N was being made, these values were converted to g or kg
of N in the soil profile. This was obtained as the product
of area, depth, average bulk density (1.54 g/cm3) and aver-
age N concentration in the profile.
In Experiments 1 and 2, the N content of the profile
was calculated for the total area fertilized (90,000 cm2)
but since the root system of the young trees occupied an
area less than 60 cm diameter, the N content was also cal-
culated for the latter area (2828 cm2). The depth used in
the calculations was 60 cm. For the mature tree experiment,
the area and depth were 28.28 m2 and 1.20 m, respectively.
The recovery of fertilizer-N was obtained by subtracting
the total soil profile N content of the unfertilized plot
from that of the fertilized plot. This value was expressed
as the percentage of the amount of N applied.
Plant Tissue Samples
All plant parts were ground in a Wiley Mill and passed
through a 20-mesh screen. Total N was determined by a
semi-micro Kjeldahl procedure using 0.5 g subsamples (12).
Nitrogen values were expressed both as % of dry weight and
in absolute amounts. The latter were obtained by multiply-
ing % N by the total dry weight of the component part. The
recovery of fertilizer-N by the trees was obtained by
subtracting the total N content of the unfertilized tree
from that of the fertilized tree and the result expressed
as the percentage of the amount of N applied. In the ma-
ture tree experiment, the N content of fruits and leaves
from the unfertilized tree was similarly substracted from
those of the fertilized tree.
In Experiments 1 and 2, data pertaining to tree growth,
tissue N content, and soil total N content were submit-
ted to analysis of variance. In Experiment 1, these data
were analyzed in a 3 x 3 factorial arrangement. Data for
Experiment 2 were treated as a randomized complete block
design. Significant differences were determined by Duncan's
multiple range test. Soil water data and those from the
mature tree experiment were not analyzed.
RESULTS AND DISCUSSION
Experiments 1 and 2
Experiment 1 was prematurely terminated 14 weeks af-
ter fertilizer application and 6 weeks following the first
irrigation because the trees were beginning to show the
effects of freeze damage. This was especially apparent in
the unfertilized control trees where extensive leaf drop
occurred. Since the complicating effect of the freeze
damage on N absorption was not known, and NH4NO3-fertilized
trees appeared to have higher N content in comparison to
IBDU-fertilized trees,Experiment 2 was started to deter-
mine if the same trend would prevail in warm weather.
Therefore, the trees in Experiment 2 were deliberately
grouped such that IBDU-fertilized trees were larger in
trunk diameter than control and NH4NO3-fertilized trees
Tree Growth Analysis
The young trees in both experiments generally showed
vigorous growth regardless of treatment. Trees fertilized
with NH4NO3 and IBDU had denser canopies than the unfer-
tilized control trees, and those fertilized with NH4NO3
SH H H
^y rm ^
m < ^
f< r1 r-
O> I 1
%D 00 1
o H 0
.Q Cd ..
-4 I 0
u H z
had darker foliage than IBDU-fertilized trees. Leaves
from the control trees in Experiment 1 were dull green and
began to shed toward the end of the study, indicating that
N was becoming a limiting factor (22) and the trees were
responding to the cold weather. Leaves from the control
trees in Experiment 2 were also dull green, but they did
Ammonium nitrate and IBDU appeared to increase the
dry matter content of nearly all tree component parts as
compared to the control in Experiment 1 (Table 2). A dif-
ference in dry matter content was particularly apparent in
the leaves collected from both of the fertilizer-N treat-
ments which were statistically greater than the control.
The NH4NO3-fertilized trees showed numerically greater dry
weight values compared to the IBDU treatment in the more
succulent components of the tree, but they were not statis-
It is not plausible to compare the effect of N
sources on the growth of the tree parts in Experiment 2,
because of the initial tree size differences. Nevertheless,
there were differences (nonstatistical) particularly in
green twig dry weight indicating a greater response to
NH4NO3 fertilization as compared to the other treatments
(Table 2). Trees fertilized with NH4NO3 were the smallest
and yet had numerically greater green twig dry weight than
the rest. This trend was similar to the one in Experiment
1. Significant differences between treatments in fibrous
U I 0 (0 (0 (
U) I 0H m
I> C N C M 130 LO in 1-1
00HI a 0 C H
C4 I m l A;
I H H
Q.) 0o 0o co ,- o r-i
OU I .)
0 r I n H L o
u 1 4 c;
0 0 CD CT l
N -H)I ., (
* I .
, Q I
S I U)
o0 N C CN H H
S 4OJ I VC C
0 I I-I I
04 I o m n 0
aI N N m
0 U0 1 z o
(0 0i .
f p I O,
' 0 c c Lo
O O *.
0 C D H O1 0 C 00 I
L4 LA I.0 C( (0 H
4-4 i Ia
4I O 5 I .
( o S0 0 z 0
0 m 0 m
z zH z
H z z a)
and lateral roots, and the scion trunk in Experiment 2, but
not in Experiment 1, showed that these differences were a
reflection of the initial tree size differences in Experi-
ment 2 rather than treatment effects.
Irrigation level did not statistically influence the
distribution of dry matter or the N content of the various
tree fractions in Experiment 1 (Table 3). There were no
significant interactions between N source and irrigation
level. Data related to irrigation level effects, except
for those in Table 3, have been omitted from this study be-
cause irrigation level effects were not statistically dif-
ferent and there were no irrigation level x N source in-
The dry weight of the total root system and the above-
ground parts of the tree in Experiment 1 were consistently
highest for the NH4NO3 treatment (Table 1), followed by the
IBDU and control treatments; however, the data did not show
any statistical differences. Nitrogen source also had no
statistically significant effect on total tree dry weight
or % trunk diameter increase. Nevertheless, N fertiliza-
tion resulted in larger % trunk diameter increase over the
control trees in both Experiments 1 and 2 (Table 1).
Trunk growth was greater in Experiment 2 than in Experi-
ment 1 presumably because of the beneficial effect of the
warm spring weather on tree growth. Total tree dry weight
in Experiment 2 did not appear to have been altered by
I i r-
I N N N
I H %D m
I H H
o m m
I 0 H
I N N
I Lf l (N
I ) m
O O --l
o N m
0 0 0
0 0 -
o O -'-
z >- 33
treatments presumably due to the short duration of the ex-
periment. It reflected the trend in the initial trunk di-
ameter measurements (Table 1).
