Relation of magnesium deficiency in grapefruit leaves to yield and chemical composiion of fruit

Material Information

Relation of magnesium deficiency in grapefruit leaves to yield and chemical composiion of fruit
Series Title:
Bulletin University of Florida. Agricultural Experiment Station
Fudge, B. R ( Bonnie Reid )
Place of Publication:
Gainesville Fla
University of Florida Agricultural Experiment Station
Publication Date:
Physical Description:
36 p. : ill., charts ; 23 cm.


Subjects / Keywords:
Grapefruit -- Composition ( lcsh )
Grapefruit -- Diseases and pests -- Florida ( lcsh )
Magnesium deficiency diseases ( lcsh )
bibliography ( marcgt )
non-fiction ( marcgt )


Bibliography: p. 36.
General Note:
Cover title.
Bulletin (University of Florida. Agricultural Experiment Station)
Statement of Responsibility:
by B.R. Fudge.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
027189315 ( ALEPH )
18214379 ( OCLC )
AEN5188 ( NOTIS )


This item has the following downloads:

Full Text

I1 Bulletin 331








Bulletins will be sent free to Florida residents upon request to

January, 1939


S/John J. Tigert, M.A., LL.D., President of
the University
S Wilmon Newell, D.Sc., Director
Harold Mowry, M.S.A., Asst. Dir., Research
J. Francis Cooper, M.S.A., Editor
Jefferson Thomas, Assistant Editor
Clyde Beale, B.J., Assistant Editor
Ida Keeling esap, Librarian
Ruby Ne\,L.ia, Administral e Myager
K. H. Graham, Business Manager
Rachel McQuarrie, Accountant

W. E. Stokes, M.S.., Agronomist'
W. A. Leukel, Ph.D., Agronomist
G. E. Ritchey, M.S., Associate2
Fred H. Hull, Ph.D., Associate
W. A. Carver, Ph.D., Associate
John P. Camp, M.S., Assistant
Roy E. Blaser, M.S., Assistant
A. L. Shealy, D.V.M., Animal Husbandman'
R. B. Becker, Ph.D., Dairy Husbandman
L. M. Thurston, Ph.D., Dairy Technologist
W. M. Neal, Ph.D., Asso. in Dairy Nutrition
D. A. Sanders, D.V.M., Veterinarian
M. W. Emmel, D.V.M., Veterinarian
N. R. Mehrhof, M.Agr., Poultry Husbandman
O. W. Anderson, M.S., Asst. Poultry Hush.
W. G. Kirk, Ph.D., Asst, An. husbandman
R. M. Crown, B.S.A., Asst. An. Husbandman
P. T. Dix Arnold, M.S.A., Assistant Dairy
L. L. Rusoff, M.S., Asst. in An. Nutrition"
R. V. Allison, Ph.D., Chemist'
R. M. Barnette, Ph.D., Chemist
F. B. Smith, Ph.D., Soil Microbiologist
C. E. Bell, Ph.D., Associate
R. B. French, Ph.D., Associate
H. W. Winsor, B.S.A., Assistant
J. Russell Henderson, M.S.A., Assistant
L. W. Gaddum, Ph.D., Biochemist
L. H. Rogers, M.A., Spectroscopic Analyst'
Richard A. Carrigan, B.S., Asst. Chemist
C. V. Noble, Ph.D., Agricultural Economist'
Bruce McKinley, A.B., B.S.A., Associate
Zach Savage, M.S.A., Associate
A. H. Spurlock, M.S.A., Assistant
Ouida Davis Abbott, Ph.D., Specialist'
Ruth Overstreet, R.N., Assistant
J. R. Watson, A.M., Entomologist'
A. N. Tissot, Ph.D., Associate
H. E. Bratley, M.S.A., Assistant
G. H. Blackmon, M.S.A., Horticulturist'
A. L. Stahl, Ph.D., Associate
F. S. Jamison, Ph.D., Truck Horticulturist
R. J. Wilmot, M.S.A., Spec. Fumigation Res.
R. D. Dickey, B.S.A., Assistant Horticulturist
J. Carlton Cain, B.S.A., Asst. Horticulturist
Victor F. Nettles, M.S.A., Asst. Hort.
W. B. Tisdale, Ph.D., Plant Pathologist'
George F. Weber, Ph. D., Plant Pathologist
R. K. Voorhees, M.S., Assistants
Erdmar West, M.S.. Mycologist
Lillian E. Arnold, M.S., Assistant Botanist


R. P. Terry, Chairman, Miami
Thomas W. Bryant, Lakeland
W. M. Palmer, Ocala
H. P. Adair, Jacksonville
Chas. P. Helfenstein, Live Oak
J. T. Diamond, Secretary, Tallahassee

L. O. Gratz, Ph.D., Plant Path. in Charge
R. Kincaid, Ph.D., Asso. Plant Pathologist
J. D. Warner, M.S., Agronomist
Jesse l.,eves, Farm Superintendent
A. F. Camp, Ph.D., Horticulturist in Charge
John H. Jefferies, Superintendent
Michael Peech, Ph.D., Soils Chemist
B. R. Fudge, Ph.D., Associate Chemist
W. L. Thompson, B.S., Asst. Entomologist
W. W. Lawless, B.S., Asst. Horticulturist
J. R. Neller, Ph.D., Biochemist in Charge
J. W. Wilson, Sc.D., Entomologist
F. D. Stevens, B.S., Sugarcane Agronomist.
Thomas Bregger, Ph.D., Sugarcane
Jos. R. Beckenbach, Ph.D., Asso. Horticul.
Frederick Boyd, Ph.D., Asst. Agronomist
G. R. Townsend, Ph.D., Asso. Plant Path.
R. W. Kidder, B.S., Animal Husbandman
W. T. Forsee, Ph.D., Asst. Chemist
B. S. Clayton, B.S.C.E., Drainage Engineer2
W. M. Fifield, M.S., Asst. Horticulturist
S. J. Lynch, B.S.A., Asst. Horticulturist
Geo. D. Ruehle, Ph.D., Asso. Plant Pathologist
W. F. Ward, M.S., Asst. An. Husbandman
in Charges

M. N. Walker, Ph.D., Plant Pathologist in
K. W. Loucks, M.S., Asst. Plant Pathologist
C. C. Goff, M.S., Assistant Entomologist
Plant City
A. N. Brooks, Ph.D., Plant Pathologist
R. N. Lobdell, M.S., Asst. Entomologist
A. S. Rhoads, Ph.D., Plant Pathologist
A. H. Eddins, Ph.D., Plant Pathologist
Samuel O. Hill, B.S., Asst. Entomologist'
David G. Kelbert, Asst. Plant Pathologist
R. W. Ruprecht, Ph.D., Chemist in Charge,
Celery Investigations
W. B. Shippy, Ph.D., Asso. Plant Pathologist
E. S. Ellison, Meteorologists
B. H. Moore, A.B., Asst. Meteorologist'

'Head of Department.
'In cooperation with U.S.D.A.
'On leave.





M ETHODS OF A NALYSIS ....................... .................................................. 6
R ESULTS ........... .. ...........................- .... .. ...... ......... ................. 7
Seed Production .................................... .. ..................................... 8
Composition of Whole Fruit and Seed ....... ................................ 9
Composition of Foliage in Relation to Bronzing ....................... 13
Tree Performance ........... ...... ..... ..... ............. 22
Alternation of Bearing ................... ......... ................... 26
DISCUSSION OF RESULTS ..................... ............ .......................... 32
SUM M ARY ..................... ........... ........ ..................-....... 34
LITERATURE CITED ........................ ....... ................... 36


It has been observed for a long time that there exists con-
siderable difference in fertilizer and cultural requirements
between seedy varieties of grapefruit and the so-called seedless
varieties such as Marsh Seedless. Although the greater vigor
and productivity of Marsh is quite common, growth and pro-
lificacy have been accentuated on light soils because of the
difficulties attendant upon growing the seedy varieties on such
soils. These observations have been widely recognized, although
there has been no explanation of the underlying causes.
To arrive at some explanation for the observed differences,
a study was started in 1934 of fruit, seed and foliage composi-
tion of Duncan, Excelsior, Walters and Marsh varieties of grape-
fruit. The block used in this experiment included trees budded
on rough lemon rootstock which were planted in 1925 and came
into bearing in 1930. Fertilizer programs for the four varieties,
as to both amounts and analyses, have been the same since the
trees were set. At present the Marsh trees are larger, more
vigorous and possess more abundant foliage than trees of the
other three varieties. Likewise, Marsh has produced the larg-
est crops for several years prior to and during the time this
study was made.
Foliage on trees of seedy varieties producing heavy crops
begins to "break" in color about mid-August as indicated by
the loss of green color (Fig. 1). The disappearance of the


Florida Agricultural Experiment Station

green color becomes more and more accentuated until the leaves
are practically devoid of green pigment except for a triangular
area of green extending from the base of the leaf up to the
apex of the triangle which lies on the leaf midrib (Fig. 2, B).
In severe cases the green area may entirely disappear (Fig. 2, C).
Observations made during three consecutive years indicate that
the various greenish-yellow patterns are progressive stages lead-
ing toward complete loss of green color in the leaves. Apparently
this yellowing process reaches a maximum about November 1.



