NFREC Res. Rpt. 94-16
SOURCE OF P, LIME APPLICA TION
RESIDUAL EFFECTS ON P
A AVAILABILITY TO WA TERMELON
F. M. Rhoads and S. M. Olson
NORTH FLORIDA RESEARCH AND
EDUCA TION CENTER,
Florida Agricultural Experiment Stations
Instititute of Food and Agricultural Sciences
University of Florida Gainesville
SITY OF 4
-L IJLA..L -.
Institute of Food and Agricultural Sciences
The value of watermelon [Citrullus lanatus (Thunb.) Matsum. &
Nakai] production in Florida was reported at more than $62 million
in 1991-92. Watermelon has a relatively high phosphorus (P)
requirement. Availability of P from triple superphosphate (TSP) to
some crops was reported to be greater than from diammonium
phosphate (DAP). Also, lime and P application sequence was
reported to influence availability of DAP but residual
effects have not been investigated. Our objectives were to
determine the influence of P source and lime and P application
sequence on P availability to watermelon and determine the
availability of residual P with respect to P source and lime
treatment. Two crops of watermelon were grown consecutively in the
same soil in pots containing 2 kg (4.4 lb) of soil each. The first
planting was made in the fall and the second in the spring. The Ap
horizon of a Norfolk loamy sand (fine-loamy, siliceous, thermic,
Typic Kandiudult) containing <5.0 mg kg'' Mehlich-1 extractable P
was used. Rates of P were 40 and 80 mg kg'1 (80 and 160 Ib acre"')
from either TSP or DAP. Calcium oxide was added at a rate of 1 g
kg1' (2000 lb acre"') either 4 wk before or 4 wk after P
application. The experiment consisted of two P sources, two P
rates, three lime treatments and a control with no lime or P added.
There were four replicates and the analysis of variance was
conducted on the 2x2x3 factorial treatment arrangement. Dry-matter
yield and P content of watermelon tissue were determined.
Availability of P from TSP was 23% greater than with DAP in the
first crop (immediate effect) and 14% greater in the second crop
(residual effect), averaged over P rates and lime treatments, as
shown by analysis of variance. Without lime, availability of P was
about 23% greater with TSP than with DAP in both crops. Liming
before P application increased P availability by 17% in the first
crop and by 22% in the second crop over the control (no lime).
Liming after P application decreased P availability by 12% in the
first crop, but it increased P availability by 11% in the second
In Florida, watermelons are important to the agricultural
economy of the state, having a value of more than $62 million in
1991-1992 (Freie and Young, 1993). Most growers plant their
watermelon crop on land taken out of pasture or on land cleared
prior to planting to avoid plant diseases that reduce yield and
quality. Where soil-test phosphorus (P) is low, about 70 lb acre"'
is recommended for watermelon production (Hanlon et al., 1990).
A total of more than 3 million lb of P would be required for
watermelons in Florida if all 53,000 acres planted in 1991-1992
(Freie and Young, 1993) received 70 lb of P acre'1 Considerable
savings in fertilizer costs could be realized over a period of
several years if growers used the most efficient source of P in
terms of fraction available to plants. Information on the effect
of P source on watermelon response is limited (Locascio et al.,
1968). Effects of lime and P application sequence on P
availability have only been investigated on a limited number of
crops. Therefore, studies showing the influence of lime and P
application sequence on P availability to watermelon and similar
crops were not found in the literature.
Phosphorus availability to snap bean from diammonium phosphate
(DAP) has been shown to be less than from triple superphosphate
(TSP) on an acid coastal plain soil (Rhoads, 1991). Lime and P
application sequence had strikingly different effects on P
availability to snap bean in comparisons between TSP and DAP
(Rhoads et al., 1993). Lime application had no effect on P
availability from TSP, however, lime applied 4 wk before P
increased availability of DAP to a greater extent than lime applied
4 wk after P (69% vs 26%). Other researchers have shown increased
crop growth and/or P availability when lime was applied before P in
comparison to lime applied after P with various sources of P (Singh
and Seatz, 1961; Soltanpour et al., 1974; and Wang and Yuan, 1989).
Phosphorus uptake by plants is used as an index of P availability
in this report.
Objectives of this research were to (1) determine the
influence of P source and sequence of lime and P application on P
availability to watermelon and (2) determine the availability of
residual P with respect to source and lime treatment.
MATERIALS AND METHODS
Two crops of watermelon plants were grown consecutively in the
same soil at Quincy, Florida in greenhouse pots; one during the
fall of 1992 and the second in the spring of 1993. Each pot
contained 2 kg (4.4 lb) of soil material from the Ap horizon of a
Norfolk loamy sand (fine-loamy siliceous, thermic, Typic
Kandiudult) containing <5mg P kg'' (less than 5 parts per million).
Phosphorus sources were triple superphosphate containing 20% P and
diammonium phosphate which contains 23% P and 20% nitrogen (N).
Rates of P were 40 and 80 mg kg'1 (80 and 160 lb acre'). Lime
source was calcium oxide (CaO) applied at the rate of 1 g kg'' (2000
lb acre"1) either 4 wk before P application or 4 wk after P
application. Lime and P were applied only to the first crop as
was 0.25 g of K kg'" (500 lb of K acre-1) of soil as K2SO4 .
