Bulletin 461
June, 1949
UNIVERSITY OF FLORIDA
AGRICULTURAL EXPERIMENT STATIONS
HAROLD MOWRY, Director
GAINESVILLE, FLORIDA
Fertilizer Experiments with
Citrus on Davie Mucky Fine Sand
By T. W. YOUNG and W. T. FORSEE, JR.
Fig 1.-A typical orange grove in the Davie area.
BOARD OF CONTROL
J. Thos. Gurney, Chairman, Orlando
N. B. Jordan, Quincy
Thos. W. Bryant, Lakeland
J. Henson Markham, Jacksonville
Hollis Rinehart, Miami
W. F. Powers, Secretary, Tallahassee
EXECUTIVE STAFF
J. Hillis Miller, Ph.D., President of the
University:'
H. Harold Hume, D.Sc., Provost for Agr.s
Harold Mowry, M.S.A., Director
L. O. Gratz, Ph.D., Asst. Dir., Research
W. M. Fifield, M.S., Asst. Dir., Admin.
J. Francis Cooper, M.S.A., Editur3
Clyde Beale, A.B.J., Associate Editor3
WV. W. Mosher, Assistant Editor
Ida Keeling Cresap, Librarian
Ruby Newhall, Administrative Manager3
Geo. F. Baughman, M.A., Business Manager'
Claranelle Alderman, Accountant3
MAIN STATION, GAINESVILLE
AGRICULTURAL ENGINEERING
Frazier Rogers, M.S.A., ApEr. Engineer3
J. M. Johnson, B.S.A.E., Asso. Agr. Engineer5
J. M. Myers, B.S., Asso. Agr. Engineer
R. E. Choate, B.S.A.E., Asst. Agr. Engineer3
A. M. Pettis, B.S.A.E., Asst. Agr. Engineer2 a
AGRONOMY
Fred H. Hull, Ph.D., Agronomisti
G. E. Ritchey, M.S., Agronomist2
G. B. Killi:iwer, Ph.D., Agronomist3
H. C. Harris, Ph.D., Agronomist:
R. W. Bledsoe, Ph.D., Agronomist
S. C. Litzenberger, Ph.D., Associate
W. A. Carver, Ph.D., Associate
Fred A. Clark, B.S., Assistant
M. N. Gist, Collaborator'
ANIMAL INDUSTRY
A. L. Shealy, D.V.M., An. Industrialist1 1
R. B. Becker, Ph.D.. Dairy Husbandman3
E. L. Fouts, Ph.D., Dairy Technologist3
D. A. Sanders, D.V.M., Veterinarian
M. W. Emmel, D.V.M. Veterinarian3
L. E. Swanson, D.V.M., Parasitologist
N. R. Mehrhof, M.Agr., Poultry Hush."
G. K. Davis, Ph.D., Animal Nutritionist3
R. S. Glasscock, Ph.D., An. Husbandmana
P. T. Dix Arnold, M.S.A., Asst. Dairy Husb.'
L. E. Mull, M.S., Asst. in Dairy Tech.
Katherine Boney, B.S., Asst. Chem.
J. C. Drigfers, Ph.D., Asst. Poultry Hushb.a
Glenn Van Ness, D.V.M., Asso. Poultry
Pathologist
S John Folks, B.S.A., Asst. An. Husb.3
W A. Krienke, M.S., Asso. in Dairy Mfs."
S P Marshall, Ph.D., Asso. Dairy Husb.'
I Simpson, D.V.M., Asso. Veterinarian
C. r. Winchester, Ph.D., Asso. Biochemista
ECONOMICS, AGRICULTURAL
C. V. Noble, Ph.D., Agri. Economist a
R. E. L. Greene, Ph.D., Agri. Economist
Zach Savage, M.S.A., Associate
A. H. Spurlock, M.S.A., Associate
D. E. Alleger, M.S., Associate
D. L. Brooke, M.S.A., Associate
M. I. Godwin, Ph.D., Associate
H. W. Little, M.S., Assistant
Tallmadge Bergen, B.S., Asst.
Orlando, Florida (Cooperative USDA)
G. Norman Rose, B.S., Asso. Agr. Economist
J. C. Townsend, Jr., B.S.A., Agr. Statistician2
J. B. Owens, B.S.A., Agr. Statistician2
J. F. Steffens, Jr., B.S.A., Agr. Statistician2
ECONOMICS, HOME
Ouida D. Abbott, Ph.D., Home Econ.'
R. B. French, Ph.D., Biochemist
ENTOMOLOGY
A. N. Tissot, Ph.D., Entomologist'
L. C. Kuitert, Ph.D., Assistant
H. E. Bratley, M.S.A., Assistant
HORTICULTURE
G. H. Blackmon, M.S.A., Horticulturist'
F. S. Jamison, Ph.D., Horticulturist'
H. AM. Reed, B.S., Chem., Veg. Processing
Byron E. Janes, Ph.D., Asso. Hort.
t. A. Dennison, Ph.D., Asso. Hort.
R. K. Showalter, M.S., Asso. Hort.
Albert P. Lorz, Ph.D., Asso. Hort.
R. H. Sharpe, M.S., Asso. Hort.
R. J. Wilmot, M.S.A., Asst. Hort.
R. D. Dickey, M.S.A., Asst. Hort.
Victor F. Nettles, M.S.A., Asst. Hort.'
F. S. Lagasse, Ph.D., Asso. Hort."
L. H. Halsey, B.S.A., Asst. Hort.
Forrest E. Myers, B.S.A., Asst. Hort.
PLANT PATHOLOGY
W. B. Tisdale, Ph.D., Plant Pathologist'
Phares Decker, Ph.D., Plant Pathologist
Erdman West, M.S., Mycologist and Botanist
Howard N. Miller, Ph.D., Asso. Plant Path.
Lillian E. Arnold, M.S.. Asst. Botanist
SOILS
F. B. Smith, Ph.D., Microbiologistt'
Gaylord M. Volk, Ph.D., Chemist
J. R. Hendorson, M.S.A., Soil Technologist'
J. R. Neller, Ph.D., Soils Chemist
Nathan Gammon, Jr., Ph.D., Soils Chemist
C. E. Bell, Ph.D., Associate Chemist
R. A. Carrigan, Ph.D., Asso. Biochemist3
H. W. Winsor, B.S.A., Assistant Chemist
Geo. D. Thornton, Ph.D., Asso. Microbiologista
R. E. Caldwell, M.S.A., Asst. Chemist3
J. B. Cromartie, B.S.A., Soil Surveyor
Ralph G. Leighty, B.S., Asso. Soil Surveyor
V. W. Cyzycki, B.S., Asst. Soil Surveyor
R. B. Forbes, M.S., Asst. Soils Chemist
W. L. Pritchett, M.S., Asst. Chemist
Jean Bcem, B.S.A., Asst. Soil Surveyor
Head of Department.
In cooperation with U. S.
SCooperative, other divisions, U. of F.
On leave.
BRANCH STATIONS
NORTH FLORIDA STATION, QUINCY
J. D. Warner, M.S., Vice-Director in Charge
R. R. Kincaid, Ph.D., Plant Pathologist
W. H. Chapman, M.S., Asso. Agron.
L. G. Thompson, Ph.D., Soils Chemist
Frink S. Bakr.,r. ,r., B.S.. \A-. An. Hu,b.
W. C. Rhoads, M.S., Entomologit
Mobile Unit, Monticello
R. W. Wallace, B.S., Associate Agronomist
Mobile Unit, Marianna
R. W. Lipscomb, M.S., Associate Agronomist
Mobile Unit, Chipley
J. B. White, B.S.A., Associate Agronomist
Mobile Unit, DeFuniak Springs
R. L. Smith, M.S., Associate Agronomist
CITRUS STATION, LAKE ALFRED
A. F. Camp, Ph.D., Vice-Director in Charge
W. L. Thompson, B.S., Entomologist
J. T. Griffiths, Ph.D., Asso. Entomologist
R. F. Suit, Ph.D., Plant Pathologist
E. P. Ducharme, M.S., Plant Pathologist4
R. K. Voorhees, Ph.D., Asso. Horticulturist
C. R. Stearns, Jr., B.S.A., Asso. Chemist
J. W. Sites, M.S.A., Horticulturist
H. O. Sterling, B.S., Asst. Horticulturist
J. A. Granger, B.S.A., Asst. Horticulturist
H. J. Reitz, M.S., Asso. Horticulturist
Francine Fisher, M.S., Asst. Plant Path.
I. W. Wander, Ph.D., Soils Chemist
A. E. Willson, B.S.A., Asso. Biochemist
J. W. Kesterson, M.S., Asso. Chemist
R. N. Hendrickson, B.S., Asst. Chemist
Joe P. Barnett, B.S.A., Asst. Horticulturist
J. C. Bowers, B.S., Asst. Chemist
D. S. Prosser, Jr., B.S., Asst. Horticulturist
R. W. Olsen, B.S., Biochemist
F. W. Wenzel, Jr., Ph.D., Superviskry Chem.
Alvin H. Rouse, M.S., Asso. Chemist
L. W. Fayville, Ph.D., Asst. Chemist
EVERGLADES STATION, BELLE GLADE
R. V. Allison, Ph.D., Vice-Director in Charge
F. D. Stevens, B.S., Sugarcane Agronomist
Thomas Bregger, Ph.D., Sugarcane
Physiologist
J. W. Randolph, M.S., Agricultural Engineer
W. T. 'orsee, Jr., Ph.D., Chemist
R. W. Kidder, M.S., Asso. Animal Hu b.
T. C. Erwin, Assistant Chemist
Roy A. Bair, Ph.D., Agronomist
C. C. Seale, Asso. Agronomist
N. C. Hayslip, B.S.A., Asso. Entomologist
E. H. Wolf, Ph.D., Asst. Horticulturist
W. H. Thames, M.S., Asst. Entomologist
C. B. Savage, M.S.A., Asst. Horticulturist
D. L. Stoddard, Ph.D., Asso. Plant Path.
W. N. Stoner, Ph.D., Asst. Plant Path.
W. H. Hills. M.S., A, o. Horticultlioi M
WV. Genunr, B.S.A., A-t. Entomolouict
SUB-TROPICAL STATION, HOMESTEAD
Geo. D. Ruehle, Ph.D., Vice-Dir. in Charge
D. 0. Wolfenbarger, Ph.D., Entomologist
Francis B. Lincoln, Ph.D., Horticulturist
Robt. A. Conover, Ph.D., Asso. Plant Path.
R. W. Harkness, Ph.D., Asst. Chemist
Milton Cobin, B.S., Asso. Horticulturist
W. CENT. FLA. STATION, BROOKSVILLE
William Jackson, B.S.A., Animal Husband-
man in Charge2
RANGE CATTLE STATION, ONA
W. G. Kirk, Ph.D., Vice-Director in Charge
E. M. Hodges, Ph.D., Associate Agronomist
D. W. Jones, B.S., Asst. Soil Technologist
H. E. Henderson. M.S.. A t. An. Hush.
CENTRAL FLORIDA STATION, SANFORD
R. W. Ruprecht, Ph.D., Vice-Dir. in Charge
J. W. Wilson, Sc.D., Entomologist
Ben F. Whitner, Jr., B.S.A., Asst. Hort.
WEST FLORIDA STATION, MILTON
C. E. Hutton. Ph.D. Agronomi-ti
H. W. Lundy, B.S.A., Associate Agronomist
FIELD STATIONS
Leesburg
G. K. Parris, Ph.D., Plant Path. in Charge
Plant City
A. N. Brooks, Ph.D., Plant Pathologist
Hastings
A. H. Eddins, Ph.D., Plant Path. in Charge
E. N. McCubbin, Ph.D., Horticulturist
Monticello
A. IM. Phillips, B.S., Asso. Entomologist2
John R. Large, M.S., Asso. Plant Path.
Bradenton
J. R. Beckenbach, Ph.D., Hort. in Charge
E. G. Kelsheimer, Ph.D., Entomologist
David G. Kelbert, Asso. Horticulturist
E. L. Spencer, Ph.D., Soils Chemist
Robert 0. Magie, Ph.D., Gladioli Hort.
J. M. Walter, Ph.D., Plant Pathologist
Donald S. Burgis, M.S.A., Asst. Hort.