Fig. 1.-Excelsior grapefruit tree showing a typical area of bronzed foliage which is
closely associated with fruit production. Note the absence of bronzing in areas where
little fruit is being produced.

Magnesium Deficiency in Grapefruit Leaves

After several weeks in this condition the leaves begin to drop
and severe loss of foliage results. The fruit will persist on
the defoliated limbs for a time and then drop, and, more
often than not, the weakened limb will die back to a point well
within the canopy of the tree. The loss of green color and
subsequent loss of leaves is commonly referred to as "crop
strain". Marsh trees are conspicuously free of the above condi-
tion, as are trees of the other three varieties when not bearing
a crop. The greenish-yellow leaf patterns of this type are being
specifically referred to as "bronzing", and this term will be used
throughout this paper in referring to the condition. The type
of bronzing produced in this grove is similar to type "B'' de-
scribed by Bahrt and Hughes (2).1

*^ -^ ^ .... *,i .......L- ..^(

..*'. .. .. .

Fig. 2.-Excelsior leaves showing normal condition (A) and two stages of advanced
bronzing (B and C). These leaves represent the chemical compositions of green and
bronzed leaves, respectively, given in Table 6.

The object of this study has been to find in the composition
of fruit, seed and leaves an explanation, if possible, for the dif-
ferences in the behavior of these varieties of grapefruit.

'Italic figures in parentheses refer to "Literature Cited" in the back
of this bulletin.

Florida Agricultural Experiment Station

The grapefruit trees are growing in rows, first Excelsior,
followed by Duncan, Marsh and Walters in the order named.
There are eight rows of a variety each containing 16 trees, or
a total of 128 trees of each variety in the experiment. Sampling
of the fruit was done in the packinghouse where each variety
was sized and graded separately. Fruit of size 70 was used
for this work and further selection of fruit in this size for
analytical samples was made for uniform weight and size.
Ashing of the dried plant material was done in an electric
muffle which was usually quite cool when the charge was placed
in it. The temperature was increased slowly until charring
was complete. The temperature of the muffle, which was regu-
lated with a pyrometer at 5000 C., was maintained for several
hours when a white fluffy ash was obtained. Whenever possible
colorimetric methods were employed, the values being taken
from deflection-concentration curves prepared from standards
using a photoelectric colorimeter.
Nitrates in Sap.-Phenoldisulphonic acid method according to
Frear (4) using sodium hydroxide as recommended by Harper
(5). Activated charcoal was found necessary to complete the
clearing (6) in order that the nitrate color could be obtained
free of off-tints. This method has shown the presence of large
quantities of nitrates in certain tissues.
Ammonia in Sap.-The method given by Schlenker (14) with
subsequent Nesslerization was used.
Total Nitrogen Including Nitrate-Nitrogen.-Modification of
the official salicylic acid-thiosulphate method by Ranker (12).
Phosphorus. 1:2:4:aminonaphtholsulphonic acid method
with 60 percent perchloric acid (8).
Potassium.-Sodium cobaltinitrite precipitation followed by
color development with choline hydrochloride and potassium
ferrocyanide (10).
Calcium.-Official method (1) using sintered glass crucibles
instead of filter paper.
Magnesium.-8-hydroxyquinoline precipitation in ammoniacal
solution (7) followed by the colorimetric determination of the
phenolic properties of the 8-hydroxyquinoline-magnesium com-
plex (15).
Iron. -, c'-bipyridine as used by McFarlane (9) except
that cobalt nitrate was not used as standard. The values of

Magnesium Deficiency in Grapefruit Leaves

the unknowns were taken from a deflection-concentration curve
carefully prepared from known amounts of iron.
Manganese.-The Richards (13) modification of the Willard
and Greathouse method (16) was used.

Fertilizer applied 1930-35, inclusive, was composed entirely
of inorganic nitrogen, superphosphate and muriate of potash.
The nitrogen was derived from nitrate of soda and/or sulfate
of ammonia. With only a few exceptions the formula used
consisted of 4-6-5 mixtures which were applied in the spring,
summer and fall. During this period nitrogen was officially
expressed on the ammonia (NH3) basis. On the present
nitrogen (N) basis this formula would represent a 3.3-6-5
mixture. The three applications of fertilizer in 1936 and the
spring application of 1937 consisted of 50 percent organic nitro-
gen mixtures. Throughout this entire period, 1930 to 1937,
none of the so-called secondary elements was applied separately
or as part of the mixed fertilizer except as these may have
been contained in very small amounts in the materials used
in the mixtures. Unpublished soil studies by Dr. M. Peech
show that this Norfolk fine sand soil which has received no
magnesium with an acid inorganic fertilizer program contains
in the surface six inches of soil nine pounds of exchangeable
magnesium per acre. This amount is considered to be extremely
low. Copper, lime and sulfur have been used to some extent
in the form of sprays and dusts for the control of diseases
and insects.
Although data comprising this study have been collected on
the 1934-35, 1935-36 and 1936-37 crop years, actual data pre-
sented in this paper constitute those of the 1936-37 crop year.
Since concordant results have been obtained for the three-year
period, it is not deemed necessary to present the work of each
year. However, the average data for the three years are pre-
sented in two tables for comparison with results of 1936-37.
This period in the production of the trees represents a heavy
production year, 1934-35; a light production year, 1935-36; and
another heavy production year, 1936-37.
Quality of the fruit in the light production year, 1935-36,
was poor compared to that of the other two years. This was
reflected in the fruit composition by somewhat higher nitrogen

Florida Agricultural Experiment Station

and lower calcium content. Likewise, the nitrogen content of
the seed was higher in 1935-36. Otherwise the composition
of the fruit and seed has been quite consistent throughout the
three-year period.
To determine the relative number of seeds produced and to
obtain adequate seed material for analysis, 100 fruits, size 70,
were selected from each variety. After these samples were
weighed the seeds were removed, counted, cleaned with dilute
alcohol and weighed. The data of Table 1 show the physical
aspect of seed production. The mean number of seed per fruit
of the three seedy varieties varied from 45.5 to 55.6 as com-
pared with only 3.0 for Marsh and constitute about 2.4 percent
of the weight of seedy fruit and 0.18 percent of the Marsh
fruit. Hereafter the term "seedy varieties" will be used when
referring collectively to Duncan, Excelsior and Walters.

Duncan Excelsior Walters Marsh

No. of fruit used .................. 100 100 100 100
Green wt. of fruit (gms.) ....... 52,537.0 51,688.0 50,540.0 50,717.0
Aver. wt. of each fruit (gms.) 525.37 516.88 505.40 507.17
Green wt. of seed in
100 fruits (gms.) ................ 1,250.0 1,238.0 1,289.5 91.65
Dry wt. of seed in
100 fruits (gms.) ................ 629.0 669.1 682.1 43.85
No. of seed in 100 fruit .---.......... 4,745 4,502 5,522 304
47.80 45.555 55.65 3.04
Mean no. seed per fruit .......... --0.664 -0.698 0.750 0.128
Green wt. seed in ea. fruit (%) 2.38 2.39 2.55 0.18
Dry matter in seed (%) ........ 50.32 54.04 52.90 47.85

The difference in seed production between seedy and seedless
varieties is apparently basic to any explanation for the wide
variation in growth and yield responses of these varieties under
field conditions. Production of seed is often accompanied by
appreciable translocation of food from leaves and stems of the
plant to fruit and seed. Unquestionably translocation is a major

Magnesium Deficiency in Grapefruit Leaves

internal process in citrus and especially in seedy varieties which
tend to produce fruit in clusters. The Walters variety shows
the greatest mean number of seed per fruit. It is interesting
to note that Camp and Jefferies (3) found Walters to. be the
least responsive of the four varieties as indicated by yield and
growth records. They found that this particular strain of
Walters, however, does not fit the original description of the
Walters variety.
The term "crop strain" indicates that the supply of elements
and organic foods are inadequate for producing fruit and at
the same time maintaining a healthy tree condition. Analyses
of whole fruit are given in Table 2 to show the comparative
nitrogen and mineral requirements of seedy and seedless grape-
fruit varieties. Percentages of dry matter, total nitrogen and
ash are higher in seedy than seedless fruit. The elements that
were determined in the ash are expressed as percent of the dry
matter and of the ash. All mineral elements determined, with
the exception of aluminum, are higher in seedy than seedless
fruit. Due to the lower ash content of Marsh fruit, the mineral
elements expressed as percentages of the total ash do not reflect
the differences that actually exist in these varieties. Although
the differences in percentage composition are small, it is appar-
ent that these, expressed as a function of the dry matter, reflect
significant differences in a given quantity of seedy and seedless
Analyses of the respective seed, Table 3, show that seed of
seedy varieties are higher in dry matter, ash, potassium and
magnesium, and are lower in calcium and iron than seed of
Marsh. The other elements vary only slightly. Here again the
differences in percentage composition are small. However, when
these data are calculated to show total amounts of these elements
in seeds of the respective fruits the differences become very
great because of the much greater number of seeds contained
in each fruit of the seedy varieties. The high percentage of
dry matter in seeds indicates that they are largely storehouses
of organic food for the young embryos. This is further indi-
cated by the fact that the percent total nitrogen, phosphorus,
magnesium, iron and manganese are approximately double those
of the corresponding whole fruit. The percent ash in seed is
much lower than that in whole fruit because of the large amount