Irrigation level had no statistically significant ef-
fect on the dry weight of the root or shoot system of the
trees in Experiment 1, or the dry weight and the % trunk
diameter increase of the tree. There was, however, a ten-
dency towards greater trunk growth with an increase in
irrigation level. Koo (55) reported a similar trend for
mature citrus trees in Florida.
Several factors may have contributed to the absence of
an irrigation effect in Experiment 1. These would have in-
cluded duration of the experiment, the quantity and fre-
quency of irrigation as well as the time of year. The
trees were irrigated only during the last 6 weeks of the
experiment. It is possible that irrigated trees did not
have enough time to respond to the irrigation treatments.
Furthermore, it has been shown in Florida that extra water
applied to mature citrus trees between June and December,
the approximate period of this study, did not contribute
substantially to tree growth (55).
The lack of significant N source effects on tree
growth in Experiment 1 can possibly be attributed to a di-
minished growth rate in winter (25). Also, the 14 weeks of
experimentation probably were not long enough for the
treatments to show their maximum effects on tree growth.
Patten and Domoto (78) and Miller (70) recently showed that
N treatments had no significant effect on apple tree growth
during the first growing season.
In Experiment 1, N source statistically altered the
pattern of dry matter distribution (Table 4). A greater
proportion of the total dry weight was located in the tap-
root of the NH4NO3-fertilized and control trees as com-
pared to the IBDU-treated trees. Both NH4NO3- and IBDU-
fertilized trees had statistically greater proportion of
their dry weight distributed in the crotch branches in com-
parison to control trees. The % dry weight of the scion
trunk of the NH4NO3-treated trees was less than that of
either the control or the IBDU-treated trees. The fact
that the dry weight distribution was similar to the actual
dry weight of certain tree parts (Table 2) makes it doubt-
ful that all of these were true treatment differences.
Moreover, treatment means in dry weight were not statis-
tically different except for the leaves. Some of the sig-
nificant treatment differences may very well be a reflec-
tion of initial differences in the trees. Nevertheless,
there was a trend for NH4NO3-treated trees to have a
greater dry matter in crotch branches, green twigs and
leaves in comparison to IBDU-fertilized and control trees.
As was the case with the dry weight of the tree fractions,
irrigation level did not significantly affect the distri-
bution of dry matter.
p a) I
0 I 0
0 0 -
rd H H
< 3 <
e e z
C Q v
U H 1
Hm r OD
, N C m
u I H
The distribution of dry weight was similar in Experi-
ments 1 and 2 (Table 4). In Experiment 2, the control
trees had proportionally more of their dry weight in the
rootstock and scion trunks as compared to the fertilized
trees. The % dry weight in the green twigs of the control
trees was, however, significantly less than in the NH4NO3-
treated trees; but the greatest difference was in the leaf
fraction. The % dry matter in the leaves of the fertilized
trees was about twice as much as in the unfertilized con-
trol trees. Furthermore, the leaves of the fertilized
trees in Experiment 2 represented a higher % of the total
than in Experiment 1. This is a reasonable expectation be-
cause of the stronger growth flush in the spring than in
Tree N Content and Distribution
The main effects of the N sources in Experiment 1
showed that the N concentrations were highly significantly
different in all component parts of the tree, with NH4NO3
treated trees having the highest values followed by the
IBDU-treated and control trees (Table 5). Nitrogen con-
centration was generally highest in leaves followed by
fibrous and lateral roots, green twigs, taproot, crotch
branches, rootstock trunk, and scion trunk. This order is
similar to the data of Barnette et al. (5) for a mature
tree of a different cultivar. Furthermore, the N con-
centrations in the current study were higher than the
w I U U 4
> I i-4 r-H Ln IV (N co
0 I .
I oo r r L
SN m (n (N CO Cn -
c u < um< a
0 U 0 r N 0N 0 00 4)
SOHH O H
4 I d
U 4 a m to
0) U 1 U 11 H
0 z 1 ," I V ,- D N 0-
O CI ^ o C M -
UO I 0 H o 1 Ln 0
(0 -*H t n o LA 0 0)
0- 4 1
U I 0 4 I 0; ; C
0) 94 u m < Z S .Q
4 mr r- co C 1
0 m0 rn ko m v o +1
U p F. .
04 000 1 C Hr ( )
0) 0 H
S I 0 H o -i C
0 1 m IV 0o r, 4 j
Cz p I o cN EL )
4 0 Ioi
z to u
a0 ro 0O
0 0d8 1
U I H H.
S WHI U m u m 0-
::j 0 I r- H Ln ON m r LA
0 0 p It0 (N 00 M 4W dp
to 0 L
Q41 0 Hr N 0 -l (N H
Z *H *O CdI
4 q 4J C
4J rH r-1 C
o 4 O Z0 n 0 Z 0 .4
0) p- I 0 1 1 0 -i H )
W 0Q m 0 m p a
U H Z U H Z Cr
in W1d z
E-4E N N
values reported by Barnette et al. (5). Such differences
are not unusual when a young, vigorously growing tree is
compared to a mature, fruiting tree (99). Also, the rate
of N application in the current study was higher.
No significant differences due to irrigation level
were detected in the N concentration of the various tree
components- Table 3).