Lakeland
Warren O. Johnson, B.S., Meteorologist2
SHead of Department.
SIn cooperation with U. S.
3 Cooperative, other divisions, U. of F.
SOn leave.
Contents
Page
INTRODUCTION ...... .... .... .. ........ ...... ......5...-- 5
PURPOSE OF THE EXPERIMENTS ......... .............. ....... ....... 7
EXPERIMENTAL GROVE ..... .... .... ...... ..... ...........--- 7
EXPERIMENTAL DESIGN .... .. . ...........-- ...-- ...... -. 11
FERTILIZER TREATMENTS ... ........ ..... .... ..... ........ .......... 11
R E SU LTS 1.... .... .. .......... ......... ......... .......... ................... 15
DISCUSSION OF RESULTS ........ ....... ... ........ -......- 18
N itrogen -. ... .. ... ........ --.. ... ..... 18
Phosphate ........................... .................... ......- ... -- 26
P otash ..... .. ... ..... ... ................. .. 34
GENERAL DISCUSSION ... .................... 37
CONCLUSIONS ............. ........... .. .. ......----------------- .---------------- 39
ACKNOWLEDGMENTS -.. ...... ..........................---- -.--------.-----.. 39
LITERATURE CITED ............................. .....- .............. ....... ... --- 39
Fertilizer Experiments with Citrus
On Davie Mucky Fine Sand
By T. W. YOUNG and W. T. FORSEE, JR.
Introduction
The Davie citrus area, in which the experiments reported
here were located, lies on the eastern edge of the Florida Ever-
glades about 10 miles southwest of Ft. Lauderdale. This area
extends roughly from State Road 7 westward about eight miles
and southward from the North New River Canal to about a
mile beyond the South New River Canal. In this area of ap-
proximately 25 square miles, about 10,000 acres have been
planted to citrus. Practically all the commercial plantings in
existence in the region at present have been made since 1926.
The great majority of these plantings are of two late oranges,
Valencia and Lue Gim Gong varieties.
The topography is essentially flat. Average elevation above
sea level is about 10 feet. There are a few sand ridges in the
area with elevations somewhat higher.
Drainage is primarily by gravity through a series of lateral
canals and two main outlet canals, the North and South New
River Canals, which empty into the Atlantic Ocean. Each of
the outlet canals has locks about eight miles inland to help regu-
late the flow. The water level in the North New River Canal
above the locks is normally maintained about three feet higher
than that in the South New River Canal above the locks. By
regulating the canal levels in this manner the water table in
most of the area is held at an approximate average depth of
three feet. It is the general opinion of the majority of growers
in the area that this arrangement provides ample irrigation and
allows for adequate drainage. However, at least in some cases
the groves occasionally suffer from drought. Most of them have
from time to time suffered more or less from water-logged soils
or flooding during periods of excessive rains. A few growers
have consequently installed pumps on their groves to supple-
ment the drainage and irrigation as provided by the water-
control system for the area in general. Drainage has been
accomplished with a certain amount of difficulty because of the
porous limerock underlying the area.
The surface layer of the soil in most of the area originally
Florida Agricultural Experiment Station
consisted of a fibrous sawgrass (Everglades) peat from 18 to 24
inches deep. This soil is locally known as "muck". In the
process of planting citrus in most groves the soil was plowed
and scraped into broad low ridges or beds on which the trees
were planted for better drainage. With improved drainage and
aeration there was a shrinkage of the surface peaty layer due
to loss of water and increased oxidation of the organic matter.
This caused a subsidence of the surface layer, which was further
hastened by the compacting effect of the heavy equipment used.
Usually a considerable amount of sandy or marl subsoil was
mixed with the organic surface layer in making the beds. As
a result of all these processes the mineral matter content of
the surface layer has increased, the organic matter content de-
creased and depth of the surface layer decreased. Over most
of the area the subsoil consists of gray to light gray sands
which overlie marl or soft porous limerock at a depth of a few
inches to about five feet or more. In some places, however,
the organic surface layer rests directly on the marl or limerock.
Occasionally spots are encountered where the marl or limerock
comes to the surface, as does the sand in others, particularly
on the sand ridges.
With a few minor exceptions, the soil on which citrus has
been planted in the Davie area is classified as Davie mucky fine
sand. Some idea of the chemical composition of the Davie soils,
after a number of years of drainage but before being worked
and planted to citrus or other crops, may be gained by an
examination of the data in Table 1. These data are from chemi-
cal analyses made on soil samples from two virgin tracts in the
vicinity of the experimental grove described later.
Observations by some of the first growers around Davie on
the growth and performance of the earlier citrus plantings led
them to conclude that commercial fertilizers were not necessary
to successful citrus culture on this organic soil. Within a few
years, however, some indications were found that a favorable
response apparently was obtained on older trees, at least in some
cases, to applications of certain commercial fertilizer elements
(1).1 Up to that time, no systematic experimental data had
been collected relative to the fertilizer requirements of citrus
growing on these soils. As the acreage planted to citrus in the
Davie area increased the need for such information became more
evident and pressing. In response to this need for nutritional
'Italic figures in parentheses refer to Literature Cited.
Fertilizer Experiments with Citrus
investigations, the Everglades Experiment Station, in cooper-
ation with the Florida Agricultural Research Institute and
Flamingo Groves, Incorporated, initiated in March, 1934, the
experiments described and discussed in this bulletin. These
experiments were turned over to the Citrus Experiment Station
for handling in 1944.
Purpose of the Experiments
Organic soils, as shown for the Davie series in Table 1, are
inherently high in nitrogen and low in phosphates and potash,
particularly the latter. With most crops on such soils the use
of phosphate and potash fertilizers is axiomatic for successful
culture (16).
The primary object of these experiments was to determine
for the Davie soils by a field plot technique the relative response
of citrus to phosphate fertilizers when applied in various amounts
and from various sources. Treatments were also included to
test the response to various rates of application of potash
fertilizers. While nitrogen is not generally at a limiting low
level in soils of this nature, some attention was given in these
tests to nitrogen fertilization at different levels.
The Experimental Grove
The site selected for these experiments was one of the Fla-
mingo Groves, Incorporated, properties located in the western
portion of the Davie area. It was planted to Lue Gim Gong
oranges in 1929 and was five years old when the plots were
established. The soil, a Davie mucky fine sand, appeared to be
rather typical for the area. The dark organic surface layer
was about 12 to 16 inches deep on the beds and had a reaction
ranging from about pH 4.8 to 5.5. According to analyses made
just prior to beginning the experiments, it contained about 57
percent organic matter and was well supplied with calcium and
magnesium. The surface layer was underlain by a light gray
sand to a depth of two or three feet. The pH of the subsoil,
which rested on limerock, was somewhat higher than that of
the surface layer. In 1942, after the experiments had been in
progress eight years and the trees were 13 years old, more com-
prehensive chemical analyses were made on soil from the (con-
trol) plots receiving the fertilizer treatment most comparable
to that in general commercial use in the area at that time.
Fertilizer Experiments with Citrus
investigations, the Everglades Experiment Station, in cooper-
ation with the Florida Agricultural Research Institute and
Flamingo Groves, Incorporated, initiated in March, 1934, the
experiments described and discussed in this bulletin. These
experiments were turned over to the Citrus Experiment Station
for handling in 1944.
Purpose of the Experiments
Organic soils, as shown for the Davie series in Table 1, are
inherently high in nitrogen and low in phosphates and potash,
particularly the latter. With most crops on such soils the use
of phosphate and potash fertilizers is axiomatic for successful
culture (16).
The primary object of these experiments was to determine
for the Davie soils by a field plot technique the relative response
of citrus to phosphate fertilizers when applied in various amounts
and from various sources. Treatments were also included to
test the response to various rates of application of potash
fertilizers. While nitrogen is not generally at a limiting low
level in soils of this nature, some attention was given in these
tests to nitrogen fertilization at different levels.
The Experimental Grove
The site selected for these experiments was one of the Fla-
mingo Groves, Incorporated, properties located in the western
portion of the Davie area. It was planted to Lue Gim Gong
oranges in 1929 and was five years old when the plots were
established. The soil, a Davie mucky fine sand, appeared to be
rather typical for the area. The dark organic surface layer
was about 12 to 16 inches deep on the beds and had a reaction
ranging from about pH 4.8 to 5.5. According to analyses made
just prior to beginning the experiments, it contained about 57
percent organic matter and was well supplied with calcium and
magnesium. The surface layer was underlain by a light gray
sand to a depth of two or three feet. The pH of the subsoil,
which rested on limerock, was somewhat higher than that of
the surface layer. In 1942, after the experiments had been in
progress eight years and the trees were 13 years old, more com-
prehensive chemical analyses were made on soil from the (con-
trol) plots receiving the fertilizer treatment most comparable
to that in general commercial use in the area at that time.
TABLE 1.-CHEMICAL COMPOSITION OF Two VIRGIN* DAVIE SOILS.
Depth of Exchange Pounds per Acre-Six-Inches of Soil
Tract Profile Capacity Total Organic
No. Layer in pH m.e./ Exchangeable Bases P N Matter
Inches 100 Gms. I Acid- Water- % %
Ca Mg. K Soluble Soluble
1 0-10 5.30 30 3,100 223 Tr. 14 3 1.65 30
10-24 5.78 3 1,157 35 Tr. Tr. Tr. 0.04 1
2 0-10 4.20 73 2,039 21 27 11 1 2.62 67
Subject to drainage since about 1927 but never cultivated.
TABLE 2.-CHEMICAL ANALYSES OF SOIL FROM PLOTS RECEIVING A 3-6-12 FERTILIZER *
Exchange Pounds per Acre-Six-Inches of Soil
Capacity Total Organic
Profile pH m.e./ Exchangeable Bases P N Matter
100 Gms. Acid- Water- % %
Ca Mg K Soluble Soluble
Surface .... 4.62 27.22 1,953 128 100 45 12.8 1.270 42.81 "
Subsoil ........ 5.44 1.64 507 17 14 3 1.4 0.032 0.83
Samples collected in 1942.
TABLE 3.-MINIMUM, MAXIMUM AND AVERAGE AMOUNTS OF VARIOUS CHEMICAL CONSTITUENTS FOUND IN DAVIE MUCKY
FINE SANDY SOIL FROM 26 CITRUS GROVES NEAR DAVIE, FLORIDA.*
Pounds per Acre-Six-Inches of Soil |
Range Exchange Total Organic
Profile of pH Capacity Exchangeable Bases P N Matter
Values m.e./100 G. Acid- Water- | % %
Ca Mg K Soluble Soluble
Surface Minimum 4.00 11.88 866 71 0 3 0 0.007 4.72
Maximum 7.55 100.80 11,235 660 302 299 39.6 2.650 88.64
Average** 5.22 39.46 3,868 225 98 48 8.5 1.574 49.81
Subsoil Minimum 5.22 0.64 264 2 0 1 0 0.009 0.20
Maximum 8.12 29.60 6,676 652 160 35 10.6 0.272 14.83
Average** 6.82 4.63 1,474 72 20 13 2.1 0.087 2.64
*Samples collected 1912 to ll9;.
** Average of the 26 samples.
Florida Agricultural Experiment Station
The results of these later analyses indicated still further that
the chemical composition of the soil in the plots was fairly rep-
resentative of that from most commercial groves of similar age
in the area. This is made more evident by a comparison of the
data in Table 2, which gives the results of the analyses of soil
from the control plots (a 3-6-12 treatment), with those in Table
3 which represents the entire area, being analyses from 26
representative grove soils taken over the entire area. It is seen
that the organic matter content of the surface soil of the plots,
which was about 57 percent when the experiments were started
eight years previously, had decreased, evidently through oxida-
tion, to about 43 percent. This was slightly lower than the
average for the area, but is to be expected when one considers
that soils from a number of younger groves, in which oxidation
would not yet have been so great a factor, were included in the
analyses for Table 3. In this connection it should be mentioned
also that the depth of the dark organic surface layer decreased
about three inches during this period. The pH (4.6 in surface
layer) also was slightly lower in 1942 than when the plots were
established. The difference was so slight that it may well have
been due to sampling error or natural seasonal variation. Al-
though no liming material was applied to the plots during the
course of the experiments, the close proximity to limerock would
probably tend at least partly to neutralize any excess acidity
that might develop.