Dry Tot. N. Tot. Ash P. (%) K. (%) Ca. (%) Mr. (%) II Fe. (%) Al. (%) Mn. (%)
Matter (% Dry (% Dry D [ _-lI II 1tI
_____Var Matter) M e Matter Ash MMatter I Ash Matter Ash IMatter Ash iMatter Ash Matter Ash
Duncan ...... 11.51 1.09 5.32 0.152 2.86 2.07 38.93 0.492 9.25 0.112 2.10 0.0081 0.153 0.0105 0.197 0.00053 0.0099
Excelsior .... 10.77 1.09 5.2 0.138 2.59 1.96 36.87 0.617 11.62 0.117 2.21 0.0058 0.109 0.001 0.024 0.00053 0.0099
Walters ...... 11.41 1.16 5.72 0.156 2.72 2.15 37.52 0.607 10.62 0.128 2.23 0.0037 0.064 0.0009 0.015 0.00061 0.0106
Marsh ........ 9.63 0.95 4.53 0.129 2.84 1.71 37.7 0.458 10.11 0.103 2.27 0.00 0.079 0.0057 0.125 0.00041 0.0090
Three-Year Average 1934-37*
Duncan .. 12.50 1.19 5.21 0.165 3.18 2.18 41.88 0.490 9.38 0.116 2.23 0.0044 0.085 0.0080 0.152 0.00043 0.0081
Excelsior .... 11.71 1.21 5.21 0.152 2.93 2.18 41.87 0.560 10.74 0.117 2.24 0.0044 0.083 0.0070 0.134 0.00048 0.0092
Marsh ........ 9.97 1.06 4.54 0.137 3.05 [[ 2.02 44.60 0.410 0.108 2.40 0.0027 0.060 0.0057 0.130 | 0.00038 0.0084

Iry i II MnII .( II M.
DryVariety Tot. N. Tot. Ash P. (%) K. (%) Ca. () Mg. (%) II Fe. (%) Al. () 1 Mn. ()
SVariety Matter (% Dry (% Dry I
% Matter) Matter)il Dry [Dry [Dry I Dry I Dry Dy Dry
iMatt) Matter Ash IMatter Ash iMatter Ash [Matter Ash [IMatterI Ash Matter I Ash Matter IAsh M
Dunca .... II I[ I I' I. 0.0007
Duncan 50.08 2.46 2.98 0.249 8.36 0.924 31.06 0.279 9.38 I 0.198 6.66 0.0056 0.188 0.0007 0.023 0.0007 0.0248
Excelsior 54.04 2.42 3.05 0.276 9.03 0.939 30.78 0.311 10.19 0.199 6.53 0.0164 0.538 0.0004 0.013 0.0009 0.0288
Walters 52.90 2.45 2.98 0.274 9.20 0.892 29.96 0.294 9.88 0.198 6.64 0.0167 0.561 0.0002 0.006 0.0008 0.0255
Marsh ........ 47.85 2.48 2.77 0.284' 10.25 0.760 24.45 0.339 12.24 110.188 6.77 0.0235 0.849 0.00061 0.021 0.0008 0.0281
Three-Year Average 1934-37*
Duncan .... 51.44 2.55 3.17 0.274 8.48 0.893 29.19 0.290 9.25 0.189 6.09 0.0053 0.172 0.0060 0.186 00008 0.0251
Excelsior .... 53.03 2.57 3.27 0.276 8.56 II 0.890 27.94 0.321 9.88 0.195 6.11 0.0100 3 0. 0.0020. 0 .0 0.0 0.0009 0.0270
Marsh ........ 49.95 2.53 3.14 0.302 9.78 | 0.787 25.53 0.316 10.30 0.201 6.54 0.0135 0.463 I 0.0062 0.186 0.0007 0.0235
*Results for only 1936-37 available on Walters' variety.

Magnesium Deficiency in Grapefruit Leaves

of organic foods in the seeds-fats, oils and proteins. The seeds
are lower in potassium, calcium and aluminum than the whole
fruit. This change in the ratio of minerals found in the seeds
as compared with the ratio in whole fruit indicates that the
seeds exert a selective absorption of minerals. The low ash
content of seed may be due partly to the large reduction in
potassium and calcium which constitute 45 to 50 percent of the
total ash of whole fruit. It is interesting to note that the
calcium content of Duncan seed, like that of the whole fruit
(Table 2), is lower than that of other varieties.
That a comparison may be made of the composition of a unit
quantity of fruit, data of Table 4 have been compiled to show
the quantities of principal elements removed from the tree in
100 whole fruits, size 70, and amounts of these elements removed
in the seed. The data of Tables 1, 2 and 3 were used as a basis
for this computation. Although fresh weights of Duncan and
Excelsior are slightly greater than that of Marsh, these differ-
ences are small as compared with the differences in dry weights
and in the elements determined. The data show that the quan-
tities of these elements removed in 100 whole fruits are much
greater in seedy Varieties than in Marsh, the percentage increase
as compared with Marsh varying from 22 to 60 percent.
Of these total quantities removed in the whole fruits of these
varieties, the amounts removed in the respective seeds are pre-
sented to give a measure of the effect of seed production and
to show the magnitude of the quantities in the seeds. It is
obvious that the larger amounts of total dry matter produced
in the seeds of seedy fruit are due to greater numbers of seed
which they contain and not to any large difference in the per-
centage composition of the seed as shown in Table 3. The values
obtained by multiplying the quantities of the elements found
in Marsh seed by 15 approximately equals the amounts in the
seeds of seedy varieties. The data of Table 1 show that there
are about 15 times as many seeds produced in seedy fruit as
in Marsh.
Subtracting the quantities of these elements contained in the
seeds from the whole fruit gives the amounts in the pulp and
rind. Fruits of seedy varieties, after exclusion of the seed,
still contain greater quantities of these elements than Marsh.
However, the differences have become greatly reduced. For
instance, differences in nitrogen are of the same magnitude as
the differences in dry matter. The mineral elements have not



Weight of 100 mature fruit (gms.) .............................................. 52,537.0
Dry weight of 100 mature fruit (gms.) ..................................... 6,047.0
Total nitrogen in 100 mature fruit (gms.) ................................. 65.912
Total phosphorus in 100 mature fruit (gms.) ................................ 9.204
Total potassium in 100 mature fruit (gms.) ................................. 125.173
Total calcium in 100 mature fruit (gms.) ....................................... 29.751
Total magnesium in 100 mature fruit (gms.) ............................... 6.773
Total manganese in 100 mature fruit (gms.) ................................ 0.032

Dry weight of seeds in 100 mature fruit (gms.) ....- .- ... 629.0
Total nitrogen in seeds in 100 mature fruit (gms.) .................... 15.473
Total phosphorus in seeds in 100 mature fruit (gms.) ...........- 1.754
Total potassium in seeds in 100 mature fruit (gms.) ....................... 5.812
Total calcium in seeds in 100 mature fruit (gms.) ................. 1.755
Total magnesium in seeds in 100 mature fruit (gms.) .........-.. 1.245
Total manganese in seeds in 100 mature fruit (gms.) .................... 0.0044


nitrogen in 100 fruits excluding seeds (gms.) ........................
phosphorus in 100 fruits excluding seeds (gms.) ....................
potassium in 100 fruits excluding seeds (gms.) ...........- -
calcium in 100 fruits excluding seeds (gms.) ...................
magnesium in 100 fruits excluding seeds (gms.) .................
manganese in 100 fruits excluding seeds (gms.) ....................