The results obtained in Experiment 2 were consistent
with those in Experiment 1 in that the order of N concen-
tration in all tree parts was NH4NO3- > IBDU-N > control
(Table 5). Except for rootstock and scion trunks, these
values were significantly different from one another at the
1% level. The N concentration for all tree parts was
higher in the NH4NO3-treated trees than in their counter-
parts in Experiment 1, particularly in the leaves, fibrous
and lateral roots, and in the taproot. It would not be un-
reasonable to assume that these relatively higher concen-
trations were partly a consequence of concentration ef-
fects, in view of the smaller trees assigned to the NH4NO3
treatment. Also, the higher soil temperature in the spring
could have been conducive to more N absorption by the
The total amount of N in a tree part, the product of
the dry weight of that part and the N concentration on a
dry weight basis, was not statistically affected by irri-
gation level in Experiment 1. There was, however, a
statistically significant relationship between N source and
total N content of tree parts (Table 6). Numerically, the
trend was similar to the N concentration in that NH4NO3-
fertilized trees had the highest values followed by IBDU-
fertilized and control trees. The data were, however, less
The N content of the tree parts was reflected in the
total amount of N in the tree (Fig. 2). Differences due
to N source were highly significant with NH4NO3-treated
trees containing about 3 and 1.5 times greater total N than
the control and IBDU-treated trees, respectively. No sta-
tistically significant differences were detected due to ir-
In Experiment 2, the total amount of N in each tree
part (Table 6) and consequently in the entire tree (Fig.
2) was lower than in Experiment 1 probably because of the
smaller trees used in Experiment 2. No meaningful impor-
tance can be attached to the statistical differences in
Experiment 2 in the total N content because of the inequal-
ity of tree sizes. Despite the fact that the trees selected
for the IBDU treatment were larger and had 1.6 times the dry
weight of the NH4NO3-treated trees, total N in the NH4NO3-
fertilized trees was 1.05 times that of the IBDU-fertilized
trees. Equivalent biomass, including the root system, would
absorb more N from NH4NO3 than it would from IBDU in a short
period as in Experiment 1.
T I m Q a
0) I 3 r -- C) 0o --
> I 1 n Lf) N 1( 1 H i
4> I a a CN H N D
S I .
1 I 0 0 H 0 0 0
I 0 0n c m )
0 z 0 0 oH r
M I ci)
O.I o o o o o o
-H i m 0 rJ C
MI n m o o <
U ) ,CIN cy r N i
00 0 0 ,0 0 0 4
0 I U)
U-4I '. 0 H mCr 0 4
C NMI N NcLtU O
0 OCW p m 4 C
1 4HCI 0 H N 0 0 0 CO
u 0 I .H4-1i
CI a )4
0& CI 00 0 0 0 0 )P
c O IC .U) t i
Sr m 0 rN
4 Oflc 0 O 000 N
o H H 0 0 H
0 M .
0 0 I w a n d
w aI 0 D w O .))
H0 0 o o o o 0 o0-1 i
O 0 N "
I CC 1)
0 m 0 m -o
WI 4I &
TOO ud O m .
-H0 0H00 0
Ui U E am
u 0 4^) z r s
0 ^ 0 C r H
r0 4 (
I I I 1 1 11 1 1 11 I
S'N 1V101 33yIJ
-< !llllll I I llll 1111 1111
There was no significant relationship between the dis-
tribution of N in the various tree parts and irrigation
level in Experiment 1. Nitrogen source significantly af-
fected the distribution of N in the taproot and scion trunk
(Table 7). Regardless of N source, 30-34% of total tree N
was located in fibrous and lateral roots. Leaves repre-
sented only 10-13% of the total dry matter of the tree but
contained 30-33% of the total N in the tree. These figures
correspond to those reported in earlier work for mature
citrus trees (20, 63). Sixty, 57, and 55% of the total N
in the tree were found in the aerial parts of the control,
IBDU- and NH4NO3-fertilized trees, respectively. This may
suggest that the more readily available the N source, re-
sulting in luxury consumption, the greater the proportion
of N stored in the root system (86).
As in Experiment 1, N source significantly affected
the distribution of N in Experiment 2 (Table 7). Compared
to Experiment 1, however, the proportion of N was greater
in leaves and less in the fibrous and lateral roots of the
fertilized trees than the control. The proportion of N
in the taproot of NH4NO3-fertilized trees was higher than
for IBDU-fertilized and control trees in both experiments.
Whether it is a characteristic of citrus trees to accumu-
late NH4NO3-N in the taproot or some factor related to the
time of sampling needs further investigation. In contrast
to the data in Experiment 1, 57, 67 and 66% of the total N
0 Z I
0 S4 dP
*H ( I
Er r 1
00 CM CN
m r- a.
LA C 00
01 c' M
0 0 0z
0 om o
m < <
LA 0 LA
m m o
N l ,w
.0Q Q (
00 '-0 C
(d IQ OQ
0ri m 0
U H 0
0 .0 ,Q '
t N (N
in the tree were found in the aerial parts of control, IBDU-
and NH4NO3-treated trees, respectively. The greater pro-
portion of total N in the aerial parts of the fertilized
trees in the spring than in the fall-winter may reflect the
more vigorous growth activity in the spring. Also, trans-
location of N from roots may be greater in the spring. The
difference probably resulted from a lower activity of the
nitrate reducing enzyme system during periods of low soil
temperature in the winter (38).
Soil Profile Content and Distribution of N
Soil samples in Experiment 1 were taken to the 45-cm
depth after 1.42 and 2.26 cm of cumulative rainfall on
October 3 and October 19, 1978, respectively. Analysis of
the first set of samples indicated that NO3-N from NH4NO3
had moved to this zone (Fig. 3-A). As a result the sam-
pling depth was extended to 60 cm. A third set of samples
was collected after 4.87 cm of cumulative rainfall on
October 31, 1978. After the initiation of the irrigation
treatments, a fourth set of soil samples was obtained on
January 9, 1979.
At the first sampling (Fig. 3-A), there was evidence
that some N03-N from NH4NO3 had moved down to the 45-cm
depth while NH4-N from the same source had only moved to
the 30-cm depth. A greater proportion of the NH4-N was
still retained in the 0- to 15-cm zone. Only a small
amount of NH4-N from IBDU had moved to the 30 cm depth
Fig. 3. Effect of N source on the exchangeable NH and
NO3 soil content. A, B, and C correspond to the
1st, 3rd and final soil sampling dates on October
3, October 31, 1978, and January 9, 1979, respec-
tively. (Expt. 1).
1BOU NH4 N
;- -- --' -5-,,,
0-15 15-30 30-45 45-60
presumably because of the sparingly soluble nature of IBDU
which remained in the surface soil. Eighty-five percent of
the applied N from NH4NO3 was accounted for in the 0- to
45-cm depth at the first sampling. The difficulty of ex-
tracting IBDU with 1 N KC1 precluded the determination of
NH4 and NO3 in the 0- to 15-cm depth. The total N proce-
dure, however, indicated that 94% of the applied N from
IBDU remained in the 0- to 45-cm depth at the first sampling.