A topographical survey showed practically no natural slope
to the surface of the experimental plot area. Before planting,
the soil had been worked into beds to improve the drainage
within the grove. These beds were about 12 to 18 inches high
and on 30-foot centers. A single row of trees was planted on
each bed crown. The trees were spaced 20 feet in the row. The
"furrrows" thus provided between the tree rows led into a ditch
on the west end of the grove. The distance from the most remote
plot to this ditch was about 900 feet. The west ditch connected
with a ditch adjoining the south side of the grove. The most
remote plot was about 300 feet from the south ditch, which
emptied into a lateral canal about 400 feet east of the southeast
corner of the experimental plots. Fair surface and sub-drainage
was secured through this system of furrows, ditches and canals.
There was some evidence, however, of water damage to scat-
tered trees in the plots by the time the grove was about 12 or
13 years old. Commercial groves of the same age and variety
Fertilizer Experiments with Citrus
adjoined the plots on the east and north. An "Australian pine"
(Casuarina lepidophloia) windbreak running north and south
divided the experimental site. The layout of the experimental
grove is shown in Fig 2.
Experimental Design
Sixteen different fertilizer treatments were applied in tripli-
cate to the 48 plots arranged as shown in Fig. 2. The different
treatments are indicated by the arabic numerals 1 to 16. Repli-
cations of the same treatment are indicated by the letters A,
B and C. As originally laid out, each plot contained a minimum
of eight trees, four trees in each of two rows. All trees in any
single plot received the same treatment, but experimental data
were taken from only the four central trees. Thus the end trees
in each of the two rows in each plot acted as buffers between
treatments to the east and to the west in the same row. This
served to eliminate the border effects between different treat-
ments in the same rows. No buffer trees were provided between
different treatments in the rows to the north or to the south.
Since the tree rows were on beds 30 feet apart it was felt that
the border effects between different treatments across the rows
would be negligible.
Some of the trees in the groups of four in each plot, from
which data were originally taken, later died or declined from
causes other than those induced by the treatments. When these
conditions developed on plots that had originally contained more
than eight trees it was usually possible to substitute another
tree of similar size and vigor for data-taking purposes. On the
plots containing only eight trees originally, no substitutions
could be made except to include buffer trees, which would ob-
viously be undesirable. Therefore, when for any reason it be-
came undesirable to take further data from any of the four
trees originally designated for this purpose on any one of the
8-tree plots, it was necessary to secure subsequent data for that
particular plot from less than four trees. For this reason the
following data for each treatment are not necessarily based on
12 trees (four trees per plot replicated three times) every year.
In no instance were data taken from the buffer (end) trees in
the plots.
Fertilizer Treatments
The grove in which these plots were later established was
planted in April, 1929, and at that time each tree received %:
Fertilizer Experiments with Citrus
adjoined the plots on the east and north. An "Australian pine"
(Casuarina lepidophloia) windbreak running north and south
divided the experimental site. The layout of the experimental
grove is shown in Fig 2.
Experimental Design
Sixteen different fertilizer treatments were applied in tripli-
cate to the 48 plots arranged as shown in Fig. 2. The different
treatments are indicated by the arabic numerals 1 to 16. Repli-
cations of the same treatment are indicated by the letters A,
B and C. As originally laid out, each plot contained a minimum
of eight trees, four trees in each of two rows. All trees in any
single plot received the same treatment, but experimental data
were taken from only the four central trees. Thus the end trees
in each of the two rows in each plot acted as buffers between
treatments to the east and to the west in the same row. This
served to eliminate the border effects between different treat-
ments in the same rows. No buffer trees were provided between
different treatments in the rows to the north or to the south.
Since the tree rows were on beds 30 feet apart it was felt that
the border effects between different treatments across the rows
would be negligible.
Some of the trees in the groups of four in each plot, from
which data were originally taken, later died or declined from
causes other than those induced by the treatments. When these
conditions developed on plots that had originally contained more
than eight trees it was usually possible to substitute another
tree of similar size and vigor for data-taking purposes. On the
plots containing only eight trees originally, no substitutions
could be made except to include buffer trees, which would ob-
viously be undesirable. Therefore, when for any reason it be-
came undesirable to take further data from any of the four
trees originally designated for this purpose on any one of the
8-tree plots, it was necessary to secure subsequent data for that
particular plot from less than four trees. For this reason the
following data for each treatment are not necessarily based on
12 trees (four trees per plot replicated three times) every year.
In no instance were data taken from the buffer (end) trees in
the plots.
Fertilizer Treatments
The grove in which these plots were later established was
planted in April, 1929, and at that time each tree received %:
oolo 0 0 00 0oo
I 6-C 11 5-C
I1 9-A 11 10-B
o_oL o__oQ_oL o_o_
II 4-C I 3-C II 2-C I
oL o oLQoo o ooL o o oLo_
000
9-B
ooo__
oD o ooo-oo oo oooo
S II-B II 14-B i 13-B '1 12-B
olqo o oLo_o_o _O__ooj LoO_
L000010
I 16-C G I 15-C
_oooo L_ o_o__
Doooo o!!
I 8-C iI
O0 oo_ o_ 0
uool
9-C I
0 0oo
Fo-boooooToo-oolohoo
L o o o opo Lo o o oo
S I -C i 2-B II 3-B
i 0 -QQQ 0_0_ii 0
ooo o o o
II I-B 11 5-B
_ojLooooaajlp_oo
oHooolldT
1I 6-B 11
oI Lo oDoL
7-oo
7-B
_O0pOO
-oo o oogoo0
S 16-B 15-B
LQo__o__o__o__o _
:-o o ol
11 14 -C I
oiio o_o oo
u0 0 0 0 0
7-A
aQ oooo
S0ooo 00
i I-A
o-Q__o__o
dd o o:d o o D-o o dA o o d0 I
i1 13-A II 12-A I 14-A 11 IO-A i
o Lo__o _o Lo_oojQ o o oLooo__o__O
oodi-ooo
4-B 11 8-A
o_o jL_o__o__o
II 6-A II 5-A 11 4-A 1I 3-A II 2-A II -A
dlooo Lo_o dLo_o oLo_od o_o__opo _
ooo
10-C
000
-do0 o I o
I I II-C 1I 13-C I
dLo _oo:d_ o__o__oi
S 12-C II 16-A II 15-A
Looo__o__oo _oLo_oi_
Fig. 2.-Arrangement of experimental plots.
l[~D
0a
000
7-C
000
8-B
000
Fertilizer Experiments with Citrus
pound of goat manure. In the five-year period between plant-
ing and the establishment of the experimental plots the trees
also received applications of commercial fertilizer at fairly
regular intervals. The average approximate analyses of these
pre-experiment fertilizers was 3-8-8. The first application after
planting was at the rate of 1,4 pound per tree. This was gradually
increased until the last application before experimental treat-
ments were started was at the rate of 214, pounds per tree.
The experimental treatments consisted of applying 16 differ-
ent fertilizers with varying N-P-K ratios derived from various
materials to plots in triplicate. Only one application was made
annually, this in the spring, until April, 1942. Beginning at
this time a second application was made each year in the fall
until the experiments were terminated. The two annual fertil-
izer applications were the same for individual treatments with
respect to the phosphate and potash percentages. Except for
one treatment, no nitrogen was used in the fall fertilizer mix-
tures. In addition to the regular experimental N-P-K treat-
ments, occasional supplemental soil or spray treatments were
made with the sulfates of copper, zinc and manganese at the
same rate to all treatments. A complete summary of all treat-
ments, materials used, together with dates and rates of applica-
tion, is given in Table 4. Treatment 6 (3-6-12), in which nitro-
gen was derived from nitrate of soda, sulfate of ammonia and
castor pomace, phosphate from super or triple superphosphate
and potash from sulfate or muriate of potash, was considered
the "normal" or control treatment for these trials.
Reference to Table 4 will show that in some cases certain
changes were made from the original fertilizer formulae during
the course of the experiments. These changes Were found de-
sirable for various reasons as the work progressed. Where there
is evidence that these changes had any bearing on the results
obtained it will be discussed later under the particular element
concerned. Also, in some instances, the fertilizer materials used
originally became unprocurable and substitutions were necessary.
These are shown in Table 4. There were no indications that
any of these substitute materials altered the course of any of
the treatments.
The experimental treatments were carried on for 11 years.
At the end of this time rooting across the middles between trees
under different treatments to the north and south was beginning
to occur. This rendered the plots unsatisfactory for further
TABLE 4.-CHRONOLOGY OF TREATMENTS AND MATERIALS::: APPLIED,
Date I
Applied
March,
1934
March,
1935
March,
1936
March,
1937
March,
1938
March,
1939
April,
1940
April,
1941
April,
1942
Aug.,
1942
March,
1943
Oct.,
1943
April,
1944
Oct.,
1944
Legend
a
b-
c
d-
e -
f -
g -
h -
i-
j-
k-
1 -
Formula
Materials
Formula
Materials
Formula
Materials
Formula
Materials
Formula
Materials
Formula
Materials
Formula
Materials
Formula
Materials
Formula
Materials
Formula
Materials
Formula
Materials
Formula
Materials
Formula
Materials
Formula
Materials
| 1 2 | 3 [ 4
0-0-6 0-0-12 0-0-241 3-0-121
Sk I k I k a k
S0-0-6 0-0-12 0-0-241 3-0-12
k k k a k
0-0-6 1 0-0-12 0-0-241 3-0-121
k I k| ka k I
0-0-6 0-0-1 0-0-12 24 3-0-12
k ki k a k|
0-0-6 0-0-121 0-0-24 3-0-121
k k k a k1
3-12-24 3-12-24 6-12-24 3-0-12
a c k a c k a k a k
10-12-1210-12-24 6-12-24 3-0-12|
Se k ck ac ka kl
0-12-12 0-12-24 6-12-24 3-0-12
ck cka c k a k
0-12-12 0-12-2416-12-241 3 0-12
ck e k ac k a kl
0-12-1210-12-24 -12-24' 0--12
c c ac i !
10-12-1210-12-24i6-12-24, 3 0-12
e k ) e k a c k a k
0-12-1210-12-24 6-12-24 0-0-121
Sc j cj a c 5
[0-12-12 0-12-24 6-12-24 3-0-12
I cjb c ji b jP
5 6
6-0-121 3-6-
a kl a
6-0-12 3-6-
a k ac
6-0-12 3-6-1
a k ac
6-0-12 3-6-1
a k ac
6-0-12 3-6-1
a k ac
6-0-12 3-6-1
a k ac
6-0-12 3-6-1
a k ac
6-0-121 3-6-1
a k ac
6-0-121 3-6-1
a k ac
0-0-121 0-6-1
j c
6-0-12 3-6-12 3-6-24 3-12-12 3-12-241
a k' ac k ac ka c kha c kl
0-0-121 0-6-121 0-6 2410-12-1210-12-24
j I c ji c c jl c l
6-0-12 3-6-121 3-6-2413-12-12 1-12-24
b ji be i be jib c jib c jl
0-12-12i0-12-24 6-12-24 0-0-121 0-0-121 0-6-12 0-6-24 0-12-12 0-12-24
d j d j b d j i j jl d ji d j' d ji d j1
to Fertilizer Materials:
1/3 castor pomace, 1/3 nitrate of soda, 1/3 sulfate of ammonia.
1/2 nitrate of soda, 1/2 sulfate of ammonia.
triple superphosphate--44', P2-Os.
superphosphate-20' P-Os.
rock phosphate-formulation based on total P.O., content.
1/2 triple superphosphate-44 ,, 1/2 rock phosphate.
colloidal phosphate- (approx. 26',; total P..O-).
basic slag-(approx. 10'/ total P-Os).
dicalcium phosphate-38'/e P2O..
muriate of potash-62', KO.
sulfate of potash-48'/r KO.
potassium carbonate-65% K(O.