Magnesium Deficiency in Grapefruit Leaves

been proportionately reduced since the differences in mineral
composition between seedy and Marsh fruits after excluding
the seeds are greater than the differences in dry matter. Potas-
sium and calcium have not been greatly reduced because the
seed requirement of these elements is comparatively low. Al-
though the amount of magnesium removed in fruit is small as
compared with nitrogen, potassium and calcium, and of about
the same magnitude as phosphorus, it may very likely become
the "limiting element" where the soil supply is low and where
it is not being added in the fertilizer in the manner that nitro-
gen, phosphorus, potassium and calcium are applied.
In October of the 1936-37 season, which was a heavy crop
year, the trees of all three seedy varieties were showing ex-
treme effects of what has been called "crop strain". The dis-
coloration or bronzing of the foliage on a particular limb, as
already described, was definitely associated with the amount
of fruit being produced (Fig. 1) and the more abundant the
fruit the more complete had become the disappearance of green
color (Fig. 2, B and C). Samples of green and bronzed leaves
were collected and analyzed, the green leaves being taken off
limbs which were not bearing fruit. In the case of Marsh,
which rarely produces bronzed leaves, analogous samples were
taken "near the fruit" and "away from the fruit". The object
of the analysis was to determine the effect of fruit production
upon composition of the foliage and the change in composition
that gives rise to the bronzed leaf condition. The leaf sample
was macerated in a food chopper and thoroughly mixed before
removing the 250.0 gram aliquot that was used for sap extrac-
tion after having been frozen at -16 to -18 C. The sap was
extracted by a hydraulic press. The amount of sap obtained
and its partial analysis are given in Table 5.
The differences of ammonia in the sap are very slight, espe-
cially in the light .of extreme differences in appearance and
condition of the leaves. No nitrates could be detected even in
green leaves. After the heavy summer rains and before the
fall application of fertilizer is a period during which a low
level of nitrate-nitrogen exists in the trees as shown by previous
work on the Nitrogen Source Grove (unpublished). Apparently
the assimilation of nitrates is as rapid as the level of absorption
will permit at this time of year. The total nitrogen (N) in the

OCTOBER 27, 1936.

Bronzed I Green

Bronzed I Green

Bronzed I Green

Green I Green

I Leaves I Leaves I Leaves I Leaves Leaves | Leaves Leaves*I Leaves

Dry matter (%) ........................ ........................... 35.00 37.20 34.40 36.70 35.30 37.75 36.00 36.30

Ammonia in sap (p.p.m.) .......................................... 6.10 6.00 5.75 5.20 6.15 5.90 5.00 7.20

Nitrates in sap (p.p.m.) .............................................. none none none none none none none none
Total nitrogen in sap (mgm/cc) .............................. 3.96 3.59 3.84 3.24 3.86 3.84 3.30 3.30
Phosphorus in sap (p.p.m.) ........................................ 310 255 215 290 210 235 245 295

Potassium in sap (p.p.m.) .......................................... 12300 9600 12100 10300 10500 9900 10200 10600

Total amount of sap extracted from 250.0 gms.
green tissue (cc) ............................................... 92.0 92.0 88.0 96.0 91.0 92.0 103.0 94.0

*Leaves near fruit which correspond to the location of bronzed leaves of seedy varieties.

Magnesium Deficiency in Grapefruit Leaves

sap is presented as milligrams per cubic centimeter (mgm/cc)
of sap. These values for nitrogen can be expressed as p.p.m.
by multiplying by 1,000. It is evident that the nitrogen in the
sap of seedy varieties is higher in the bronzed leaves than in
the green leaves, whereas there is no difference in the sap
nitrogen of Marsh leaves. In the growth and function of normal
citrus shoots, it has been found that high total nitrogen in the
leaf sap reflects a high total nitrogen in the dried leaf tissue
and that approximately 26 to 28 percent of the total nitrogen
in the dried tissue is soluble nitrogen, i. e., sap nitrogen. That
this is not the condition found in these leaves will be shown
presently. The concentration of elements in the sap fluctuates
widely as a result of changes in the available nutrient supply.
For this reason sap analysis reflects the effects of fertilizer
programs to a greater degree than will total analysis.
Phosphate phosphorus shows a mixed trend; that is, Duncan
bronzed leaves are higher in phosphorus than the green leaves,
while the other three varieties including Marsh are showing the
reverse condition. Therefore, the weight of evidence indicates
that green leaves are higher in phosphates in the sap. The
potassium content of sap of bronzed leaves is definitely higher
than that of the respective green leaves of seedy varieties. The
difference iii Marsh leaves, although small, indicates that the
leaves away from the fruit are slightly higher in potassium.
This element has never been identified as a part of any organic
complex in plant tissue and in some cases 100 percent of the
potassium found in the plant is in the sap. Apparently, a great
majority of it stays in solution as an inorganic salt. Approxi-
mately 80 percent of the total potassium in citrus leaves can
be computed from sap analysis. In young maturing leaves the
amount in the sap is nearer 100 percent.
An aliquot of the macerated leaf tissue was weighed for mois-
ture determination and the remainder was dried, ground and
used for total nitrogen and total mineral analysis. Results of
the total analyses of leaves are given in Table 6. In every case
green leaves contain appreciably more dry matter than analogous
bronzed leaves. Since these leaf samples are all of the same
age and flush of growth, it may be assumed that the bronzed
leaves probably have had removed from them such soluble solids
that may be transported to the fruit, causing the low percentage
of dry matter. Results for total nitrogen lend support to this
contention because of the much lower total nitrogen content in

Variety and Dry Tot. N. Tot. Ash P- K. Ca. Mg. Fe. Al. II Mn.
Foliage Matter (% Dry (% Dry _____ II II
Condition % Matter) Matter) Dry i Dry Dry Dry I Dry | Dry I IDry
ii_ IMatter I Ash I Matter Ash IIMatter Ash IMatter I Ash I Matter I Ash IMatter I Ash i Matter I Ash
Bronzed 35.00 1.71 14.53 0.104 0.71 3.22 22.16 3.48 23.95 0.017 0.11 0.041 0.282 0.0097 0.066 0.0027 0.019

A Green 37.20 2.17 15.30 0.109 0.71 2.56 16.76 4.11 26.84 I 0.156 1.02 0.032 0.207 0.0076 0.049 0.0033 0.022

. Bronzed 34.40 1.72 15.02 0.110 0.73 3.52 23.44 3.70 24.63 0.014 0.09 0.022 0.143 0.0106 0.070 0.0024 0.016

SGreen 36.70 2.15 13.93 0.140 1.01 2.78 19.99 3.84 27.59 0.208 1.49 0.016 0.116 0.0087 0.062 0.0033 0.024

SBronzed 35.30 1.69 13.75 0.105 0.76 3.04 22.10 3.74 27.21 I 0.013 0.09 0.019 0.140 0.0088 0.064 0.0029 0.021

Green 37.75 2.25 14.29 0.123 0.86 2.52 17.66 4.17 29.18 0.143 1.00 0.018 0.124 I 0.0049 0.034 0.0040 0.028
-I- __ __ 1 I _1 IIT i _

36.00 1.92 14.10

36.30 2.23 13.99

*Leaves near fruit.

0.104 I 0.74

0.131 0.93

3.56 25.28

3.59 25.66

0.87 0.019 0.132 0.0072 0.051 0.0028

1.63 0.018 0.128 I 0.0090 0.064 0.0033





Magnesium Deficiency in Grapefruit Leaves

bronzed leaves or leaves near the fruit in the case of Marsh,
Green leaves of the four varieties show about the same level of
nitrogen. This might be expected since the fertilization program
has been uniform. On the other hand, bronzed leaves of seedy
varieties show a lower nitrogen content than Marsh leaves borne
in the analogous position with reference to the fruit. This indi-
cates very definitely that the production of fruit, especially
large quantities of seeds found in seedy varieties, draws upon
the leaf nitrogen. In this connection, leaf sap analyses (Table
5) show the opposite condition; namely, bronzed leaves contain
more nitrogen in the sap. This high soluble nitrogen is asso-
ciated with a low total nitrogen in the leaves. This condition
is construed to mean that the production of fruit, and especially
of large quantities' of seeds, is drawing upon the leaf nitrogen.
The high soluble nitrogen could not be the result of increased
absorption by the root system because (1) the green leaves
do not show it, (2) there are no free nitrates in the sap and
(3), as already pointed out, the supplying power of the soil is
particularly low in October.
Percent total ash shows no definite relationship to leaf condi-
tion. This may be due, as is apparent from the data, to some
elements being high in green leaves and others high in bronzed
leaves. The differences are too small to be considered signifi-
cant. The ash elements determined are expressed as percentages
of the dry matter and of the total ash. Phosphorus is slightly
higher in green leaves than in bronzed leaves. This is in agree-
ment with sap analysis previously given (Table 5). Phosphorus
is much lower in leaves near the fruit regardless of whether
or not they are bronzed. This indicates that phosphorus is not
a factor in producing the bronzed foliage although it is materially
reduced in leaves in the vicinity of fruit. Thus the green leaves
of Marsh near the fruit are just as low in phosphorus as the
bronzed leaves of seedy varieties.
The behavior of potassium is opposite to that of phosphorus;
that is, bronzed leaves and Marsh leaves near the fruit are
higher in potassium than green leaves away from the immediate
influence of fruit production. Bronzed leaves of seedy varieties
show a particularly high potassium content as compared with
green leaves. The same condition is indicated by the sap analy-
ses already presented. This relationship is not construed to
mean that high potassium is the cause of bronzing. A probable
explanation is that the selective removal of other elements and