These trends in inorganic N concentrations were also
revealed in the soil total N content in that values for
the fertilized plots were higher than those for the control
plots (Fig. 4-A). Soil samples were not sufficient for the
determination of exchangeable NH4 and NO at the second
sampling. Nevertheless, there was an indication that N
from NH4NO3 had probably moved past the 45-cm depth (Fig.
At the third sampling it was apparent that NH4-N and
NO3-N from NH4NO3 had leached down to the 45- to 60-cm
depth (Fig. 3-B). This suggests that at least a part of
the downward movement involved the un-ionized NH4NO3
fertilizer. The well-drained nature of Astatula fine sand,
and its low organic matter content could have been factors
contributing to this movement. Some NH4-N and NO3-N from
IBDU had also moved down to the 60-cm zone, but on a smal-
ler scale relative to NH4NO3. Ninety-one percent of
Fig. 4. Effect of N source on soil total N at the 1st (A)
and 2nd (B) sampling dates on October 3 and Octo-
ber 19, 1978, respectively. Mean separation with-
in depth by Duncan's multiple range test, 1%.
2.26 cm of
the applied N from IBDU remained in the 0- to 60-cm depth
at the third sampling; 82% of the NH4NO3-N could be ac-
counted for in the same zone. Data for soil total N (Fig.
5-A), although less consistent in the 15- to 45-cm depth
than those for exchangeable NH4 and NO3, supported the
trend in the downward movement of N from both fertilizer
The final soil sampling was completed 14 weeks after
the trees were fertilized. Much of the N from NH4NO3 had
been depleted from the soil profile (Fig. 3-C). Only 30%
of the applied N from this source remained in the 0- to
60-cm zone. A part of the depletion could be attributed to
root absorption. Soil water contents (Table 9) were gener-
ally below the maximum water storage capacity; however,
rainfall events of 5.0, 1.3 and 2.6 cm on December 28,
December 29, 1978, and January 3, 1979, respectively,suggest
that leaching of some NH4NO3-N may have occurred beyond the
60-cm depth. Furthermore, provisions were not made in
this study for the determination of gaseous losses of N
which may have also occurred. As in the previous samplings
the concentrations of exchangeable NH4 and NO3 from IBDU
in the 15- to 60-cm depth were lower than those from NH4NO3
(Fig. 3-C). There was also no indication of IBDU-N having
been leached beyond the 60-cm depth since 66% of the applied
N could be accounted for within this depth. In comparison
to the other N sources, a greater proportion of the IBDU-N
was retained in the surface soil (Fig. 5-B).
(O 0 4
C Hl rl
4-J (I q)
0 f4 0 4-)
d 0 Q) P *
0 $ >i ( 4-J
4J 0 r-4 4 4J
O ) 4)
S> U) r-
0 ) -H H
S 4 0r-4 U0
0 ( a t
4-4 -< a ra
0 >1 zri
Srd 1 -l
:3 34J E
4-1 *~' r '
oE >i a
C n A(
-_ Z 1 = # -(f>
Table 8. Quantitative estimates of
transpiration from mature
evaporation and evapo-
Period Rainfall & Class "A" pan Evapotrans- Potential
irrigation, evaporation piration net leach-
cm cm/day cm/day ing, cm
Oct. 25- 4.87 0.39 0.22 +1.35
Nov. 9- 2.63 0.32 0.23 -4.96
Dec. 12, 18.40 0.27 0.13 +9.69
ZExcess water (+) and deficit (-) to effect percolation.
Soil water content of young tree plots 24 hr.
after a rainfall or irrigation event.
Sampling Cumulative rainfall Soil water Field capacity
date and irrigation, cm (0-60cm), (0-60cm),cm
Oct. 3, 1.42 3.91 4.18
Oct. 19, 2.26 3.47 4.18
Oct. 31, 4.87 3.77 4.18
Jan. 9, 25.12 4.00 4.18
In Experiment 2, sets of soil samples were taken 3
times on April 2, April 27 and May 4, 1979, following 2.28,
9.14 and 17.87 cm, respectively, of cumulative rainfall
and irrigation. These samples were not sufficient for the
determination of exchangeable NH4 and NO Soil total N
concentrations in this experiment were higher than those
from Experiment 1 (Figs. 6-8). Contamination from the
regular fall (1978) and spring (1979) fertilization of
adjacent mature trees may have contributed to the higher
soil values in Experiment 2 as compared to Experiment 1.
As in Experiment 1, however, there was a trend towards
decreased soil N content with increasing cumulative rain-
fall and irrigation (Figs. 6-8).
At the first sampling there was an indication of
some NH4NO3-N having moved to the 60-cm depth (Fig. 6),
while only a small amount of IBDU-N may have moved to the
30-cm depth. Some N from IBDU appeared to have reached
to the 45-cm depth as indicated by the bulge at the second
sampling (Fig. 7), but the data from Experiment 1 (Fig. 3B-
C) suggest that this could be a sampling error or error
inherent in the soil total N procedure. As in Experiment
1, a higher proportion of the N from IBDU, relative to the
other sources, was still retained at the surface soil at
the last sampling (Fig. 8).
SI I I- >
0 0 0 0
C /0 0 N O
co 4-4 r-
0 0 I 0
C-) o 0 oa2
iM A8Q % 'N I- lS
-' / <" "
e 0 2 "g
.LM A80 % 'N lb0S
1M AdO % 'N IDOS
Recovery of Fertilizer-N in the Soil-Plant System
The recovery of fertilizer-N in the soil profile and
in the tree was calculated by subtracting the N content of
the control from the fertilized plots and expressing it as
a percent of the amount of N applied. For the NH4NO3 treat-
ment in Experiment 1, both soil total N and inorganic N were
used in the calculations. Data for the inorganic N procedure
indicated that, after 14 weeks, 30% of N apparently remained
in the 0- to 60-cm soil profile where NH4NO3 had been applied
(Table 10). Sixty-six and 41% of N from the IBDU and NH4NO3
sources, respectively, remained in this zone when the cal-
culations were based on soil total N values. The discrep-
ancy in the NH4NO3-N figures could have resulted from the
larger errors inherent in the total N procedure as compared
to the inorganic N procedure (12, 13).