6-6-12 3 18-12 3-6-121
a c kla e kh a g k
0-6-12 0-18-121 0-6-12
c J e j g j
6-6-12 3-18-12 3-6-12
be j b e ji bg j
0-6-1210-18-12 0-6-12
d ji e i g i
Plot Series Treatment Numbers
1 7 8 9 1 10 1 11 12
12 3-6-24 3-12-1213-12-241 6-6-12 3-18-12 3-6-12
kl ac kla c kla c k ac kha e kl ag kl
12 3-6-24'1-12-1213-12-241 6-6-12 3-18-12 3-6-12
k| ac kia c k|a c kl ac kaa e k a g k
121 3-6-24 3-12-12 3-12-24 6-6-1213 18-12 3-6-121
kW ac kla c kha c kl a c kla e kl ag k1
12 3-6-24 3--12-13-12-241 6-6-1213 18-121 3-6-12
k! ac kha c ke a c kl a c ko a e k a g k
2 3-6-24 3-12-1213-12-24 6-6-1213-18-121 3-6-12
kl ac kha c kha c kl ac kha e k a g kX
12 3-6-241 -12-12 3-12-241 6-6-12 3-18-121 3-6-12
k ac k'a c ka c k a c k a e k ag k
12 3-6-2413-12-12' -12-241 6-6-1213 18-12T 3-6-121
kl a c kla c kha c kl a c k)a e kh a g kj
12| 3-6-24 3-12-12 3-12-24| 6-6-12 3-18-121 3-6-121
2 3-6-24[3-12-12 3-12-24| 6-6-1213 18-12 3-6-121
kh ac kha c kJa c kl a c ka e kl ag kl
2 0-6-2410-12-12|0-12-24 0-6-12 0-18-121 0-6-12
ij c jI c ji c j! c ji e | J i
3-6-12 3-6-121 3-6-121
bh j bi j| b f j
0-6-121 0-6-12 0-6-121
h jl i j f jl
3-6-481
be jl
0-6-48
d 1
SApril, 1940-sulfates of Cu. Mn and Zn 0t 1 lb. e
2 March, 1911--copper sulfate 0r 2 lbs./tree.
: April, 1944 -copper sulfate 0, 1/4 lb./tree. Nutrition
contained 3 lbs. each of the sulfates of Cu, Mn
per 100 gallons.
4October, 1914 copper sulfate 0, 1/4 lb.'tree.
Application
Lbs./Tree/
2
13 14 | 15 | 16 I
3-6-12 3-6-12 3-6-121 3-6-121
ah k a i k a f k ac 1!
3-6-12 3-6-12 3-6-12 3-6-121
ah k a i kla f k ac 1
3-6-12 3-6-12 3-6-12! 3-6-121
ah kh a i k a f k ac 11
3-6-12 3-6-12 3-6-12 3-6-121
ah k' a i k1 a f k ac 1|
3-6-12 3-6-12 3-6-12 3-6-12
ah h' ai ka f k ac i
3-6-12 3-6-12 3-6-12 3-6-24
a h ka i k af k' ac i
3-6-12 3-6-12 3-6-121 3-6-24
ah k| a i k a f k ac jj
3-6-12 3-6-12 3-6-12| 3-6-241
ah k1 a i k| a f k| ac
3-6-12[ 3-6-12 3-6-121 3-6-24
ah kj a i k a f kh ac jl
0-6-121 0-6-12 0-6-121 0-6-481
h il i j f j c JI
3-6-12 3-6-12 3-6-12[ 3-6-241
ah k a i k[ a f ki ac Jl
0-6-121 0-6-121 0-6-12 0-6-48
h j i j f ji c
4
4 2.
4
-5^
12' .
12
12
8
12
8
12 3
8
and Zn
Fertilizer Experiments with Citrus
investigations of this particular nature. The experiments were
therefore discontinued with the 1945 harvest.
Results
The index of the relative effectiveness of the various treat-
ments was based primarily on yield. The weight of fruit pro-
duced under each of the 16 treatments was obtained at harvest
time for nine successive years, beginning in 1937. Table 5
summarizes these yield data. All plots were picked on the same
date in any one year, but this date varied from April to August,
inclusive, in different years. The loss of fruit through pre-
harvest dropping when harvest was late in the season was ap-
preciable. In Table 5 the harvest dates are shown, together
with the average annual production per tree. Some idea of the
relationship between yield and date of picking may be obtained
from these data.
Included at the base of Table 5 are the least differences neces-
sary for statistically significant difference in yield between
treatments each year. The differences at both the 5 percent
and the 1 percent point are given. The nine-year tree average
POU!
FRI
26(
241
221
20(
181
16(
14(
12(
10
8
6(
4(
2
NDS DATE
F HARVESTED
UIT
00n
POUND
OF
FRUI
S260
0D 240
00 -220
00 200
0 /180
00DO 160
00 7-11-45 140
4-5-44 1
00 120
00 -8-2-43 I00
8-10-42 0
00 6-30-41 80
00 7-11-40 60C
00 -5-18-39 400
5-12-38
00 __ 200
4-19-37
0A~
4 5 8 9 3 I 10 13 12 6 2 II 14 15 7 16
TREATMENT NUMBERS
Fig. 3.-Cumulative totals of the average annual production per tree.
lS
T
0
'0
0
'0
10
)0
10
)0
TABLE 5.-AVERAGE ANNUAL PRODUCTION IN POUNDS PER TREE.
Treatment No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
*L.D.N.S. @ 5%9
L.D.N.S. @ 1%
4-19-37
94
123
132
108
126
132
121
89
152
135
134
121
123
175
156
147
71
95
S5-12-38
222
221
199
197
210
217
234
231
242
258
230
224
235
268
252
224
55
74
5-18-39
160
162
164
161
182
137
184
139
161
156
167
175
183
168
207
183
58
79
Harvest Date
7-11-40 6-30-41 8-10-42 8-2-43 4-5-44 7-11-45
265 275 322 281 323 296
271 300 343 308 304 371
241 285 313 270 320 312
138 145 174 77 287 153
143 195 197 119 313 186
208 285 343 322 365 345
301 286 353 357 357 349
228 233 265 261 275 303
239 240 300 269 252 378
245 264 311 306 310 312
265 308 341 313 328 332
231 264 329 265 392 326
256 280 286 252 356 344
261 294 277 319 355 317
264 274 351 343 309 385
302 277 390 334 360 351
69 65 83 49 67 91
93 87 112 66 90 123
* Lease difference necessary for significance.
Fertilizer Experiments with Citrus
annual production also was analyzed statistically. Results of this
analysis are shown in Table 6. The actual rank, according to
tree-average yields, of the 16 different treatments for each year
and the rank of the nine-year average are shown in Table 7.
These same data are presented graphically in Fig. 3, which
ranks the cumulative totals of the average annual production
per tree from the lowest to the highest yielding treatment.
TABLE 6.-SUMMARY OF STATISCAL ANALYSIS OF YIELDS, NINE-YEAR
AVERAGE PER TREE.
Treatment Yield Nine-Year Avg.
No. in Pounds/Tree
L.D.N.S.
Yields enclosed by same brackets are statistically equal
indicated at the head of each column.
Statistically Equal at
5 ( 1%
40
to the degree of significance
After these experiments had been in progress a few years
some treatments induced excessive dropping of mature fruits.
This was an important factor in the yield obtained from certain
treatments, particularly when the harvest was late in the season.
Each year, except 1940, when harvest was after May, drop
counts were made on all plots in order to be able to distinguish
between low yield caused by dropping of mature fruits and that
caused by the setting of a light crop or an early drop. In the
years these counts were taken they were started early in June,
which usually was about the beginning of the commercial harvest
in the grove that year. They were made at intervals frequent
enough that individual fruits on the ground did not deteriorate
I I
Florida Agricultural Experiment Station
TABLE 7.-YEARLY PRODUCTION RANK* BASED ON AVERAGE YIELDS.
Treat- ** Nine-
ment Year Year
No. 1937 1938 1939 1940 1941 1942 1943 1944 1945 Average
1 15 11 13 5 9 8 9 8 14 11
2 11 12 10 3 2 4 7 13 3 6
3 7 15 9 10 5 9 10 9 11 12
4 14 16 12 16 16 16 16 14 16 16
5 9 14 5 15 15 15 15 10 15 15
6 8 13 16 14 6 5 4 2 6 7
7 12 6 2 2 4 2 1 4 5 2
8 16 7 15 13 14 14 13 15 13 14
9 3 4 11 11 13 11 11 16 2 13
10 5 2 14 9 11 10 8 11 12 10
11 6 8 8 4 16 7 8 5
12 13 9 6 12 12 7 12 1 9 8
13 10 5 3 8 7 12 14 5 7 9
14 17 3 13 5 6 10 4
15 2 3 1 6 10 3 2 12 1 3
16 4 10 4 1 8 1 3 3 4 1
Ranked from 1 (highest yield) to 16 (lowest yield).
** The nine-year average is based on the actual nine-year average yield and is not the
average of the nine annual rank figures given.
to a point where they could not be recognized. All dropped fruits
were removed at each count so as not to be included in the next
count. A summary of the dropped fruit counts is given in
Table 8.
The influence of the various treatments on external appear-
ance and fruit size was examined each year, for the nine suc-
cessive years, through packinghouse studies. The fruit from
each treatment series was segregated and taken to the packing-
house where the standard commercial grades and sizes were
obtained. Results of these packinghouse studies are summarized
for grades in Table 9 and for sizes in Table 10.
Ammoniation (copper deficiency symptoms on citrus fruit)
became severe after several years under some treatments. In
addition to standard grades, the number of ammoniated fruits
from each treatment series was determined at the packinghouse
in 1943, 1944 and 1945. Table 11 summarizes the results of this
phase of the packinghouse studies.
Discussion of Results
Nitrogen.-The rates of N in the fertilizers for the various
treatments in this experiment were 0, 3 and 6 percent. The
nitrogen was derived equally with respect to N from nitrate of
soda, sulfate of ammonia and castor pomace until April, 1944,
TABLE 8.-PRE-HARVEST DROP COUNTS.*
Harvest Date-June 30, 1941 August 10, 1942
Avg. Avg. Avg. Avg.
Treat- No. No. I No. No.
ment Fruits Fruits Drop- Fruits Fruits Drop-
No. Har- Drop- ped t IHar- Drop- ped
vested ped ** vested ped
1 768 33 4 677 52 8
2 746 36 5 707 52 7
3 691 37 5 588 55 9
4 400 142 36 418 198 47
5 551 134 24 479 197 41
6 786 59 7 653 71 11
7 767 45 6 733 71 10
8 586 32 5 527 46 9
9 572 33 6 570 41 7
10 681 51 7 636 56 9
11 813 57 7 750 57 8
12 732 80 11 704 73 10
13 707 81 11 632 79 12
14 769 65 8 625 58 9
15 738 42 6 729 63 9
16 686 31 5 803 60 7
August 2, 1943 July 11, 1945
Avg. Avg. A Avg. Avg.
No. No. | No. No. c
Fruits Fruits Fits Drop- Fruits Fruits Drop-
Har- Drop- ped Har- Drop- ped
vested ped vested ped
669 168 25 612 138 23
688 119 17 728 142 19
658 157 24 659 149 23
186 364 196 332 140 42
248 387 156 436 199 46
781 267 34 677 149 22
832 177 21 676 136 20
564 146 26 636 108 17
618 111 18 749 114 15
744 209 28 613 178 29
871 232 27 680 182 27
602 353 59 649 92 14
588 363 62 704 135 19
780 313 40 691 198 29
818 230 28 837 163 19
700 151 22 684 120 18
* With exception of 1940, data on dropped fruits secured each year when harvest was after May.
** Fruits dropped prior to harvest and not included in number of fruits har;i\std.
- Calculated on basis of number of fruits harvested.
TABLE 9.-PERCENTAGE OF FRUITS IN VARIOUS GRADE CLASSIFICATIONS.
Year Grade
1937 1 + 2
3
Culls
1938 1+ 2
3
Culls
1939 1 +2
3
Culls
1940 1 2
3
Culls
1941 1 2
3
Culls
1942 1+2
Culls
1943 1+2
3
Culls
1944 1+ 2
3
Culls
1945 1 + 2
3
Culls
Avg. 1 + 2
3
Culls
1
87
12
1
82
14
4
82
16
2
81
17
2
78
20
2
58
38
4
69
28
3
45
54
1
17
81
2
67
31
2
2
71
28
1
78
17
5
82
17
1
80
19
1
63
34
3
45
51
4
50
47
3
38
61
1
7
91
2
57
41
2
3
68
31
1
66
27
7
77
22
1
65
33
2
65
33
2
47
50
3
60
35
5
51
49
0
11
87
2
57
41
2
4
88
11
1
68
30
3
76
22
2
41
31
28
48
27
25
24
22
54
4
38
58
54
45
1
13
78
9
46
34
20
Treatment No.