Florida Agricultural Experiment Station

organic materials from bronzed leaves causes the potassium to
appear as a higher percentage of the dry matter and of the ash.
For instance, the increase in potassium in bronzed leaves over
green leaves is almost the exact counterpart of the decrease in
calcium. The fact that potassium, being the highest ash element
in whole fruit, is not being reduced in leaves near the fruit
indicates that it is not a deficiency factor in the production of
crop strain or bronzing. Since potassium is highly soluble and
has never been found tied up as a part of any organic com-
pound in plants, it may be considered to possess the ability to
move directly through the conducting system and into the fruit
without passing through the elaborating leaf mechanism. If
this assumption is true, it would become possible to explain
the lack of a reduction of potassium in leaves near fruit in
spite of the high content of potassium in the fruit.
The calcium content of bronzed leaves is about 10 percent
lower than that in analogous green leaves of seedy varieties.
The variation in Marsh is considered insignificant. The calcium
content of leaves is highest of all ash elements, whereas, the
potassium content is highest in whole fruit. These two elements
constitute approximately 45 percent of the ash of leaves.
Greatest significant difference in elements thus far deter-
mined was found in magnesium. Green leaves of seedy varieties
contain 10 to 15 times more magnesium than analogous bronzed
leaves. Even Marsh leaves near the fruit which do not lose
their green color contain about half as much magnesium as the
green leaves away from the fruit. Also, Marsh leaves near
the fruit contain almost as much magnesium as green leaves
of seedy varieties away from the effect of crop production. It
may be assumed that Marsh leaves would become bronzed like
leaves of seedy varieties when the magnesium content reached
a level as low as that found in bronzed foliage. Just what the
level of magnesium is at which bronzing occurs has not yet been
determined. However, it is at some percentage below the 0.122
percent found in green Marsh leaves near the fruit. Largest
amounts of magnesium were always found in Marsh leaves
located away from the effects of fruit production-this quantity
is perhaps none too high. If it may be assumed that the Marsh
leaves near the fruit which contain 0.122 percent magnesium
at one time contained approximately 0.228 percent magnesium,
the difference, 0.106 percent, represents roughly the amount
required of the leaves in the production of the Marsh fruit. It

Magnesium Deficiency in Grapefruit Leaves

has been established by analysis that the amount of magnesium
required to produce seedy fruit is greater than for Marsh.
Therefore, by subtracting the quantity, 0.106, from the average
amount of magnesium found in green leaves of seedy varieties,
the difference (0.063 percent) obtained represents the level of
magnesium that should be in bronzed leaves of seedy varieties
if the requirement were no greater than for Marsh. This dif-
ference (0.063 percent) is greater than the averaged percentage
(0.015) of magnesium found in bronzed leaves by actual analysis.
The averaged analysis of foliage of seedy varieties indicates
that fruit production caused the magnesium content of the
leaves to be reduced from 0.169 percent to 0.015 percent. The
difference, 0.154 percent, represents the change caused by crop
production. Thus, crop production caused a reduction in the
magnesium content of foliage of seedy varieties equal to 91.1
percent of the amount in leaves not influenced by fruit produc-
tion. The analogous reduction in Marsh was equal to 46.5
percent. Although these figures are only relative approxima-
tions, they indicate that bronzing can be prevented in seedy
varieties when the magnesium requirement for fruit production
can be satisfied without reducing the magnesium content of
leaves below 0.12 percent, which is the lowest amount deter-
mined to date in a green leaf showing no bronzing-the critical
point may be below this figure.
Iron and aluminum are slightly higher -in bronzed leaves
than in green leaves of seedy varieties. The difference between
the Marsh foliage from the two locations on the tree is probably
not significant. Manganese is consistently lower in leaves near
the fruit in all varieties and the amounts of manganese in
bronzed leaves and Marsh green leaves near the fruit are of
the same magnitude. Therefore, these data do not indicate
that manganese is an important factor in bronzing. Obviously,
to what extent the three last-mentioned elements may appear
as limiting factors in the nutrition of these grapefruit varieties
depends upon building the magnesium supply to a point where
it is no longer the primary limiting factor. The same is perhaps
true of all the mineral elements in their practical application.
It is apparent from the data for whole fruit, seed and foliage
composition already presented that magnesium is very closely
associated with the bronzed leaf condition. In order to show
that the variation is truly in the magnesium, the ratios of
calcium (which is fairly constant) to magnesium are presented

Florida Agricultural Experiment Station

in Table 7. The ratios Ca/Mg are approximately 5.0 in whole
fruit and 1.5 in the seed. No significance is attached to the
observed variations in the whole fruit and seed ratios since
these have been found to vary somewhat from year to year.
The narrow ratio in the seed has been brought about by a reduc-
tion of calcium in the seed on the one hand and a sharp rise
in magnesium on the other when compared with the whole fruit
composition. The ratios of green and bronzed foliage of seedy
varieties show the very great differences that exist. The high
ratio values of bronzed leaves are not due to an increase in
calcium since the data (Table 6) show calcium decreases, but
to pronounced decreases in the magnesium content. The foliage
of Marsh Seedless near the fruit does not show the wide ratio
found in the bronzed leaves of seedy varieties. In fact, the
Ca/Mg ratio in Marsh leaves near the fruit (29.2) is of about
the same magnitude as that in green leaves of two of the seedy
varieties. The deficiency of magnesium in the leaves in relation
to crop production is apparent from these ratios. Obviously,
the production of large quantities of seeds in seedy varieties
is one of the factors augmenting the deficiency.

Ratio Ca/Mg
Whole Seed Foliage
Fruit Bronzed I Green
Duncan 4.41 1.41 210.1 26.3
Seedy Excelsior 5.26 1.56 273.6 18.5
Walters 4.75 1.49 299.0 29.1

Seedless Marsh 4.45 1.81 29.2* 15.7

*Marsh foliage not bronzed, but taken "near the fruit".

Since bronzing is the most pronounced expression of the
effects of crop strain, the average compositions of bronzed
and green leaves, respectively, of seedy varieties are compared
as percent differences with the analogous composition of Marsh
foliage; that is, the composition of bronzed leaves is compared

Magnesium Deficiency in Grapefruit Leaves

with that of Marsh leaves "near the fruit" and the green leaves
of seedy varieties is compared with that of Marsh leaves "away
from the fruit". Since Marsh did not show the bronzed leaf
symptom it serves as a logical base for comparison as shown
in Fig. 3 in which the average composition of leaves of seedy
varieties is referred to the composition of analogous Marsh
leaves as percent differences. Quantities of these elements in
leaves of seedy varieties of more than plus or minus 10 percent
are considered significant variation from Marsh composition.
Analyses of Marsh leaves "near the fruit" and "away from the
fruit", respectively, are represented by the straight line drawn
through the zero point. The percentage differences in com-
position of the respective leaves of seedy varieties are shown
as plus or minus quantities, depending upon whether a particular


- 20








1~Sw ~n.

Bronzed Leaves

Green Leaves


Fig. 3.-The percentage differences in average compositions of bronzed and green leaves
of seedy varieties as compared with those of Marsh Seedless "near the fruit" and "away
from the fruit", respectively.

Florida Agricultural Experiment Station

element is greater (+) or smaller (-) in quantity than that
in respective Marsh leaves. The first column of each pair repre-
sents the percentage differences shown by bronzed leaves as
compared with Marsh leaves "near the fruit"; the second column,
the differences shown by green leaves of seedy varieties as
compared with green leaves of Marsh away from the direct
influence of fruit production.
Potassium in bronzed leaves and calcium in green leaves are
the only elements that are significantly greater in leaves of
seedy varieties; whereas, nitrogen in bronzed leaves and mag-
nesium in bronzed and green leaves of seedy varieties are signifi-
cantly lower than in Marsh. The large deficit of magnesium in
the leaves of seedy varieties, especially bronzed leaves, over-
shadows the variations of all other elements determined, the
green leaves of seedy varieties containing approximately 25
percent less magnesium than green leaves of Marsh. This may
indicate that trees of seedy varieties are becoming more rapidly
depleted of magnesium than Marsh as a result of crop produc-
tion since 1930 when they reached bearing age. The magnesium
content of bronzed leaves is 87 percent less than the amount
in Marsh leaves near the fruit. This offers conclusive proof
that the wide ratios presented in Table 7 are the result of the
very low percentage of magnesium in bronzed leaves and not
to an increase in the calcium.
Yield records constitute the most important indication of tree
performance because they are a measure of tree efficiency. The
mean tree yields of the three seedy varieties are given for com-
parison with Marsh over a period of four consecutive years,
the last three of which constitute the period during which the
analytical studies were made. Results presented in Table 8
show that the first year is comparatively low yielding followed
consecutively by a high, another low and another high yielding
year. There are 128 trees of each variety in this block. How-
ever, the number of trees used in compiling the mean yields is
somewhat less, due to the elimination of all cut-back, reset
and diseased trees. It is apparent that the tendency for alter-
nate bearing is more pronounced in seedy varieties even though
the mean yields as compared with Marsh were less in each of
the four years, there being a small degree of alternation shown
,by Marsh. It is evident that production cost per box of Marsh