In Experiment 2, where the initial soil total N con-
tents were higher as compared to Experiment 1, the total
amount of N in the 0- to 60-cm depth was also greater
(Table 10). On the basis of total N calculations, 70 and
62% of the applied N were present in the soil from IBDU and
NH4NO3 sources, respectively. Higher soil N recoveries in
this experiment in comparison to those in Experiment 1
probably represented the difference in duration of each
experiment, 6 and 14 weeks, respectively.
The relatively high N recoveries from soil where IBDU
had been applied in both experiments suggested very little,
l l 0
(N 'D Ln
.0 N N
O H H
I I I
I I oI
I I I
I I I
I I I
I 0 (N
I I I
I I I
if any, leaching loss of N from this source. Some leaching
of N from NH4NO3 may have occurred. Other complex processes
involved in N reactions, including denitrification and other
gaseous losses, may have contributed to lower soil recover-
ies of N from NH NO3 relative to IBDU.
The recovery of N by the tree was calculated for the
total area fertilized (90,000 cm2); but as the root sys-
tem occupied a circular area of less than 60 cm diameter
(2828 cm2), N recovery was also calculated for the latter
area. In Experiment 1, on the basis of the smaller area,
25 and 45% of the N added were recovered in those trees
fertilized with IBDU and NH4NO3, respectively. It was only
0.79 and 1.42% for the same respective sources when the
larger area was the basis of calculation. The differences
in N recoveries from the 2 areas imply a higher efficiency
of N use if the fertilizer is applied on a smaller area
around the tree rather than on a larger one where the po-
tential for N loss may be great.
It has been suggested that plant recovery of N by
the difference method generally yielded higher values com-
pared to the tracer procedure, presumably because ferti-
lization may stimulate additional N uptake from soil
sources (109). This might have contributed to the high N
recoveries, particularly where NH4NO3 was the source. The
higher tree recovery of NH4NO3-N appeared to be related to
the ready availability of this fertilizer in comparison to
IBDU, a greater proportion of which was retained in the
On the basis of the smaller area calculations, trees
fertilized with IBDU and NH4NO3 in Experiment 2 recovered
only 15 and 16%,respectively of the applied N (Table 10).
Recoveries from the respective sources were 0.47 and 0.52%
for the large area. The lower tree recovery of applied N
in Experiment 2 vs. Experiment 1 was partly related to the
smaller trees used in the former experiment and partly be-
cause the experiment was run only for 6 weeks. The trees
did not therefore have enough time to absorb N to the same
extent as in Experiment 1. Also, the narrow range in the
recovery of applied N seemed to be due to the fact that
IBDU-fertilized trees were larger than NH4NO3-fertilized
trees. If NH4NO3-fertilized trees had been of equal size,
the difference in the recovery between the 2 fertilizer
sources could have been larger.
A common approach for comparing the relationship of
fertilizer sources with mineral nutrient absorption by
plants in the vegetative growth phase is to relate the to-
tal dry matter of the plant to its total nutrient content,
i.e., dry weight/total nutrient ratio. The lower this ratio,
the greater the nutrient absorption from that source. As-
suming the ratio is linear with time, the N status of the
trees was projected by plotting the ratio data for
0 0 0i
0 c0 0 o
N / IM A80 OIVd
Experiments 1 and 2 (Fig. 9). Theoretically, up to 26
weeks after fertilizer-N application NH4NO3-N would con-
tinue to be absorbed more than IBDU-N, but after 28 weeks
the reverse would be expected. There is evidence that the
period of extended availability from soluble N sources,
after the initial flush of growth of vegetable crops,
approaches that of the more slowly available N sources
(115). In view of the extremely soluble nature of NH4NO3,
however, most of the N will have probably been depleted
from the soil after 26 weeks.
Total apparent N recovery in the soil-plant system in
Experiment 1 amounted to 76 and 91 for NH4NO3- and IBDU-N,
respectively (Table 10). The total N procedure indicated
a higher recovery of N from the NH4NO3 plots in comparison
to the soluble N procedure. The apparently variable total
N recovery was probably due to the larger errors associated
with the use of the total N procedure in soil analysis vs.
the soluble N approach (12, 13).
In Experiment 2, where only soil total N was analyzed,
85 and 78% of the added N from IEDU and NH4NO3, respec-
tively, were apparently recovered in the soil-plant system
when the small area was the basis of calculation. It was
not possible to account for the deficit in N balance in
both experiments. Nevertheless, soil N data in Experiment
1 (Fig. 3A-C) and rainfall events of 5.0, 1.3 and 2.6 cm with-
in a 6-day period imply that some leaching of N, at least from
NH4NO3, may have contributed to the deficit. Processes
involved in the N cycle, including denitrification and
other gaseous losses of N may also have been reflected in
the unaccounted for part of the balance.
Nitrogen Distribution in Bearing Orange Trees
Fruit yield and estimate of leaf number
Each of the control, IBDU- and NH4NO3-fertilized trees
showed fresh fruit yields of 0.7, 3.7 and 3.7 field boxes
of 40.8 kg, respectively (Table 11). These data indicated
that yields can be severely limited by a lack of N. Ni-
trogen-fertilized trees had about 4 times as much total
fruit dry weight as the control tree. The unreplicated
nature of the experiment, however, precluded the establish-
ment of definite conclusions pertaining to the yield data.
Nevertheless, it is generally accepted that increased N
supply results in increased citrus fruit yield. Data from
a Florida experiment (87) with 'Pineapple' orange showed
that as the rate of N was increased, up to 202 kg/ha/year,
The estimated leaf number of the fertilized trees was
in reasonable agreement with the 70,200 leaves estimated
by Barnette et al. (5) for a grapefruit tree in Florida,
but lower than the 92,708 leaves sampled from a 12-year-old
'Valencia' orange tree in California (110). Trees in the
E-4 -4 Z
I -H U
4o) 1 ,
California study were, however, larger as indicated by
their height and width. The control tree in the current
study had a lower leaf number than either of the fertilized
trees by virtue of a lesser leaf density which can be at-
tributed to a lack of N.