S7 8 9 10 11
80
I
68
74 71
26 28
0 1
60 54
33 38
7 8
75 67
24 32
1 1
68 70
31 29
1 1
59 52
30 36
11 12
48 40
49 56
3 4
42 34
51 48
7 18
42 51
58 49
0 0
8 6
90 92
2 2
53 50
44 45
3 5 3 3 5 5
88
11
1
59
34
7
69
30
1
69
29
2
77
19
4
48
47
5
41
56
3
44
55
1
14
84
2
56
41
13
65
34
1
60
35
5
68
31
1
67
29
4
75
21
4
47
42
11
59
30
11
77
22
1
26
71
3
60
35
S14 15 16
83 85 86
16 14 13
1 1 1
63 45 62
31 43 33
6 12 5
68 78 72
30 21 27
2 1 1
77 71 73
20 27 25
3 2 2
78 79 65
19 18 32
3 3 3
45 55 56
50 42 41
5 3 3
46 46 44
50 49 54
4 5 2
60 55 51
40 45 49
0 0 0
16 7 7
81 91 91
3 2 2
60 58 57
37 39 41
3 3 2
Fertilizer Experimen ts with Citrus
when castor pomace became unobtainable. The nitrogen for
the last two applications of the experiment, which were made
in April and October, 1944, was derived equally from nitrate
of soda and sulfate of ammonia. The period under this change
in nitrogen sources was of too short duration to have any ob-
servable influence on the course of the treatments.
Treatments 1, 2 and 3 received only potash in the fertilizer
from March, 1934, through March, 1938. Harvest data were ob-
tained in 1937 and 1938. Table 5 shows there was no statistical
difference in yields from these treatments and those from treat-
ments 4 and 5, which received nitrogen and potash but no
phosphate.
Following a period of unfavorable growing conditions in 1938
it was felt advisable to make certain changes in treatments 1,
2 and 3. In March, 1939, treatments 1 and 2 received a 3-12-24
fertilizer and treatment 3 a 6-12-24 mixture. Thereafter treat-
ment 1 received an 0-12-12 and treatment 2 an 0-12-24 at each
application for the duration of the experiments. Treatment 3
was retained on the 6-12-24 fertilizer.
With a few exceptions after 1939, yields from the no-nitrogen
treatments 1 and 2 were statistically equal to those from any
of the other treatments where phosphate was not the controlling
factor. From Table 5 it is found that only in one case was the
yield larger, to a high degree of significance, from a complete
fertilizer treatment than from either of the no-nitrogen treat-
ments. This was in the case of treatment 7 (3-6-24) as com-
pared to treatment 1 in 1943. With low degrees of significance,
treatments 15 and 16 were better than treatment 1 and treatment
7 was better than treatment 2 in 1943. In 1944 treatment 12
was, with slight statistical significance, better than either treat-
ment 1 or 2. As shown in Table 6, there was no significant
difference in yields for the average of the nine years between
either of the no-nitrogen and any of the complete fertilizer
treatments.
It is pertinent at this point to note the production rank of
the various treatments, as given in Table 7. After the changes
made in 1939 there were 12 treatments receiving a complete
fertilizer with nitrogen at 3 or 6 percent (Table 4), as compared
with two receiving no nitrogen. As will be seen from later dis-
cussion, nitrogen, phosphate and potash available to the tree
were at least adequate when a complete fertilizer or one without
nitrogen was used. Thus, while there were six times as many
TABLE 10.-PERCENTAGE OF FRUITS IN VARIOUS SIZE CLASSES:
Fruit Size Classes
Medium
176 i 200
Treat-
ment
No.
1
2
3
4
5
6
7
8
9
10
12.7
14.2
13.3
13.6
13.8
11.5
12.4
12.7
14.7
11.1
12.5
14.3
13.2
12.4
12.4
13.0
216
17.8
20.0
17.0
16.2
15.5
16.5
15.7
15.6
14.7
13.9
15.9
16.2
15.3
16.3
11.6
15.2
Small
250 288 324 Smaller
20.8 14.8 5.9 8.0
22.6 12.3 4.6 3.5
20.4 12.1 4.6 7.5
17.3 15.1 4.4 8.9
20.4 17.0 5.2 9.9
16.0 17.7 6.4 10.4
19.9 14.6 4.8 7.3
20.3 13.8 5.8 10.4
18.9 11.8 4.6 6.3
21.6 15.4 5.8 9.7
22.2 16.4 5.6 9.0
20.0 14.8 4.0 6.9
21.9 14.7 3.8 7.9
24.0 19.5 5.9 7.1
23.2 17.5 6.9 9.1
19.0 15.4 4.9 8.0
NINE-YEAR AVERAGE.
Summary-Percentage
cf Large, Medium and
Small Sizes
96 176 250
to 200 and
150 216 Smaller
8.4 42.1 49.5
10.8 46.2 43.0
11.6 43.8 44.6
12.2 42.1 45.7
8.0 39.5 52.5
11.6 37.9 50.5
12.4 41.0 46.6
10.3 39.6 50.3
15.0 43.4 41.6
11.0 36.5 52.5
8.7 38.1 53.2
11.2 43.1 45.7
10.9 40.8 48.3
6.8 36.7 56.5
8.8 34.5 56.7
11.7 41.0 47.3
* Does not include culls which were not sized.
11.6
12.0
13.5
12.3
10.2
9.9
12.9
11.1
14.0
11.5
9.7
12.6
12.3
8.0
10.5
12.8
Large
112 126
.3 2.4
.5 2.7
1.3 3.4
1.2 3.3
.7 1.9
1.5 2.8
1.4 3.7
.9 2.5
1.2 4.4
1.2 2.8
.9 2.1
1.1 2.7
1.4 2.9
.3 1.4
1.0 1.9
1.0 2.8
Fertilizer Experiments with Citrus
complete fertilizer treatments equally as well supplied with
phosphate and potash as the no-nitrogen treatments, the com-
plete fertilizer treatments failed considerably to outrank in pro-
duction the no-nitrogen treatments in this ratio. After the
inclusion of phosphate in treatments 1 and 2 in 1939 the pro-
duction rank for the average of the two treatments was between
seventh and eighth. Several complete fertilizer treatments
ranked below this on the average. There was little consistency
in production rank for most of the treatments. In the instances
cited above where complete fertilizer treatments gave signifi-
cantly higher yields than the no-nitrogen treatments, the su-
perior treatments were in one of the top three positions in
production rank for that season.
The relative performance with respect to tree-average yields
of all the various treatments is shown graphically in Fig. 3
for the nine years in which production data were secured. Note
that the cumulative total of the average annual production from
treatment 2 is slightly higher than from treatment 1. Treat-
ment 2 appears to be slightly better than the average of the
complete fertilizer treatments (all treatments except 1, 2, 3, 4
and 5 prior to 1939 and all except 1, 2, 4 and 5 after 1939), while
treatment 1 is slightly below.
The dropping of mature fruits was apparently no more se-
vere, or perhaps slightly less severe, from the no-nitrogen treat-
ments than from those receiving a complete fertilizer. From
Table 8 it is found that the dropping from treatment 1 was
slightly larger than that from the average of the complete
fertilizer treatments in only one out of the four years in which
drop counts were made. The dropping from treatment 2 was
less each of the four years than the average from the complete
fertilizer treatments. Both treatments 1 and 2 averaged con-
siderably less dropping than the average of all treatments.
The influence on grades of the various treatments is shown in
Table 9. From this it will be observed that on the average the
no-nitrogen treatments generally produced fruit of as good ex-
ternal quality as any of the other treatments, or better. This
tendency is in agreement with the views held by early Florida
growers as reported by Webber (20). While inspecting these
data it should be kept in mind that treatments 1, 2 and 3 con-
tained only potash up to March, 1939. As will be brought out
in the discussion of phosphate, the absence of phosphate in the
fertilizer resulted in a striking reduction in fruit quality rather
Florida Agricultural Experiment Station
early in these experiments. According to the data in Table 9,
these symptoms did not materialize where nitrogen as well as
phosphate was omitted. External appearance of the fruit from
treatment 1 was somewhat better than that from treatment 2.
This may perhaps be accounted for by the lower potash content
of treatment 1. While it is not apparent from these data, high
potash tended to give a relatively large number of large, coarse-
textured fruit. This was most pronounced in treatment 16 after
48 percent potash was used and will be discussed further under
potash.
In the trade, medium size fruits are generally most accept-
able. The most desirable treatment with respect to sizes would
be one producing a large percentage of the crop in the size 176
to 216 range. Over the period of nine years treatment 2 aver-
aged the highest percentage of fruits in this size range, as shown
in Table 10. Treatment 1 was in fifth place with treatment 4.
The difference in the sizes from treatments 1 and 2 perhaps can
be accounted for by the higher potash in treatment 2. The per-
centage of sizes 176 to 216 from either treatment 1 or 2 was
higher than the average of all the other treatments. While the
size differences were probably insignificant, it seems evident
that nitrogen was not necessary in the fertilizers applied on these
plots in order to obtain a relatively high percentage of desirable
sizes.
According to Camp and Fudge (4), ammoniation may be classi-
fied as either an excess of nitrogen as compared with copper
or a deficiency of copper as compared with nitrogen. They con-
sidered the latter viewpoint the more practical. In either case
a broad N:Cu ratio results in the tree. Although copper was
applied to all treatments at the same rate from time to time,
as indicated in Table 4, the incidence of ammoniation was much
greater under some treatments than under others. Contrary
to what would normally be expected, it was not necessarily
associated in these experiments with high nitrogen fertilization
(Table 11). Rather it appeared to be correlated with high phos-
phate fertilization where superphosphate was the source. The
implications and significance of this will be considered fully
under the discussion on phosphate. It is sufficient to point out
here that for the two seasons prior to the correction of this
trouble on all plots by a copper spray the no-nitrogen treatments
1 and 2 ranked fifth and third, respectively, in percentage of
ammoniated fruits.
Fertilizer Experiments with Citrus
TABLE 11.-AMMIONIATION COUNTS ON HARVESTED FRUIT.
Harvest Date-August 2, 1943 April 5, 1944 July 11, 1945
669 9.9 1.5 835 33.5 4.0 612 0.5 0.1
2 688 29.7 4.3 757 54.5 7.2 733 0.7 0.1
3 658 21.7 3.3 821 45.8 5.6 659 0.3 0.1
4 186 2.0 1.1 730 1.1 0.2 332 0.5 0.2
-Z4 Z Z Z Z
o >
1 669 9.9 1.5 835 33.5 4.0 612 0.5 0.1
2 688 29.7 4.3 757 54.5 7.2 733 0.7 0.1
3 658 21.7 3.3 821 45.8 5.6 659 0.3 0.1
4 186 2.0 1.1 730 1.1 0.2 332 0.5 0.2
5 248 1.5 0.6 811 5.8 0.7 435 0.4 0.1
6 781 1.7 0.2 1,045 3.4 0.3 677 0.8 0.1
7 832 0.5 0.1 948 2.0 0.2 676 0.3 0.1
8 565 71.5 12.7 751 130.8 17.4 636 0.4 0.1
9 620 119.4 19.3 651 198.9 30.6 749 0.3 0.1
10 744 3.3 0.4 768 2.6 0.3 613 0.1 0.1
11 871 1.8 0.2 855 1.1 0.1 680 0.0 0.0
12 602 2.7 0.5 1,053 0.7 0.1 649 0.1 0.1
13 588 3.7 0.6 868 0.5 0.1 704 0.3 0.1
14 779 3.4 0.4 931 0.4 0.1 691 0.5 0.1
15 818 3.1 0.4 857 1.0 0.1 837 0.4 0.1
16 700 7.6 1.1 1,005 0.8 0.1 684 0.0 0.0
With the exception of the short period of 1938-39, to which
attention has been called previously, the general condition of
the trees in the no-nitrogen treatments was equally as good as
in any other treatment. They made good growth. The foliage
was normal in size, color and density. Foliage analysis made
late in the course of the experiment showed the leaves from
treatments 1 and 2 (no N fertilizer) to contain 2.94 and 3.14
percent nitrogen, respectively, on a dry matter basis. Leaves
from treatment 6 (3 percent N) contained 3.00 percent nitrogen
and those from treatment 5 (6 percent N) contained 3.11 percent.