(119 trees)

(122 trees)

Walters Marsh
(107 trees) (126 trees)

Mean yield (Ibs.) per tree-1933-34 ................ 73.63 5.85 113.52 6.51 52.45 3.63 163.30 6.02

Mean yield (lbs.) per tree-1934-35 .................. 275.95 7.33 234.22 8.38 176.52 7.35 321.43 6.82

Mean yield (lbs.) per tree-1935-36 .................. 134.33 8.90 154.10 9.24 88.90 7.26 275.60 13.11

Mean yield (lbs.) per tree-1936-37 .................. 299.79 11.28 278.28 13.29 276.40 7.80 469.28 14.83


Duncan Excelsior I Walters Marsh

Total nitrogen removed in fruit produced per acre, 70 trees (lbs.).... 26.327 22.869 25.606 30.051

Total phosphorus removed in fruit produced per acre, 70 trees (lbs.) 3.675 2.891 3.437 4.074

Total potassium removed in fruit produced per acre, 70 trees (Ibs.) 50.001 41.118 47.355 54.124

Total calcium removed in fruit produced per acre, 70 trees (lbs.) .... 11.886 12.943 13.398 14.490

Total magnesium removed in fruit produced per acre, 70 trees (Ibs.) 2.702 2.457 2.828 3.255

Total manganese removed in fruit produced per acre, 70 trees (Ibs.) 0.0126 0.0112 0.0133 0.0133





Florida Agricultural Experiment Station

is less than that of seedy varieties due to the larger number
of boxes produced each year during this four-year period. The
errors of the means of seedy varieties are somewhat greater
than that of Marsh. The reason for this will be shown later.
Having already determined that a unit quantity of fruit of
the seedy varieties contains greater amounts of the principal
elements as compared with Marsh, the data of Table 10 have
been compiled to show the comparative amounts of these ele-
ments removed in the 1936-37 mean yield of fruit per tree. The
data in this table give a measure of the ability of the tree to
convert these elements into fruit. In other words, the figures
represent-the mean output of these elements per tree. Although
it has been shown that the percentage composition of Marsh
fruit is lower than that of seedy varieties, it is obvious that
greater quantities of these elements are removed from the Marsh
trees as a result of the much greater mean yields. However,
the amounts are not in proportion to the greater yield for
reasons already given. The relative amounts of these elements
removed in the corresponding seed are interesting in that they
are greater in seedy varieties than in Marsh. Thus, the total
dry matter of seed produced in fruit of a Marsh tree is still
less than that in analogous seedy fruit in spite of the yield
of whole fruit of Marsh being 39.3 percent greater. When the
amounts of these elements in the seed are deducted from those
of the corresponding elements in the whole fruit, the differences
between the seedy and seedless fruits become more accentuated
and represent more nearly the differences in mean yield.
Taking the 1936-37 mean yield per tree as a basis, quantities
of N, P, K, Ca, Mg and Mn removed in the whole fruit produced
per acre of 70 trees were calculated. The results (Table 9)
give a measure of the loss of these elements per acre due to
cropping. The Marsh tree, due to the greater mean yield, has
had removed from it more of these elements per acre of 70
trees than any of the seedy varieties. It has been shown that
the Marsh trees have consistently yielded larger crops for sev-
eral years and during the same time shown better physical
condition and growth with no bronzing. Thus, it has been im-
possible to find an element or elements removed in greater quan-
tities in the seedy varieties than in Marsh. Fertilizer records
show that the amounts of N, P, K and Ca added in fertilizers
have been more than adequate to take care of this annual loss.
However, this is not true of magnesium which has not been


Duncan Excelsior Walters Marsh

Number of trees ........................................ ..................... 119 122 107 126
Mean total yield per tree in 1936-37 (lbs.) ................................... 299.79 278.28 276.40 469.28
Total dry matter in fruit produced per tree (lbs.) ................... 34.5058 29.9707 31.5372 45.1916
Total nitrogen removed in fruit produced per tree (lbs.) .............. 0.3761 0.3267 0.3658 0.4293
Total phosphorus removed in fruit produced per tree (lbs.) ........ 0.0525 0.0413 0.0491 0.0582
Total potassium removed in fruit produced per tree (lbs.) ....... 0.7143 0.5874 0.6765 0.7732
Total calcium removed in fruit produced per tree (lbs.) ........... 0.1698 0.1849 0.1914 0.2070
Total magnesium removed in fruit produced per tree (Ibs.) ....... 0.0386 0.0351 0.0404 0.0465
Total manganese removed in fruit produced per tree (lbs.) ....... 0.00018 0.00016 0.00019 0.00019

Total dry matter in seed produced per tree (lbs.) ........................... 3.5732 3.5941 3.7285 0.4042
Total nitrogen removed in seed produced per tree (Ibs.) .............. 0.0879 0.0870 0.0913 0.0100
Total phosphorus removed in seed produced per tree (lbs.) ....... 0.0100 0.0099 0.0102 0.0011
Total potassium removed in seed produced per tree (lbs.) ........... 0.0330 0.0337 0.0333 0.0031
Total calcium removed in seed produced per tree (lbs.) ............. 0.0100 0.0112 0.0110 0.0014
Total magnesium removed in seed produced per tree (lbs.) .......... 0.0071 0.0072 0.0074 0.00075
Total manganese removed in seed produced per tree (lbs.) .... 0.000025 0.000032 0.000030 0.000003

Total nitrogen removed in fruit excluding seed (lbs.) .................... 0.2882 0.2397 0.2745 0.4193
Total phosphorus removed in fruit excluding seed (lbs.) ............... 0.0425 0.0314 0.0389 0.0571
Total potassium removed in fruit excluding seed (lbs.) .............. 0.6813 0.5537 0.6432 0.7701
Total calcium removed in fruit excluding seed (lbs.) ...................... 0.1598 0.1737 0.1804 0.2056
Total magnesium removed in fruit excluding seed (lbs.) ................ 0.0315 0.0279 0.0330 0.04575
Total manganese removed in fruit excluding seed (lbs.) ................ 0.000155 0.000128 0.000160 0.000187
1 1

Florida Agricultural Experiment Station

added to the fertilizer. The results show that approximately
3.0 pounds of magnesium are removed annually in fruit from
an acre of 70 trees. This amount of magnesium represents
about one-third of the exchangeable or available magnesium in
this Norfolk sandy soil.
Nitrogen and potassium are lost by cropping in the approxi-
mate ratio of one unit of nitrogen to two of potassium. This
ratio is about the same as the ratio in which these two elements
are generally applied to the trees as fertilizers. In this partic-
ular case the ratio of N to K has been slightly less than 1:2.
The fertilizer formula, as already noted, was a 3.3-6-5 since
1930 and through to the completion of this work. The ratio
of phosphorus to N and K is far below the ratios in which
it is applied as fertilizers.
It should be remembered that these figures are only for fruit
requirements. Amounts of these elements needed for tree
growth, amounts which are fixed in the soil, and amounts lost
through leaching are not included.
The individual tree yield records afford an excellent oppor-
tunity to study the alternate bearing habit of these varieties
in the light of the analytical data already presented. This
behavior may be considered to be a form of exhaustion which
is due to a deficiency of one or more elements that are very
necessary to tree maintenance and crop production. The term
exhaustion is not used here in the sense of fatigue or over-
production. For instance, where there is a severe shortage
of magnesium it is possible to have bronzing and alternation
of bearing associated with medium crops. Therefore, alter-
nation of bearing may be defined as the normal response of
trees to these limitations in the supply of elements resulting
from cropping. In fact, individual tree yield records indicate
that trees, after an exceptionally heavy crop, may require two
years instead of one to rehabilitate themselves to a condition
where another heavy crop is produced.
Although the mean yield data (Table 8) show the tendency
of alternation in bearing of seedy varieties, the full extent of
this alternation is not apparent because the trees were not all
yielding high or low in the same year. The mean yield data
of all the trees cover the high and low yielding trees of a
particular variety for any given year, and the alternation or

Magnesium Deficiency in Grapefruit Leaves

variation from year to year of the mean yield is due to the pre-
domination of either the low or the high yielding trees. For
this reason the mean yields show a large mean error even
though more than 100 trees were used in the calculations on
each variety.
The true picture of alternate bearing is best illustrated by
individual tree yield records, some of which are given in Table
11 and were taken from the records of Duncan and Excelsior.