Fruit and leaf N content
Several approaches have been used to determine citrus
fertilizer needs. One school of thought advocates return-
ing to the soil the amount of nutrients removed in the crop
of fruit. Embleton et al. (33) considered this approach
unsound, particularly with respect to the less mobile nu-
trients. Nevertheless, because they represent part of the
N balance sheet, data related to crop removal of N when
combined with leaf analytical data have a significant po-
tential for guidance in the N management of citrus groves.
The concentration of N in the fruit from the ferti-
lized trees was higher than that of the unfertilized trees
(Table 11). As a result, when both total fruit dry weight
and N concentration are considered, the amount of N re-
moved in fruit was in the order of NH4NO3- > IBDU-fertilized
> control trees. Five times more N was removed in fruit
from fertilized trees as compared to the control tree.
Reitz (89) reported that in Florida 36.3 kg of N are re-
moved in every 20,400 kg of fresh fruit. Recalculation
of the data in Table 11 indicates that 32.2 and 29.9 kg
of N would be removed in 20,400 kg of fruit from the NH4NO3-
and IBDU-fertilized trees, respectively.
Leaf N concentration followed the same trend as in the
fruit. Nitrogen concentration for the 3 single-tree plots
was in the order: NH4NO3- > IBDU-fertilized > unfertilized
trees. This clearly reflected the more rapid absorption of
NH4NO3-N over IBDU-N and the control. The total amount of
N in the leaves was, however, slightly larger for the IBDU-
treated tree because of its greater leaf number. It is
doubtful if the greater leaf number is related to IBDU
fertilization since the duration of the experiment was
short. The total amount of N in the leaves of the fertilized
trees was in agreement with the 200 g of N reported by Cam-
eron and Appleman (20) for a 10-year-old 'Valencia' orange
tree. This close agreement lends added credence to the pro-
cedure used in this study for estimating the leaf number.
Soil Total N Content and Distribution
Fertilized fallow plots received 33% (235.7 g/6 m dia-
meter area) of the total annual amount of fertilizer-N ap-
plied to the fertilized tree plots (707.2 g/tree). Three
sets of soil samples were collected: on November 9, December
11, 1978, and February 16, 1979. These sampling dates corres-
pond to cumulative rainfall and irrigation of 4.87, 7.50 and
25.90 cm, respectively, since October 25, 1978. Samples were
not sufficient for the determination of exchangeable NH4 and
NO3; hence, only the total N procedure was used.
At the first sampling, some N from both IBDU and NH4NO3
appeared to have moved at least down to the 45-cm depth in
the fallow plots (Fig. 10). The amount of NH4NO3-N
leached, as reflected by the lower concentration in the 0-
to 15-cm depth and higher concentration in the 15- to 45-
cm zone, was greater than that of IBDU-N. In the ferti-
lized tree plots, NH4NO3-N appeared to be depleted rapidly
from the 0- to 15-cm zone, while a considerable amount of
IBDU still remained in this zone (Fig. 11). The apparent
rapid depletion of NH4NO3-N from this zone could be attri-
buted to root absorption in view of the readily available
nature of NH4NO3 fertilizer. Higher N concentrations, par-
ticularly for the IBDU treatment, in the 75- to 105-cm
depth, as compared to the control, could have been due to
the residual N from previous applications.
Data for the second sampling date (Fig. 12) suggest
that some N from both fertilizer sources has leached to the
120-cm depth in the fertilized plots. Soil water contents
(Table 12) and evapotranspiration data (Table 8) also tend
to support this. Since the movement of N, particularly
NO3-N, in uncropped soils is closely related to water move-
ment in the profile (106), it is likely that some N leached
past the 120-cm depth. In the mature tree plots, there was
evidence that some N had moved at least to the 75-cm
depth when NH4NO3 was the fertilizer applied (Fig. 13). It
is difficult to trace the movement of N in the soil pro-
file in the presence of a tree, however, because of root
I lI lo
MO e 0
== ) 0 Io
." I = I
o c z
4 4() 411-;
L 0 0 4
S/ j 4
ls- il l ^
E Ic Za$
C "I /i
M LU O0w -i
.l. ( | rC
o 0 4)
0 0 00
.* O i
C> a. 4J
) 0O 4 l
0 ow 7 a)
.I I I
0 0 0 0
.M AUa % 'N "110
.M AdQ % 'N I OS
r-iWCmO m WLcO v
v r- v w v CN
mZ L Iz L1fll
v (n N ffo
C)00r a) k. r- a%
LO OO r- (n m 0 (n Or
aoDo m ConD
4 O (
*i (f *fd U
r- a r-ofr-N
lI I I I l I I
% ( C O 00 r, 00
%.0qv LO )n 0 r-4o
r-!I n rL %.0U'N
00CNi-f-4r-fr L 00
r-4 rn 10 110 r m> Co
*.-H I -4
0 C =Z
o 'N IIOS
As in Experiments 1 and 2, soil total N content gener-
ally decreased with time in the tree plots partly because
of tree absorption of N. At the last sampling, almost all
of the applied NH4NO3-N in the fallow and tree plots had
dissipated from the 0- to 15-cm depth, but considerable
IBDU-N remained (Figs. 14 and 15).
Recovery of Fertilizer-N in the Soil, Fruit and Leaf
The recovery of applied N in the soil, fruit and leaf
was calculated by the difference procedure used in both
Experiments 1 and 2. Compared to the young tree experi-
ments, however, less confidence can be expected in the N
recovery data for Experiment 3. The absence of replica-
tions in the mature tree experiment precluded the estab-
lishment of plausible conclusions. Furthermore, since the
trees had been fertilized for several years it is diffi-
cult to separate the effect of previous fertilization on
N contents of tree parts and the soil. The use of an
empirical procedure in the estimate of leaf total N and
the large sampling and analytical errors associated with
the soil total N procedure are additional factors which
reduce the confidence in the N recovery data.