There was no marked difference in results obtained from the
3 percent and 6 percent nitrogen treatments where phosphate
and potash were held constant. This is illustrated by compar-
ing the performance of the trees under treatment 6 (3-6-12)
with those under treatment 10 (6-6-12) and those under treat-
ment 3 (6-12-24), after 1939, with those under treatment 9
(3-12-24) for the same period. Although the yield from treat-
ment 6 averaged slightly higher than that from treatment 10
it was never statistically larger (Table 5). The dropping of
mature fruits, grades, sizes and ammoniation also were prac-
tically equal for the two treatments. There was no observable
Florida Agricultural Experiment Station
difference in tree condition under the two treatments at the
conclusion of the experiments. Average yield from treatments
3 and 9 was the same. In only one instance was there a statis-
tical difference. Treatment 3 was better than treatment 9 in
1944 at a low degree of significance (Table 5). Treatment 9
averaged a slightly lower percentage of drops than treatment 3.
They were practically equal in respect to grades and sizes. Am-
moniation was prevalent under both treatments, but much more
severe under treatment 9, which ranked the highest of all treat-
ments in this respect (Table 11). Tree condition under the two
treatments appeared to be the same when the experiments were
terminated.
From the foregoing discussion on nitrogen it is apparent that,
under conditions of these experiments, no beneficial results were
obtained from nitrogen fertilization at the rates used insofar
as fruit yield, quality or size and tree condition were concerned.
Unpublished data from a subsequent nitrogen experiment on
a considerably more extensive scale operated four years in an
adjoining grove on the same soil planted to the same ages and
variety trees substantiate these findings. In this latter case
the plots were on soil ranging from about 14 to 50 percent
organic matter. During the three years harvest data were
obtained, no increase in yield was secured over no nitrogen
from light, medium or heavy nitrogen applications applied at
various times throughout the year. This was equally as true
for the plots on soils with the lowest organic matter content as
those higher in organic matter.
Phosphate.-Phosphate was used in the fertilizer for the vari-
ous treatments of this experiment at the rates of 0, 6, 12 and
18 percent total P-O,. Superphosphate was the source in all
treatments except 11, 12, 13, 14 and 15. Rock phosphate was
used on treatment 11 at the rate of 18 percent total P,20,. Treat-
ment 12 received colloidal phosphate, treatment 13 basic slag
and treatment 14 dicalcium phosphate, each at 6 percent total
P0.,. The phosphate for treatment 15, which was also at the
rate of 6 percent, was derived about equally each application
from superphosphate and rock phosphate.
Some of the various phosphate treatments produced results
of most striking and significant interest. At the conclusion of
the experiment trees on the no-phosphate plots were smaller
and showed much less vegetative growth than the other treat-
ments. The foliage was smaller with a somewhat narrow and
Fertilizer Experiments with Citrus
stunted appearance. The reason no more serious symptoms
developed prior to the termination of the experiment was per-
haps due to the limited amounts of PO., received annually from
the castor pomace included in the mixed fertilizer and also to
the cross rooting effects whereby trees on the no-phosphate plots
obtained small amounts of P,O., from adjacent treated plots.
Phosphorus deficiency symptoms on citrus have been reported
by Haas (12) working with culture solutions and by Chapman
and Brown (5) in soil cultures. These symptoms were described
as irregular shaped "burned" spots, with the older leaves being
the first affected.
As early as 1938 some very obvious differences were noted
in appearance of the fruit. That from treatments 4, 5 and 12
had a much brighter orange color early in the season. This led
to a belief that the fruit from plots receiving limited amounts
of phosphate might be more mature. Fruit samples were col-
lected from certain treatments and tested for maturity and
examined for other fruit characteristics. Table 12 shows the
variation of brix and acidity with phosphate treatment for the
1943 crop. These variations are characteristic of those obtained
by sampling the crop from the five previous years (11, 17). When
measured by the ratio of brix to acid, the fruit from treatments
4 and 12 was actually less mature than that from the super-
phosphate treatments. Fruit from the colloidal phosphate plots
was more mature than that from the no-phosphate treatment
and less mature than that from the superphosphate plots. As
early as 1931 Takahashi (18) found that spraying citrus with
phosphoric acid or fertilizing with phosphate decreased the citric
acid content of the juice. Other investigators (19, 5, 2, 3) also
have found that a decreased content of citric acid may be asso-
ciated with the use of phosphate fertilizers. It is known that
the use of arsenical sprays decreases the acid content of citrus.
In view of the chemical relationship between phosphorus and
arsenic, it is perhaps significant that phosphorus exerts this
same effect.
Not only were there differences in fruit color and maturity
associated with phosphate treatment but there were other qual-
ity differences. Fruit from the no-phosphate treatment was
slightly smaller in size, had a rough, coarse texture and was
slightly elongated with a thick, wrinkled stem end. (See Fig. 4).
Cut half-sections of this fruit are shown in Fig. 5. Fruit from
the no-phosphate treatment 4 had a coarse internal structure
Florida Agricultural Experiment Station
with large juice sacs and a thick core. The differences in rind
thickness are obvious in the photograph. Rind thickness for
various phosphate treatments was measured and data taken
on samples from the 1943 crop are recorded in Table 12. Here
again colloidal phosphate is intermediate between the super-
phosphate source and the no-phosphate treatment.
Fig. 4.-Exterior view of fruit from treatments including no phosphate
(left) and 6 percent P.OR (right).
Phosphate treatment had a very decided effect upon yield.
The production rank for all 16 treatments is recorded in Table 7.
Treatments 4 and 5, no phosphate, show the lowest production
for the nine years. On all harvests but one from 1940 to the
termination of the experiment these two no-phosphate treat-
ments ranked lowest. It is interesting to note that treatment 5
always outranked treatment 4. This may be because treatment
5 received more castor pomace as a nitrogen source and this
Fertilizer Experiments with Citrus
Fig. 5.-Cut half-sections of fruit from no-phosphate (left) and 6 percent
P.O- (right) treatments.
material carried a small amount of P.O,. Treatments 8 and 9,
the highest two treatments in soluble phosphate, ranked 14 and
13, respectively, next to treatments 4 and 5. This is due to the
large amount of ammoniation on these high-phosphate treat-
ments, which probably induced considerable early drop of green
fruit. This ammoniation factor will be discussed later. The
high significance of the low yields of certain of the phosphate
treatments is readily seen from Table 6.
It is interesting to note that yields from the so-called slowly
available sources of phosphate, i.e., colloidal phosphate, rock
phosphate, dicalcium phosphate and basic slag, were all on the
high end of the cumulative yield curves (Fig. 3). Dicalcium
phosphate was probably more available than the other insoluble
sources on this low pH soil. Rock phosphate was applied at
three times the rate of the other materials. This may account
for the fact that treatments 11 and 14 both ranked higher in
Florida Agricultural Experiment Station
yield than treatments 12 and 13. However, there was no sta-
tistical difference between any of these plots receiving insoluble
sources of phosphate (Table 6).
It was indicated as early as 1940 that the amount of dropping
of mature fruit before harvest was associated with a lack of
available phosphorus. Drop counts for the four years on which
such counts were made are recorded in Table 8. Here it is evi-
dent that a large amount of dropped fruit is definitely associated
with low phosphate. Drop counts for the 1943 crop are listed
in Table 12 according to phosphate treatments. This was a year
of abnormally heavy loss by drops. The no-phosphate treat-
ment 4 lost approximately twice as much fruit by drops as was
harvested. Treatments 12 and 13 lost considerably more fruit
than those plots receiving applications of the more readily avail-
able superphosphate. Treatment 11, with rock phosphate,
showed a drop count approximately the same as the superphos-
phate plots. The apparent superiority of rock phosphate over
the other two insoluble sources is due to differences in the ap-
plication rates. Colloidal phosphate and basic slag were applied
in such quantities that the total P.,O, applied was the same as
the P.,O, of the suprephosphate used on treatment 6. Since the
PO.-, from superphosphate is practically all readily available,
the trees on treatment 6 had access to more available P.,O than
those on treatments 12 and 13, which reflected their lack of
available phosphorus by an increase in dropped fruit. Rock
phosphate was applied in such quantities that the total PO.,
applied was three times that of the P.,O, used on treatment 6.
Since its rate of application was triple that of the other in-
soluble phosphate sources it furnished more available POn than
did the colloidal and slag applied at the lower rates.
Differences in available phosphorus exhibited by these several
treatments is shown in Table 13. The soil tests for water-
soluble phosphorus showed a very low value for the no-phosphate
plots, a somewhat higher value for those plots receiving colloidal
and basic slag and a decidedly higher value for the plots receiv-
ing the triple rate application of rock phosphate. The available
phosphate as represented by this test was approximately the
same on the rock phosphate plots as on those receiving 6 percent
P.,O. from superphosphate. Availability of the phosphate from
these sources is reflected in the leaf and juice analyses listed in
Table 13. Phosphorus was very low in the leaves and juice of
samples from the no-phosphate plots.
Fertilizer Experiments with Citrus
TABLE 12.-EFFECT OF PHOSPHATE TREATMENT ON BRIX, ACIDITY, RIND
THICKNESS, PRE-HARVEST FRUIT DROP AND CULLS. 1943 CROP.
Phosphorus Brix*
Treatment |(S18C.
I
Citric
Acid,*
Per-
cent
Thick-
Brix ness*
of
Acid Rind,
mml.
Drops,** Culls,**
Per- Per-
cent cent
no PO- 143 199 719 5.03 196 58
6 6%C P20,
(super)
8 12'/, PLO,
(super)
11 18% PO.
(rock)
12 6% P2O,
(colloidal)
13 6%/ PTOU
(slag)
I 13.0
12.0
13.1
1.50 8.69 3.51
1.37 8.78 3.77
1.67 7.84 4.13
Average of 32 fruit from each of the three replication collected 2 18 43.
** Percentage calculated on the basis of number of fruits harvested on 8 2 43.
TABLE 13.-INFLUENCE OF PHOSPHATE TREATMENT ON THE PHOSPHORUS
IN SOIL, LEAF AND FRUIT SAMPLES.*
Phosphorus
Treatment
4 no PRO ..
5 no PO ......
6 6%/ P2O, (super) ...
10 6% P.O, (super) .....
8 12% PO, (super) ...
9 12% PO- (super)
11 18'; P.O; (rock) .....
12 6%( P.O. (colloidal)
13 6% P.O. (basic slag)
Water Soluble
Phosphorus
in Soil,
Lbs. per Acre
4.8
3.8
13.1)
12.6 /
32.1\
35.9)
Total
Phosphorus
in Leaves,
Percent**
0.140
0.149
0.186
0.179
0.188
0.190
13.5 0.172
6.3 0.166
7.6 0.174
Collected just prior to the 1944 harvest. Recorded
cations.
** Oven-dry basis.
? Determination made directly on the clarified juice.
Soluble
Phosphorust
in Juice,
pp1m.
2.04
2.43
7.22
6.09
7.94
6.97
5.27
4.39
4.17
values are averages of three repli-
Assuming that excessive dropping of mature fruit before
harvest was at least in part due to lack of available phosphorus,
34 5
26 8
27 5
59 14
62 11
' .
Florida Agricultural Experiment Station
drop count data in Table 12 and analytical data in Table 13 are
in good agreement. Those treatments showing severe dropping
were very deficient in phosphate, as indicated by soil, leaf and
juice analyses. Those showing moderate dropping analyzed
somewhat lower than normal in phosphate. Much of the fruit
that was harvested from the phosphate-deficient plots was soft
and of poor quality. This is shown in Table 12 by the relatively
large percentage of culls from treatments 4 and 5. Here again
the colloidal and slag plots showed values somewhat higher than
the rock or superphosphate plots.