Yield (lbs.)
1934-35 1 1935-36


EXCELSIOR E-8 0 300 39 429
E-9 0 302 42 397
Low yielding E-10 5 303 0 471
1933-34 Q-9 17 419 45 606
CC-15 7 318 6 548

A-15 406 40 421 135
High yielding I-11 319 44 394 2
1933-34 Q-4 420 40 649 10
Q-6 228 1 315 54
E-12 211 21 488 66

B-8 12 314 35 399
DUNCAN F-4 8 355 72 450
F-10 3 411 48 447
Low yielding J-6 0 260 4 317
1933-34 R-7 1 312 35 460

B-2 382 153 306 41
High yielding F-13 140 84 393 0
1933-34 F-14 362 24 667 43
N-15 430 20 701 1

There are two groups purposely given of each variety; namely,
a low yielding group in 1933-34 and a high yielding group in
the same year. These two groups of each variety represent
only a part of the total tree population (Table 8) used in cal-
culation of mean yields, and were selected from the low and
the high yielding groups shown in Fig. 4. The behavior of the
trees in these two groups of each variety in the subsequent
years is the same except that when one group is yielding low
the other is yielding high. Although seasonal conditions exert

Florida Agricultural Experiment Station

500 -


300 -H-igh

/ / p
400 -


1933-34 1934-35 1935-36 1936-37

Fig. 4.-Yield responses of high and low yielding groups selected in 1933-34 of Duncan
and Excelsior, respectively, and a comparison of the yields of these groups over a 4-year
period with the mean yields of all trees of each variety.

an influence upon yield, weather is not considered a factor con-
tributing to these variations. It is not a question of setting
the bloom because, after a heavy crop, the tree fails even to
bloom. Apparently the amount of fruit that any individual
grapefruit tree will produce in any given year is most closely
associated with the size of the crop the previous year. It is
logical to assume that the heavy crop produced on a tree in
any particular year causes the limitations on the crop the next
year. This indicates that the previous crop has removed from
the tree an element, or elements, vital to the mechanism pro-
ducing the next crop. The results of Duncan and Excelsior
are shown in Figure 4 in relation to the mean yields of the two
varieties. Relative positions of the three points to each other
in each of the four years indicate which of the two groups is
predominating. It is evident that the low yielding group of
trees in 1933-34 is the predominating group because the alter-
nation of the mean yields of both varieties is greatly influenced
by this group as shown by the relative positions of the three
points for each of the four years. Thus by breaking down the
mean yields in this manner a better picture is obtained of tree

Magnesium Deficiency in Grapefruit Leaves 29

performance and a better interpretation of yield data in field
plot work is given. Of course a large number of trees of each
variety produced yields intermediate to these two groups. How-
ever, the predominating group (low or high yielding) exerts a
very definite influence upon the mean yield. To illustrate, if
conventional fertilizer plots were laid out in this grove, it would
scarcely be possible to have the same number of low and high
yielding trees in one plot or its duplicate. The curves of Fig-
ure 4 show that the predominance of either the high or the low
yielding trees in any year could readily obscure the good or
bad effects of a fertilizer treatment in its comparison with
adjacent plots.
Data have already been presented which show that the actual
quantities of the elements removed in the fruit are the initial
cause of bronzing under field conditions where there is a low
available supply of these elements in the soil. Since a deficiency
of magnesium has been found in bronzed leaves, the most
prominent symptom of crop strain, it appears that the high
requirement of seedy fruit for magnesium causes a depletion
of the supply in the leaves. Therefore, bronzing impairs the
leaf area of the tree by causing a loss of foliage and general
inefficiency of that which remains to such an extent that a
poor physiological condition is created. The tree is slow to
recover because the assimilating mechanism-the leaves-is
impaired. The data of Table 6 show that the normal green
leaves of seedy varieties contain 15 times as much magnesium
as bronzed leaves. Due to the low supply of available mag-
nesium, the tree is unable to make rapid recovery or to prevent
the bronzed condition that destroys its efficiency. Therefore,
alternation of bearing in these trees is the result of low tree
Individual tree yields of two groups of Marsh trees are given
in Table 12 to illustrate the yielding behavior of Marsh and
for comparison with similar groups of trees representing seedy
varieties (Table 11). The selection of trees was based upon the
1933-34 tree yields and upon the 1933-34 mean yield (163.3 lbs.)
of all Marsh trees. The first group is representative of trees
yielding distinctly less than the 1933-34 mean yield; and the
second group, much larger yields than the 1933-34 mean yield.
The difference in the average yield of these two groups in
1933-34 was approximately 200 pounds of fruit per tree. There-
fore, these groups of trees in the subsequent years should show

Florida Agricultural Experiment Station

the effect of the 1933-34 crop size upon yields and alternation
should occur if they respond like trees of seedy varieties. The
results indicate that there is no alternation of bearing just as
there are no visual indications of crop strain and bronzing in
the trees. It is recognized that climatic factors can and do
affect alternation of bearing, but these have been effectively
eliminated in these comparisons.


Tree Yield (Ibs.)
_Number 1933-34 1 1934-35 I 1935-36 1936-37
C-10 106 281 470 368
MARSH G-13 57 240 373 254
SEEDLESS K-4 62 419 358 558
K-10 82 360 446 509
0-8 105 215 290 421
Yield less than I S-9 59 306 319 445
1933-34 mean S-13. 79 460 430 608
W-5 91 345 272 472

C-6 278 453 318 424
C-11 288 225 358 360
C-15 285 245 384 274
Yield more than G-14 260 398 283 368
1933-34 mean 0-2 259 420 310 707
W-7 288 364 585 520
AA-8 258 308 61 640
AA-16 280 298 311 598

The relationship of yields of all the trees of these two groups
to the mean yield of all the Marsh trees is shown in Figure 5.
Although both groups represent only a few trees, they show
the absence of alternate bearing as influenced by crop produc-
tion and give good agreement with the mean yield of all trees.
These results show that Marsh, by its greater efficiency, is
better adapted to the Norfolk sandy soils of Florida. This point
should be recognized by growers in the selection of varieties for
future plantings. A comparison of Figures 4 and 5 shows the
wide difference in the behavior of trees of seedy and seedless
varieties during the four-year period. If alternate bearing were
entirely the result of the loss of elements through cropping,
the Marsh trees should have shown extreme alternate bearing
because of the greater mean yield which resulted in larger quan-
tities of all elements being converted into fruit and removed

Magnesium Deficiency in Grapefruit Leaves

from the tree. Remembering that the fertilizer applications
have been uniform, observed differences must be due either to
greater inherent efficiency in Marsh or to inefficiency in seedy
varieties. If the yield behavior of Marsh may be considered
normal, the alternate bearing of trees of seedy varieties is a
condition of inefficiency resulting from bronzing of foliage which
may drop in severe cases, leaving a thin canopy of leaves and
considerable dead wood. The effects of bronzing and the ap-
pearance of dead wood in trees of seedy varieties annually
since 1930 have resulted in the better appearance and larger
size of Marsh trees at present. The level of vigor in these trees
is closely associated with the abundance of normal appearing
foliage; in comparison with trees of seedy varieties, the amount
of normal foliage on Marsh trees is proportionately greater than
either tree size or mean yield. Therefore, the maintenance of
abundant foliage is prerequisite to efficient tree performance.
Alternation of bearing of trees of seedy varieties is to a large
extent the result of a low ratio of efficient leaf area to tree size
caused by bronzing of foliage. For this reason, supplying these
nutrient deficiencies and protecting the foliage from ravages of
diseases and insects are essential to economic citrus culture.

500 -

400 Marsh

300 -

200 7


1933-34 1934-35 1935-36 1936-37

Fig. 5.-Yield responses of high and low yielding groups of Marsh trees selected in
1933-34 and a comparison of the yields of these groups over a 4-year period with the mean
yields of all Marsh trees.

Florida Agricultural Experiment Station

In comparing the appearance of bronzing as it occurs in the
field with the chemical composition of the fruit and foliage, it
is apparent that a deficiency of magnesium is a primary cause
of this type of bronzing. The close relationship of bronzing
to fruit position on a limb (Fig. 1) indicates that the mag-
nesium requirement of fruit induces the deficiency found in
bronzed leaves. The evidence that bronzing has occurred only
in the foliage of seedy varieties indicates that seeds are the
part of the fruit that causes the difference in the behavior of
seedy varieties as compared with Marsh Seedless, which pro-
duces few seeds and the foliage of which has not become bronzed.
The fact that foliage of trees of seedy varieties not producing
a crop does not become bronzed is further evidence that bronz-
ing is the "crop strain" effect of low available magnesium.
Chemical analyses have shown that seeds have a higher require-
ment for magnesium than the rest of the fruit. The actual
requirement depends more upon the number of seeds produced
than upon the magnesium percentage of seeds. The poor growth
and yield response of Walters as compared with Duncan and
Excelsior is logically explained by the fact that this variety
produces about 22 percent more seed than the other two.
The clustering habit of seedy varieties causes a somewhat
low ratio of leaves to each fruit. However, the clustering ap-
pears accentuated where the canopy of foliage is thin as in the
trees of these seedy varieties. A much more desirable distribu-
tion of fruit is found on trees having a dense canopy of foliage.
Analyses of foliage have shown that bronzed leaves are lower
in nitrogen, phosphorus, calcium, magnesium and manganese,
and higher in potassium, iron and aluminum than green leaves
of the same variety and age. This condition is considered to
be the result of crop production. Parberry (11), working on
chlorosis of orange leaves in New South Wales, found the same
relationships of calcium, potassium and magnesium. As in this
work, he found that the greatest significant difference was in
the greater magnesium content of normal leaves, but did not
state the conditions of crop production.
Crop production depletes the magnesium supply in soils low
in available magnesium to the extent that bronzing occurs and
alternate bearing results. Thus it has been possible to show
that the probable crop any tree of the seedy varieties will pro-