Nevertheless, the N recovery data (Table 13) may serve
a useful purpose in stimulating additional research. The
proportion of applied N recovered in the 0- to 120-cm soil
depth of the IBDU- and NH4NO3-fertilized mature tree plots
was 44 and 30%, respectively. The fruits accounted for
27 and 30% of applied N, and the leaves, 11 and 10%
IM ),cJ % N
4w 0 -7
W y. U o
C= C r Q4
A 0 4) I
(O Z m
"-- 3 r- V," .4
- ,-, Z l
* ON O* -
0l 0NI -
> I A I
,' -, o,
S % N lOS
S~rr *n ? 0 g
E / /0 0 ^
/ ro 1'-
Apparent applied N recovery (%) in mature tree
Treatment Fruits Leaves Soil Total
IBDU-fertilized 27.60 10.77 44.00 82.77
NH4NO3-fertilized 30.08 9.85 29.55 69.48
ZCalculated from the difference in N content, for each N
balance component, between the control and fertilized tree
plots and expressed as the % of the amount of N applied.
for the same respective sources. The apparent total re-
covery of applied N when fruits, leaves and the soil are
considered amounted to 83% for the IBDU treatment, and 69%
for the NH4NO3 treatment.
It is suggested that the deficits in the N balance
could be partly accounted for in the permanent structures
of the tree. Smith (100) estimated that 31% of the total
N was found in the leaves of 15-year-old 'Valencia' orange
trees in Florida. On this basis, control, IBDU- and NH4NO3-
treated trees in this experiment would contain 402, 648
and 627 g of N, respectively. It was also indicated that
42% of the total N of the tree was found in parts other
than leaves and fruit (100). On the basis of this assump-
tion 13 and 12% of the applied N would be expected to be in
tree parts other than fruits and leaves for IBDU- and NH4NO3
treated trees, respectively. When these figures are added
to the measured recovery of applied N, 95 and 81% of ap-
plied N from IBDU and NH4NO3 sources, respectively, could
be accounted for in the soil-plant system. No account
could be made for the remaining 5-19%.
These apparently high recoveries of applied N may in-
dicate that the trees had absorbed some N already present
in the soil in addition to that applied in this study.
Also, substantial leaching of N may not have been signifi-
cant. Although the potential for the leaching of N exis-
ted (Table 8), the large root systems of the mature trees
may have absorbed a greater portion of the applied N as
it moved in the soil profile.
In view of the low nutrient retaining capacity of
Florida soils planted to citrus, it has often been sugges-
ted that the efficiency of N utilization by citrus is very
low (25-30%). In the current study, however, higher re-
coveries for both young (45%) and bearing (40%) trees are
reported when NH4NO3 was the N source. These high recover-
ies imply that citrus trees may be absorbing a greater pro-
portion of fertilizer-N than previously suspected. The
unreplicated nature and short duration of the mature tree
experiment, however, call for additional adequately repli-
cated experiments. Moreover, it would be beneficial to in-
vestigate the N balance of young citrus trees over a period
longer than the ones used in this study.
1. Growth parameters of young citrus trees in short-term
experiments, trunk diameter increase, and dry weights
of tree component parts and the entire tree, showed no
statistical differences due to N sources. There was
a trend, however, for N from IBDU and NH4NO3 sources to
result in greater dry matter, particularly in the more
succulent aerial parts, when compared to the control.
Leaves from fertilized trees in both fall-winter and
spring studies were statistically greater in total dry
weight than those from the unfertilized leaves. The
apparent lack of statistical differences in growth
parameters due to N source was probably related to the
short duration of the studies. The trees did not have
sufficient time to respond to the treatments.
2. The dry weights of certain tree components as a percent of
the total plant were statistically different due to N
source, but it is doubtful if they were true differ-
ences because the pattern of dry matter distribution
paralleled that of the dry matter. These differences
were probably the result of selective use of different
size trees for each N source.
3. There was a highly significant relationship between N
source and N concentration and total N of component
tree parts, and total N content of the entire tree.
In every case, the order was NH4NO3- > IBDU-fertilized
> control trees. This confirms that a soluble N source
such as NH4NO3 is more readily absorbed by the trees
in the short-term than a controlled release form of N.
Other investigations, however, suggested that the ab-
sorption of N from both fertilizer sources might be
different in longer periods.
4. Regardless of N source, 30-34% of the total tree N was
distributed in fibrous and lateral roots, and 30-33% in
the leaves. This illustrates the importance of recover-
ing the entire root system as much as practical in
studies of this nature.
5. A greater proportion of the total N in the fertilized
tree was distributed in the aerial parts in the spring
study than in the fall-winter study, indicating that
N absorption and translocation to the aerial parts was
greater in spring.
6. Data related to tree growth and N content parameters,
and soil total N content did not reveal any statistical
differences due to irrigation level in the fall-winter
study. The narrow range and short duration of irriga-
tion treatments were probably not sufficient to elicit
responses to the treatments.
7. Nitrogen balance data, based on the 2828 cm2 plot area,
showed that total apparent N recovery in the soil-
plant system was 91 and 76% for IBDU- and NH4NO3-N,
respectively. Sixty-six and 30% of applied N from
IBDU and NH4NO3 sources, respectively, were retained
in the 0- to 60-cm soil profile. Twenty-five and 45%
from the same respective sources could be accounted for
in the young trees. The remaining 9-24% was presumed
to have been leached or lost to the atmosphere.
8. In the bearing tree study, no difference was detected
in fruit yield between IBDU and NH4NO3 sources. Fer-
tilized trees, however, had about 4 times as much
yield as the unfertilized tree.
9. The relatively high recovery of applied N in young and
bearing trees suggest that substantial leaching of N
may not be a major route for N loss from citrus groves
established on deep-well drained soils when the ferti-
lizer is applied over the root zone at reasonable
10. A major portion of IBDU-N was retained in the 0- to
15-cm depth, yet the growth response of the IBDU-
fertilized young trees was similar to that of NH4NO3-
fertilized trees. There is, therefore, a potential for
a single application of IBDU to nonbearing young citrus
trees which are normally fertilized several times a
order to establish if a single application of IBDU is
feasible from the standpoint of economy and horticul-
1. Adriano, D.C., F.H. Takatori, P.F. Pratt, and O.A.
Lorenz. 1972. Soil N balance in selected row-
crop sites in Southern California. J. Environ.