Treatments 8 and 9 showed an increase in culls also, but this
was due to ammoniation. The relationship between this factor
and treatment will be considered later in this discussion. Refer-
ence to Table 9 shows that treatments 4 and 5 began to show
an increase in culled fruit with the 1940 harvest and, with the
exception of 1944, when the fruit was harvested early, continued
to exhibit this characteristic. For the nine-year average these
treatments gave a high percentage of culls and treatments 12,
13 and 9 gave a somewhat greater than normal percentage of
culls. Collinson (8) and Young (21) found no evidence that
phosphate fertilizers increased fruit quality. However these
investigators were probably working with increments at higher
phosphate levels and were taking their data from earlier har-
vests of fruit. Kinnison and Albert (15) found that a higher
percent of fancy fruit was produced on plots fertilized with
P2O., and K20 than with manure alone.
Attention has been called already to the low yields and high
cull counts of treatments 8 and 9 and a possible connection be-
tween these factors and ammoniation. As early as 1941 am-
moniation symptoms had become quite evident on certain plots.
This condition was found to be restricted almost entirely to
treatments 8 and 9, which had remained a heavy superphosphate
treatment since the beginning of the experiment. The serious-
ness of this condition on these plots increased the next year and
in 1943 ammoniation counts were made on the fruit from all
treatments as it was graded through the packinghouse. These
data are recorded in Table 11, along with similar counts made
in 1944 and 1945. It is evident from the 1943 and 1944 am-
moniation counts that this condition was quite serious on treat-
ments 8 and 9 and somewhat more than normal on treatments
1, 2 and 3. These are all treatments that had received a 12
percent P20,O fertilizer using superphosphate as the source.
Fertilizer Experiments with Citrus
Treatments 8 and 9 had received this high-phosphate fertilizer
since the beginning of the experiment, while 1, 2 and 3 had been
without phosphate until March, 1939, after which time they
also received a 12 percent P2Os mixture (see Table 4). In 1944
a nutritional spray program including copper was initiated on
the entire grove. This practically eliminated ammoniation from
the 1945 crop, as may be seen from-Table 11.
Since ammoniation is a symptom of copper deficiency and all
trees in the experiment had received the same amount of copper
applied to the soil, evidence is rather strong to the effect that
in some way the higher amounts of available phosphorus in the
soil were interfering with copper assimilation by the trees.
The apparent effects of phosphorus upon the availability of cer-
tain minor elements have been reported by other investigators.
Chapman and others (7) have shown that mottled leaf (zinc
deficiency) is aggravated by increasing PO5, in the nutrient
solution. These investigators also produced iron deficiency in
the same manner. Haas (13) secured better growth of citrus
in solution culture with an intermittent rather than a continuous
supply of P.05. He attributed this to the probable effect of
phosphorus on the absorption of minor elements by the plant.
The extent and relation of copper deficiency symptoms as evi-
denced by ammoniation to fertilizer treatment in this set of plots
has been mentioned by Forsee and Neller (11) and discussed
in some detail by Forsee and Allison (10). Copper analyses
have been made on soil, leaf and fruit samples from certain
treatments representing various soil levels of phosphorus sup-
plied from soluble and insoluble sources. These analyses are
recorded in Table 14, along with ammoniation counts for the
1944 harvest.
Soil analyses for total copper indicate that trees on plots
receiving increasing amounts of superphosphate had access to
as much or even more copper than those on plots receiving no
phosphorus in the fertilizer. However, as the superphosphate
treatment increased from no phosphate to 12 percent phosphate
the assimilation of copper decreased, as is evidenced by the cop-
per contents of the leaves (old and new growth), juice and
seeds. The greater assimilation of copper by the no-phosphate
treatments suggests a larger overall removal of copper by these
treatments and could possibly account for the lower average of
total copper in the soil from treatment 4. Leaf samples from
the colloidal phosphate and basic slag treatments show copper
Florida Agricultural Experiment Station
values intermediate between the no-phosphate treatments and
the 6 percent superphosphate treatments and samples from
the rock phosphate treatment show copper values approximately
the same as the 6 percent superphosphate treatments. This
seems to correlate with the amount of available phosphate in
the soil as determined by tests for water-soluble phosphate
(Table 13) and indicates that copper assimilation by the tree
is inversely proportional to the amount of active phosphorus
in the soil.
In studying soils from rather widely different parts of Flor-
ida, Jamison (14) found little difference, by laboratory methods,
in the fixation of copper in the presence and absence of super-
phosphate. In lysimeter experiments on virgin Norfolk fine
sand Erwin (9) found that copper in the plant was decreased
and copper in the leachate was increased as the phosphorus in
the soil was increased up to a certain level. Beyond this level
of phosphorus the leaching of copper was depressed. The total
copper leached from any treatment, however, was less than 1
percent of the copper added. While the data included in Table
14 and discussed above and the experiments of Erwin (9)
definitely indicate that soil applications of soluble phosphorus
influence copper assimilation, this influence may be indirect and
due to some factor other than straight fixation of copper by
phosphates. An added explanation may lie in the stimulating
effect of high phosphate in the soil on the activity of such soil
organisms as the ammonifying and nitrifying bacteria. Such
a stimulation would release additional amounts of nitrogen from
the soil organic matter and this might create an unbalance be-
tween nitrogen and copper and induce the observed ammoniation
symptoms.
Potash.-As originally designed the potash treatments were
at the rates of 6, 12 and 24 percent KO in the fertilizer. The
sources of potash used at the beginning were sulfate of potash
and potassium carbonate. As the experiments progressed mu-
riate of potash was substituted for potassium carbonate (treat-
ment 16) and later for sulfate of potash in all treatments. There
was no evidence that any of these changes in materials in-
fluenced the course of the treatments concerned. Certain changes
were also made from time to time in the treatment rates so
that by 1942 they were essentially 12, 24 and 48 percent potash.
All details concerning the fertilizer mixtures used are recorded
in Table 4.
Fertilizer Experiments with Citrus 35
TABLE 14.-EFFECT OF PHOSPHORUS TREATMENT UPON COPPER ASSIMILA-
TION AS INDICATED BY LEAF AND FRUIT ANALYSES AND UPON THE INCI-
DENCE OF AMMONIATION OF THE FRUIT.
ppm. in Cu Per-
Phosphorus ppm. Cu Oven-Dry ppm. Cu ppm. Cu cent
No. Treatment in Leaves* in in Ammoni-
Soil Old New Juice* Seed* ation**
Growth Growth
4 No PO, 185 14.3 4.8 0.73 10.5 0.2
5 No P O ...... ..... ...... 0.40 .... 0.7
6 6% PO,
(super) 260 3.9 3.1 0.25 11.5 0.3
10 6% P.O,
(super) ...... ...... 0.28 ...... 0.3
8 12<; P .,O
(super) 263 .. .. ...... 0.16 2.5 17.4
9 12', P.,O
(super) ..... 2.2 1.6 0.15 4.3 30.6
11 18 P.O.
(rock) ..... 3.6 ...... .... 0.1
12 6', PO.,
(colloidal) .. 7.7 .... ....0.1
13 6', PO-
(slag) ..... 7.1 ..... ..... 0.1
Copper analyses were made spectrographically by T. C. Erwin of the Florida Experi-
ment Station Soils Department. Gainesville.
** Ammoniation counts were made on the harvested fruit.
As shown in Fig. 3, treatments 7 and 16 (24 and 48 percent
KsO) gave the two highest nine-year cumulative average yields
of the 16 treatments. Their production rank in Table 7 indicates
that they were rather consistently good producers throughout
the experiment. However, an examination of Table 4 will show
that where nitrogen and phosphate were held constant there
was no statistical difference in yield from the different potash
treatments at any time during the course of the experiment.
In this respect compare treatment 1 (0-12-12) with treatment 2
(0-12-24), treatment 6 (3-6-12) with treatment 7 (3-6-24), and
treatment 8 (3-12-12) with treatment 9 (3-12-24). Within these
pairs of treatments the dropping of mature fruits, grades, sizes
and ammoniation were also more or less uniform. In fact, where
phosphate and potash were both supplied in the fertilizer at
any of the levels used in these tests, regardless of the nitrogen
rate, there was no statistical difference in the nine-year average
Florida Agricultural Experiment Station
yields from any of the treatments except for treatment 8. As
indicated in Table 6, treatment 8 produced statistically less fruit
than several of the other treatments. The cause of the sig-
nificantly lower yield from treatment 8, and the relatively low
yield from treatment 9, has been considered fully in the discus-
sion on phosphate and need not be examined in detail here. It
is sufficient to point out that the controlling factor appeared to
be a phosphate-copper interrelationship rather than potash.
The rate of potash application was reflected in the potassium
content of the leaves. Foliage analyses on samples collected
near the termination of the experiment gave the following re-
sults: treatment 6 (3-6-12) contained 1.594 percent K on dry
matter basis, treatment 7 (3-6-24) contained 1.978 percent and
treatment 16 (3-6-48) contained 2.145 percent.
None of the variations in potash treatment produced outstand-
ing results with respect to grades and sizes, with the possible
exception of treatment 16. Although it is not evident from the
tabulated data, after the potash was increased from 24 to 48
percent in this treatment a relatively high percentage of large
coarse fruits was produced. These fruits were generally some-
what wrinkled and green, particularly around the stem end.
Many were slightly misshapen. Similar results have been re-
ported by Chapman, Brown and Rayner (6) for excessive potash
on young Washington Navel and Valencia orange trees growing
in water culture.
Except for the short period when treatment 1 was at 6 percent
potash, there were no potash treatments lower than 12 percent.
In the absence of prolonged potash treatments at rates lower
than 12 percent it is not possible to determine from these data
at what level this element would become a limiting factor of
sufficient magnitude to reduce the yield significantly. With the
possible exception of the previously mentioned symptoms of
potassium excess in the fruit, neither was the 48 percent potash,
treatment 16, continued sufficiently long to bring about symptoms
of potassium toxicity or excess, if such could occur at this potash
level under the conditions of this experiment. Although im-
probable, there may have been an increment between 12 and 24
percent KO at which statistically larger yields could have been
obtained. However, these data do not lend themselves to specu-
lation on this point through interpolation.
There was no difference between the various potash treatments
with respect to tree growth or appearance so far as could be
Fertilizer Experiments with Citrus
determined from observation on trees where phosphate was also
applied.
General Discussion
The absence of buffer rows on the north and south sides of
the individual plots and the occurrence of cross-rooting between
trees under different treatments late in the life of these experi-
ments has been mentioned. It was not practicable to determine
the exact extent to which cross-rooting had occurred. A fairly
extensive survey in 1944, however, indicated it was becoming
rather general. There was no accurate way to determine to
what extent the resulting cross-feeding between adjoining plots
had influenced the performance of trees under the various treat-
ments. A rough estimate of this, however, was possible by com-
paring the performance records of the individual plots in the
southern-most tier of the experimental tract (see Fig. 2), which
had opportunity to cross-feed only to the north, with the other
two plots of that respective treatment series, which had oppor-
tunity to cross-feed on both sides. For this comparison it was
necessary to refer to the original harvest data for each individual
plot, which are not published here. An inspection was made
of these data for the last five years of the experiment, beginning
with 1941 before any cross-rooting was noted. The average
yield for this period from most of the plots in the southern-
most tier was slightly less than the average yield from the
other two plots of that particular treatment series, but in no
case was the difference statistically significant. Furthermore,
with one minor exception the percentage of drops was less from
the outside tier of plots than the average from the same treat-
ments on interior plots. From these data it appears fairly evi-
dent that cross-feeding between plots had not greatly influenced
the performance of trees on the individual plots.
In the course of the statistical analyses of the harvest data
it was observed that the performance of trees under most of
the treatments varied considerably from year to year. Produc-
tion rank data in Table 7 emphasize this point. With those
treatments receiving no phosphorus or insufficient readily avail-
able phosphorus, a portion of this variation might be accounted
for by variations in the picking date. It has been previously
pointed out that the dropping of mature fruits was associated
with a lack of available phosphorus and that the later the date
of picking the greater this drop. Some indication of this is
obtained from Table 8. However, most of the variation observed
Florida Agricultural Experiment Station
could not be explained on such a basis. The most logical assump-
tion here seems to be that in some instances climatic environ-
mental factors had more influence on yield than did the treatment.
A treatment that produced good results under one set of weather
conditions may have been poor another season under different
weather conditions.