Magnesium Deficiency in Grapefruit Leaves

duce depends upon the size of the crop the previous year which
controls the degree of "crop strain" or bronzing. This is not
true of Marsh even though the mean yield per tree has been
about 40 percent larger than those of seedy varieties in each
of the four years. For this reason the amount of magnesium
removed from the Marsh tree has been greater than that re-
moved from any of the trees of seedy varieties. At the same
time the percentage composition of Marsh leaves is higher in
magnesium than that of leaves of seedy varieties due to a
greater density of foliage and a uniform distribution of fruit
on the tree.
The amounts of N-P-K applied in the spring, summer and
fall applications of fertilizer in 1936 have been computed from
analysis. The sum of these three applications represents the
total N-P-K applied that year as available nutrients for the pro-
duction of the 1936-37 crop, the analysis of which is presented
in this paper. From the records of fertilizer applied and the
data of Table 10, the amounts of N-P-K converted into fruit
in 1936 were determined. Slightly less than 50 percent of the
nitrogen and potassium applied in fertilizer was removed from
the grove in the form of fruit. Only 6 percent of the phos-
phorus applied could be accounted for in the fruit produced.
The amounts of N-P-K taken into the vegetative parts of the
tree and that lost through soil fixation and leaching may account
for the remainder of the N-P-K. Although Marsh trees have
produced 40 percent greater mean yields, the density of the
fruit on the Marsh trees in relation to the' leaf area is less
than that on trees of seedy varieties. For this reason the
trees of seedy varieties have not been as efficient as Marsh in
absorbing and converting N-P-K into fruit.
Under any conditions of production and prices, the citrus
grower is interested in maintaining a low cost of production
per box of fruit. The majority of the many items that make
up the production costs are under the direct control of the
operator or grower. However, the item of tree efficiency as
discussed in these data is one that is only indirectly under
the control of the grower. Physiological diseases caused by
nutrient deficiencies that impair the leaf system of a tree create
a low tree efficiency which tends to increase the per box costs
because yield is reduced. For instance, bronzing in the seedy
varieties of grapefruit creates an inefficient leaf area per tree
which results in a low assimilation rate. Thus, the elements

Florida Agricultural Experiment Station

that are applied, N-P-K, are not used efficiently. It is obvious
that maximum yield is the most important single item that
lowers the costs per box. This item is closely associated with
the physiological condition of the tree and is the one under the
least direct control of the grower.
Apparently the practical solution of this problem lies in the
application of magnesium to the soil for the purpose of building
a greater leaf surface per tree. Thus the loss per tree of mag-
nesium in fruit production has not been so serious as the loss
of foliage due to bronzing. Although the percentage composi-
tion of magnesium in the leaves may be increased by applications
of magnesium, the real physiological benefit of magnesium will
come in more abundant foliage through the prevention of bronz-
ing. A greater tree efficiency will be the result and alternation
of bearing will be minimized where the available magnesium
is adequate for both crop production and tree maintenance.

Results obtained in a three-year study of yield and chemical
composition of the fruit and foliage of four varieties of grape-
fruit are as follows:
1. The two chief characteristics distinguishing seedy and
seedless varieties are the greater production of seed and the..
clustering of fruit in seedy varieties. The seedy varieties studied
contain about 50 seed per fruit while Marsh produces about
2. Seedy fruit are higher than Marsh in percentage composi-
tion of nitrogen and minerals. This is partially due to the
larger number of seed contained in the former. After exclud-
ing the quantities of the elements contained in the seed, the
fruits of seedy varieties are still slightly richer in nitrogen
and minerals.
3. The chemical composition of foliage indicates that crop
production causes a significant reduction in nitrogen, phos-
phorus, calcium, magnesium and manganese of leaves in the
vicinity of fruit. The effects are more pronounced in seedy
than in seedless varieties; and, of these elements, the reduc-
tion of magnesium is the most significant.
4. Bronzing of foliage in this grove has appeared only in
seedy varieties and is closely associated with crop production.
The most pronounced difference in the composition of green

Magnesium Deficiency in Grapefruit Leaves

leaves and bronzed leaves of the same variety is in the amount
of magnesium. The bronzed leaves contain approximately one-
tenth the amount of magnesium found in green leaves of the
same variety and age. The amount of magnesium found in
Marsh leaves is greater than the quantities found in leaves of
seedy varieties. Marsh leaves have not become bronzed because
the foliage is abundant and the loss of magnesium from the
leaves is not excessive. However, Marsh leaves near the fruit
contain about one-half as much magnesium as leaves located
away from the direct influence of fruit production. Apparently
the level of available magnesium at which bronzing will occur
in Marsh foliage is below that at which foliage of seedy vari-
eties become bronzed.
5. Under the influence of crop production which began in
1930, the trees of seedy varieties have become depleted of mag-
nesium to the extent that bronzing is severe. Heavy crops
cause partial defoliation and dying back of limbs. Extreme
alternation of bearing occurs because of the slow rate of tree
recovery from the effects of crop production and especially from
6. Marsh trees in the experimental grove have not bronzed,
have maintained abundant foliage, grown rapidly and produced
more fruit since reaching bearing age than any of the seedy
varieties. Although the percentage compositions of the ele-
ments determined are lower in Marsh fruit, the trees have
converted into fruit larger amounts of these elements, includ-
ing magnesium, because of the larger mean yields per tree.
Therefore, Marsh is particularly well adapted to the poorer soil
types because of its habits of growth and efficient fruit pro-
7. Bronzing and consequent loss of foliage in seedy varieties
are the results of a low supply of available magnesium com-
bined with its removal in crop production. The poor physical
condition of trees of seedy varieties after the production of a
heavy crop causes the alternation of bearing habit. This ex-
treme alternation of bearing found in individual trees of seedy
varieties did not occur in Marsh trees. The actual removal of
magnesium is secondary to the physiological effect of bronzing
and subsequent loss of foliage in causing alternate bearing.
8. These results indicate that applications of magnesium bear-
ing fertilizer will prevent bronzing and thereby reduce alternate
bearing to a minimum.

Florida Agricultural Experiment Station


1. Association of Official Agricultural Chemists. Official and tentative
methods of analysis. Fourth edition. 1935.
2. BAHRT, G. M., and A. E. HUGHES. Soil fertility and experiments on
bronzing of citrus. Proceedings of Fla. State Hort. Society. 1937.
3. CAMP, A. F., and J. H. JEFFERIES. A comparison of seedless and seedy
grapefruit varieties. Citrus Industry, 18: 1: 5. 1937.
4. FREAR, D. E. Estimation of nitrate nitrogen in plant juice. A study
of the expression and clarification of the juice. Plant, Physiology
5: 359. 1930.
5. HARPER, H. J. The accurate determination of nitrates in soils. Ind.
and Eng. Chem. 16: 180-183. 1924.
6. HILL, H. H. Plant juice clarification for nitrate nitrogen determina-
tions. Science 71: 540. 1930.
7. HILLEBRAND, W. F., and G. E. F. LUNDELL. Applied inorganic analysis.
John Wiley & Sons. 1929.
8. KING, E. J. The colorimetric determination of phosphorus. Biochem.
Jour. 26: 292. 1932.
9. MCFARLANE, W. D. Determination of iron by titanium titration and
by ba, cc'-bipyridine colorimetry. Ind. and Eng. Chem. 8: 124. 1936.

10. MORRIS, V. H., and R. W. GERDELL. Rapid colorimetric determination
of potassium in plant tissues. Plant Physiology 8: 315. 1933.
11. PARBERY, N. H. Mineral constituents in relation to chlorosis of orange
leaves. Soil Science 39: 35. 1935.
12. RANKER, E. R. Determination of total nitrogen, nitrate nitrogen and
total nitrogen not including nitrate nitrogen: Further observations
on a modification of the official salicylic-thiosulfate method. Annals
of Missouri Bot. Garden 13: 391. 1926.
13. RICHARDS, M. B. The colorimetric determination of manganese in
biological material. Analyst 55: 554. 1930.
14. SCHLENKER, F. S. Comparison of existing methods for the determina-
tion of ammonia nitrogen and their adaptability to plant juice.
Plant Physiology 7: 685. 1932.
15. SNELL, F. D., and C. T. SNELL. Colorimetric methods of analysis.
D. Van Nostrand Co. 1936.

16. WILLARD, H. H., and L. GREATHOUSE. Colorimetric determination of
manganese by oxidation with periodate. Jour. Am. Chem. Soc.
39: 2366. 1917.