2. Allison, F.E. 1955. The enigma of soil nitrogen bal-
ance sheets. Adv. Agron. 7:214-250.
3. Avnimelech, Y., and J. Raveh. 1976. Nitrate leakage
from soils differing in texture and nitrogen
load. J. Environ. Qual. 5:79-81.
4. Banwart, W.L., M.A. Tabatabai, and J.M. Bremner. 1972.
Determination of ammonium in soil extracts and
water samples by an ammonia electrode. Comm. Soil
Sci. Plant Anal. 3:449-458.
5. Barnette, R.M., E.F. Debusk, J.B. Hester, and H.W.
Jones. 1931. The mineral analysis of a nineteen-
year-old 'Marsh' seedless grapefruit. Citrus
Indus. 12(3):5-6, 34.
6. Bartholomew, W.V., L.B. Nelson, and C.H. Werkman.
1950. The use of nitrogen isotope N-15 in field
studies with oats. Agron. J. 42:100-103..
7. Biggar, J.W. 1978. Spatial variability of N in soils.
p. 201-211. In D.R. Neilsen, and J.G. Macdonald
(eds.). Nitrogen in the environment. Vol. I.
Nitrogen behavior in field soil. Acad. Press,
N.Y., San Francisco, Lond.
Bingham, F.T., S. Davis, and E. Shade. 1971. Water
relations, salt balance, and nitrate leaching
losses of a 960-acre citrus watershed. Soil
9. Bjorkman, E., G. Lundeberg, and H. Nommik. 1967.
Distribution and balance of N-15 labelled ferti-
lizer nitrogen applied to young pine trees
(Pinus sylvestris L.). Studia Forestalia Seu-
10. Bower, C.A., and L.V. Wilcox. 1969. Nitrate content
of the Upper Rio Grande as influenced by nitrogen
fertilization of adjacent irrigated lands. Proc.
Soil Sci. Soc. Amer. 33:971-973.
11. Brams, E.A., and J.G.A. Fiskell. 1967. Variability
of root and soil analyses of a 'Valencia' grove
sampled in January. Proc. Fla. State Hort. Soc.
12. Bremner, J.M. 1965. Total nitrogen. p. 1149-1178.
In C.A. Black (ed.). Methods of soil analysis.
Part 2. Agronomy 9. Amer. Soc. Agron., Madison,
13. 1965. Inorganic forms of nitrogen. p.
1179-1237. In C.A. Black (ed.). Methods of
soil analysis. Part 2. Agronomy 9. Amer. Soc.
Agron., Inc., Madison, Wisc.
14. Cahoon, G.A., and L.H. Stolzy. 1959. Estimating root
density in citrus orchards by the neutron modera-
tion method. Proc. Amer. Soc. Hort. Sci. 74:
15. 1960. Citrus root density in relation
to water use. Calif. Citrog. 45:318, 323.
16. Calvert, D.V. 1975. Nitrate, phosphate, and potas-
sium movement into drainage lines under 3 soil
management practices. J. Environ. Qual. 4:
17. and H.J. Reitz. 1964. Effects of rate
and frequency of fertilizer applications on
yield and quality of 'Valencia' oranges in the
Indian River area. Proc. Fla. State Hort. Soc.
18. and H.T. Phung. 1971. Nitrate-nitrogen
movement into drainage lines under different
soil management systems. Proc. Soil and Crop
Sci. Soc. Fla. 31:229-232.
19. Cameron, S.H., and O.C. Compton. 1945. Nitrogen in
bearing orange trees. Proc. Amer. Soc. Hort.
20. and D. Appleman. 1933. The distribution
of total nitrogen in the orange tree. Proc.
Amer. Soc. Hort. Sci. 30:341-348.
21. Carlton, A.B., and A.A.R. Hafez. 1973. N-14 nitrogen
fertilizer tracing experiments in San Joaquin
Valley. Calif. Agr. 27(11):10-13.
22. Chapman, H.D. 1968. Mineral nutrition of citrus, p.
127-289. In W. Reuther, L.D. Batchelor, and H.J.
Webber (eds.). Citrus industry, Rev. ed..II.
Univ. Calif., Div. Agr. Sci., Berkeley, Calif.
23. Colman, E.A. 1946. A laboratory study of lysimeter
drainage under controlled soil moisture tension.
Soil Sci. 62:365-382.
24. Cooper, D.R., D.G. Hill-Cottingham, and C.P. Lloyd-
Jones. 1976. Distribution and identity of
labelled products following autumn application
of N-15-labelled urea or potassium nitrate fer-
tilizers to apple trees. J. Sci. Food Agr.
25. Cooper, W.C., A Peynado, J.R. Furr., R.H. Hilgeman,
G.A. Cahoon, and S.B. Boswell. 1963. Tree
growth and fruit quality of 'Valencia' oranges
in relation to climate. Proc. Amer. Soc. Hort.
26. DeVitt, D., J. Letey, L.J. Lund, and J.W. Blair. 1976.
Nitrate-nitrogen movement through soil as affected
by soil profile characteristics. J. Environ.
27. Duke, H.R., and H.R. Haise. 1973. Vacuum extractors
to assess deep percolation losses and chemical
constituents of water. Proc. Soil Sci. Soc.
28. Edwards, P.P., and R.D. Hauck. 1974. Nitrogen-15-
depleted versus nitrogen-15-enriched ammonium sul-
fate as tracers in nitrogen uptake studies.
Proc. Soil Sci. Soc. Amer. 38:765-767.
29. Embleton, T.W., and W.W. Jones. 1978. Nitrogen fer-
tilizer management programs, nitrate-pollution
potential, and orange productivity. p. 275-
295. In D.R. Nielsen and J.G. MacDonald (eds.).
Nitrogen in the environment. Vol. I. Nitrogen
behavior in field soil. Acad. Press., N.Y.,
San Francisco, Lond.
30. 1974. Foliar-applied nitrogen for citrus.
J. Environ. Qual. 3:338-391.