In considering results obtained from the 16 different treat-
ments included in these experiments it should be of interest,
particularly to those engaged in experimental work employing
field plot technique on tree crops, to note in Fig. 3 that prior to
1940 yields were all more or less equal. The treatments were
initiated in 1934 on plots which had received little fertilizer pre-
viously. Thus, with tree crops several years may elapse after
beginning treatment before any appreciable response appears.
Except in the cases of the no-phosphate treatments, where
the yield was reduced because of phosphorus deficiency, and that
of the high (12 percent) phosphate treatment, where a reduc-
tion in yield resulted because of an apparently unfavorable
phosphate-copper interrelationship, none of the 16 different com-
binations of N-P-K tested were, on the average, superior to any
of the others with respect to yield. The grades and sizes were
slightly better with the no-nitrogen combinations but the differ-
ence was probably insignificant. High phosphate produced high
cull counts because of ammoniation, although soil treatments
with copper had been made from time to time. This trouble
was readily corrected by a copper spray. The dropping of ma-
ture fruits was severe under the no-phosphate treatments and
generally somewhat heavier with the slowly available phosphate
sources than with superphosphate. Otherwise the slowly avail-
able phosphate sources were equal to superphosphate. Any dif-
ferences resulting from potash at the various rates or from
the various sources were insignificant.
In the past many citrus growers have hesitated to use muriate
of potash, particularly as the sole source of potash, contending
that it was deleterious to citrus. In this connection, observe in
Table 4, treatment 16 received all its potash from muriate for
seven years. The first several applications during this period were
at the rate of 24 percent K2O. Later it received four applications
of 48 percent K.O fertilizer at the rate of 8 to 12 pounds per
tree. This was far in excess of the rate of application of potash
ever used in commercial practice. The results here in no way
substantiated the contention that muriate was more toxic or
Fertilizer Experiimets with Citrus
in any way inferior to other potash fertilizer salts, at least as
far as citrus on the Davie soils was concerned.
Results of these experiments are in harmony with those one
might anticipate in the light of the nitrogen, phosphate and
potash content of these (Table 1) and similar virgin soils.
Conclusions
The practical aspects can be summed up briefly. On most
Davie soils (15 percent organic matter or over) nitrogen is not
normally a necessary fertilizer ingredient for citrus of bearing
age. Phosphate fertilization is essential to satisfactory produc-
tion over a prolonged period. Moderate amounts, from any of
the several sources tested here, give as good as or better results
than larger amounts. The essentiality of potash for citrus grow-
ing on these soils was not definitely established by these experi-
ments. The low level at which it would become a controlling
factor on production was evidently below the lowest rate (12
percent KO) at which it was employed in these treatments.
Acknowledgments
This bulletin is a report of investigations organized and initiated in 1933
and 1934, by Dr. R. V. Allison and Dr. J. R. Neller of the Everglades
Experiment Station, in cooperation with Flamingo Groves, Inc., and the
Florida Agricultural Research Institute. The authors are especially in-
debted to Mr. Floyd L. Wray, President, Flamingo Groves, Inc., who
provided the grove in which the experiments were conducted, together
with help for applying fertilizers, obtaining field records, and grade and
size data as the fruit was run through his packinghouse in Ft. Lauderdale.
A portion of the necessary funds for carrying on the work were furnished
by the Florida Agricultural Research Institute through the cooperation
of its manager, Mr. Frank L. Holland, who also made helpful suggestions
pertaining to the layout and operation of the experiments. From 1934 to
1940 the experiments were under the able direction of Dr. J. R. Neller
of the Everglades Experiment Station. Credit for the earlier detailed
harvest data and other experimental records is due chiefly to Dr. Neller.
Valuable assistance in supervising the treatments, collecting harvest data
and making laboratory analyses was given for several years by Mr. Len
S. Jones of the Everglades Experiment Station. To all these and to
others not mentioned here by name the authors extend their thanks.
Literature Cited
1. ALLISON, R. V. The importance of certain special elements in the
agriculture of South Florida. Fla. State Hort. Soc. Proc. 44: 11-22.
1931.
2. ALLWRIGHT, W. J. Progress report on fertilizer trials at Rustenburg.
W. Transvall Citrus Grower 45: 3-5, 7. 1936.
Fertilizer Experiimets with Citrus
in any way inferior to other potash fertilizer salts, at least as
far as citrus on the Davie soils was concerned.
Results of these experiments are in harmony with those one
might anticipate in the light of the nitrogen, phosphate and
potash content of these (Table 1) and similar virgin soils.
Conclusions
The practical aspects can be summed up briefly. On most
Davie soils (15 percent organic matter or over) nitrogen is not
normally a necessary fertilizer ingredient for citrus of bearing
age. Phosphate fertilization is essential to satisfactory produc-
tion over a prolonged period. Moderate amounts, from any of
the several sources tested here, give as good as or better results
than larger amounts. The essentiality of potash for citrus grow-
ing on these soils was not definitely established by these experi-
ments. The low level at which it would become a controlling
factor on production was evidently below the lowest rate (12
percent KO) at which it was employed in these treatments.
Acknowledgments
This bulletin is a report of investigations organized and initiated in 1933
and 1934, by Dr. R. V. Allison and Dr. J. R. Neller of the Everglades
Experiment Station, in cooperation with Flamingo Groves, Inc., and the
Florida Agricultural Research Institute. The authors are especially in-
debted to Mr. Floyd L. Wray, President, Flamingo Groves, Inc., who
provided the grove in which the experiments were conducted, together
with help for applying fertilizers, obtaining field records, and grade and
size data as the fruit was run through his packinghouse in Ft. Lauderdale.
A portion of the necessary funds for carrying on the work were furnished
by the Florida Agricultural Research Institute through the cooperation
of its manager, Mr. Frank L. Holland, who also made helpful suggestions
pertaining to the layout and operation of the experiments. From 1934 to
1940 the experiments were under the able direction of Dr. J. R. Neller
of the Everglades Experiment Station. Credit for the earlier detailed
harvest data and other experimental records is due chiefly to Dr. Neller.
Valuable assistance in supervising the treatments, collecting harvest data
and making laboratory analyses was given for several years by Mr. Len
S. Jones of the Everglades Experiment Station. To all these and to
others not mentioned here by name the authors extend their thanks.
Literature Cited
1. ALLISON, R. V. The importance of certain special elements in the
agriculture of South Florida. Fla. State Hort. Soc. Proc. 44: 11-22.
1931.
2. ALLWRIGHT, W. J. Progress report on fertilizer trials at Rustenburg.
W. Transvall Citrus Grower 45: 3-5, 7. 1936.
Fertilizer Experiimets with Citrus
in any way inferior to other potash fertilizer salts, at least as
far as citrus on the Davie soils was concerned.
Results of these experiments are in harmony with those one
might anticipate in the light of the nitrogen, phosphate and
potash content of these (Table 1) and similar virgin soils.
Conclusions
The practical aspects can be summed up briefly. On most
Davie soils (15 percent organic matter or over) nitrogen is not
normally a necessary fertilizer ingredient for citrus of bearing
age. Phosphate fertilization is essential to satisfactory produc-
tion over a prolonged period. Moderate amounts, from any of
the several sources tested here, give as good as or better results
than larger amounts. The essentiality of potash for citrus grow-
ing on these soils was not definitely established by these experi-
ments. The low level at which it would become a controlling
factor on production was evidently below the lowest rate (12
percent KO) at which it was employed in these treatments.
Acknowledgments
This bulletin is a report of investigations organized and initiated in 1933
and 1934, by Dr. R. V. Allison and Dr. J. R. Neller of the Everglades
Experiment Station, in cooperation with Flamingo Groves, Inc., and the
Florida Agricultural Research Institute. The authors are especially in-
debted to Mr. Floyd L. Wray, President, Flamingo Groves, Inc., who
provided the grove in which the experiments were conducted, together
with help for applying fertilizers, obtaining field records, and grade and
size data as the fruit was run through his packinghouse in Ft. Lauderdale.
A portion of the necessary funds for carrying on the work were furnished
by the Florida Agricultural Research Institute through the cooperation
of its manager, Mr. Frank L. Holland, who also made helpful suggestions
pertaining to the layout and operation of the experiments. From 1934 to
1940 the experiments were under the able direction of Dr. J. R. Neller
of the Everglades Experiment Station. Credit for the earlier detailed
harvest data and other experimental records is due chiefly to Dr. Neller.
Valuable assistance in supervising the treatments, collecting harvest data
and making laboratory analyses was given for several years by Mr. Len
S. Jones of the Everglades Experiment Station. To all these and to
others not mentioned here by name the authors extend their thanks.
Literature Cited
1. ALLISON, R. V. The importance of certain special elements in the
agriculture of South Florida. Fla. State Hort. Soc. Proc. 44: 11-22.
1931.
2. ALLWRIGHT, W. J. Progress report on fertilizer trials at Rustenburg.
W. Transvall Citrus Grower 45: 3-5, 7. 1936.
Florida Agricultural Experiment Station
3. ANDERSSEN, G. F. Citrus manuring-its effects on cropping and on
the composition and keeping quality of oranges. Jour. Pomol. and
Hort. Sci. 15: 117-159. 1937.
4. CAMP, A. F., and B. R. FUDGE. Some symptoms of citrus malnutrition
in Florida. Fla. Agr. Exp. Sta. Bul. 335: 1-55. 1939.
5. CHAPMAN, H. D., and S. M. BROWN. The effect of phosphorus deficiency
on citrus. Hilgardia 14: 161-181. 1941.
6. CHAPMAN, H. D., S. M. BROWN and D. S. RAYNER. Effects of potash
deficiency and excess on orange trees. Hilgardia 17: 619-643. 1947.
7. CHAPMAN, H. D., A. P. VANSELOW and G. F. LEBIG, JR. Production
of citrus mottle leaf. Jour. Agr. Res. 55: 365-379. 1913.
8. COLLINSON, S. E. Influence of soil and fertilizer on citrus fruits. Fla.
State Hort. Soc. Proc. 26: 168-172. 1913.
9. ERWIN, T. C. Interaction of copper and phosphorus in Norfolk fine
sand. Soil Sci. Soc. Fla. Proc. 7. 1945.
10. FORSEE, W. T., Jr., and R. V. ALLISON. Evidence of phosphorus inter-
ference in the assimilation of copper by citrus on the organic soils of
the lower East Coast of Florida. Soil Sci. Soc. Fla. Proc. 6. 1944.
11. FORSEE, W. T., Jr., and J. R. NELLER. Phosphate response in a Va-
lencia grove in the eastern Everglades. Fla. State Hort. Soc. Proc.
57: 110-115. 1944.
12. HAAS, A. R. C. Phosphorus deficiency in citrus. Soil Sci. 42: 93-118.
1936.
13. HAAS, A. R. C. Phosphorus nutrition of citrus and the beneficial effects
of aluminum. Soil Sci. 42: 187-202. 1936.
14. JAMISON, VERNON C. The effects of phosphate upon the fixation of
zinc and copper in several Florida soils. Fla. State Hort. Soc.
Proc. 56: 26-30. 1943.
15. KINNISON, A. F., and D. W. ALBERT. A progress report on fertilizer
studies with grapefruit in Salt River Valley, Arizona. Amer. Soc.
Hort. Sci. Proc. 33: 90-91. 1936.
16. LYON, T. L., and H. O. BUCKMAN. The nature and properties of soils,
p. 225. The Macmillan Co. 1937.
17. NELLER, J. R., and W. T. FORSEE, JR. Fertilizer experiments in an
orange grove in the eastern Everglades. Fla. State Hort. Soc.
Proc. 54: 1-4. 1941.
18. TAKAHASHI, I. Effects of phosphorus on citrus. Okitsu Hort. Soc.
Jour. 27: 18-30. 1931.
19. VAN DER PLANK, J. E., and F. A. S. TURNER. Are our sour oranges
due to lack of phosphorus? Farming in South Africa 11: 59-60.
1936.
20. WEBBER, HERBERT JOHN. Fertilization of the soil as affecting the
orange in health and disease. U. S. Dept. Agric., Yearbook Agr.,
pp. 193-202. 1894.
21. YOUNG, H. D. Effects of fertilizer on the composition and quality of
oranges. Jour. Agr. Res. 8: 127-138. 1917.
|