Ammonium nitrate was applied to each crop at the rate of 128 mg N
kg'' (256 Ib of N acre"') of soil without regard for the N contained
The cultivar for both crops was 'Crimson Sweet'. Six seeds
were planted in each pot and plants were thinned to 2 per pot after
emergence. The fall crop was planted September 30 and harvested 6
wk later, while the spring crop was planted April 1 and harvested
8 wk later. The longer growing period for the second crop allowed
time for greater expression of residual effects. Before planting
each crop, soil-water content was brought up to 250 mL per 2 kg of
air-dry soil to approximate field capacity. Seed for the first
crop were planted 4wk after water, lime, and P were applied.
During each growing season only 100 mL of tapwater pot"' was added
each time the soil surface was dry to avoid water logged
conditions. Soil was moistend (water content brought up to 250 mL
per 2 kg of air-dry soil) and allowed to dry three times between
crops to simulate wetting and drying cycles that occur in the
field. The purpose of two crops was to show immediate response to
lime and P application in the first and response to residual lime
and P in the second crop.
Both shoots and roots were harvested at the end of each test
and dry-weight was determined after drying to constant weight at
70 OC. One g of ground plant material was ashed at 500 OC and taken
up in dilute HC1. Phosphorus was determined by the molybdenum blue
method. Total P uptake was determined from dry-weight and P
concentration of shoots and roots. Both dry-weight of shoots and
* total P uptake were reported as a percent of the treatment
producing the maximum amount of dry-matter or P uptake in each crop
for convenience in making comparisons between experiments.
The experimental design consisted of a completely randomized
block with treatments arranged factorially in 4 replicates. There
were two P rates, two P sources, and three lime treatments for a
total of 12 treatments plus a control having no lime or P. A three
factor analysis of variance was used to evaluate main effects and
interactions between factors (Steel and Torrie, 1960). The error
mean square (ems) and student's t (2.036) for 33 degrees of freedom
were used to calculate LSD 05g's as follows: LSD=2.036(2ems/24)1/2 for
P sources and P rates, LSD=2.036(2ems/16)1/2 for lime treatments,
and LSD=2.036(2ems/4)1/2 for the twelve individual treatments. The
* control treatment was not used in the analysis of variance
procedures, but was included to show magnitude of P response in
RESULTS AND DISCUSSION
Factors significantly (P s 0.01) influencing plant growth
(dry-matter yield) and total P uptake were P sources, P rate, and
lime treatments (Table 1). The interaction of P rate by P source
influenced both plant growth (P < 0.01) and total P uptake (P <
0.05) but only in the first crop. Interaction between lime
treatment and P source influenced plant growth in both crops but
not total P uptake. Actual maximum yield of the first crop was
about 40% of the maximum for the second crop. Increased yield of
the second crop was attributed to a longer growth period (6 wk vs
8 wk) and longer days in the last part of the growth period
(October vs May).
Plant growth, expressed as percent of maximum, was 21% greater
in the first crop and 13% greater in the second crop with TSP than
DAP, averaged over P rates and lime treatments (Table 2). The P
rate of 80 mg kg'1 produced 26% more dry-weight in both the first
and. second crop than the 40 mg kg'1 rate when averaged over P
sources and lime treatments. Interaction between P sources and P
rates in the first crop is shown by a yield increase of 34% with
TSP and only 18% with DAP between 40 and 80 mg P kg''. A lack of
interaction between P rate and P source in the second crop was
demonstrated by similar yield increases between 40 and 80 mg P kg"
with each P source, i. e. 25% with TSP and 27% with DAP. These
observations indicate that differences in watermelon response
between P sources were less with residual P than with P application
to the immediate crop.
The significant interaction between lime treatments and P
sources indicates that plant response to lime with TSP was
different from that with DAP. Liming the first crop 4 wk before
applying TSP did not significantly (P > 0.05) influence plant
growth compared with no lime (Table 2). However,liming the first
crop 4 wk after TSP application decreased dry-matter yield by
[1/2(62+100-52-87)] 11% (LSD.05= 9.8%). But lime had no effect on
plant growth with TSP in the second crop. Liming 4 wk before DAP
application increased dry-matter yield by [1/2(67+84-41-68)] 21%
(LSDo.05= 9.8%) in the first crop and [1/2(81+99-28-73)] 39% (LSD0.05=
14.2%) in the second crop. However, liming 4 wk after DAP
application reduced dry-matter by [1/2(41+68-36-46)] 14% (LSD005=
9.8%) in the first crop and increased it by [1/2(77+58-28-73)] 17%
(LSD0.05= 14.2%) in the second crop. Therefore, immediate (first
crop) and residual (second crop) effects of both TSP and DAP with
no lime and liming 4 wk before applying P were similar. However,
liming after P application reduced the yield of the first crop with
both TSP and DAP, but the residual response to DAP was positive and
to TSP not significant compared to no lime.
Phosphorus uptake, expressed as per cent of maximum, was 23%
greater in the first crop and 14% greater in the second crop with
TSP than with DAP, averaged over P rates and lime treatments (Table
S 3). The P rate of 80 mg kg-1 increased P uptake of watermelon by
28% in the first crop and 32% in the second crop, averaged over
P sources and lime treatments, in comparison to the 40 mg kg'' rate.
The interaction between P rates and sources in the first crop was
shown by an increase in P uptake of 36% with TSP and only 19% with
DAP between 40 and 80 mg of P kg'. The lack of interaction between
P rates and sources in the second crop was demonstrated by similar
increases in P uptake between 40 and 80 mg P kg'1 with each P
source, i.e. 34% with TSP and 31% with DAP. Therefore, we
conclude that DAP with lime was considerably more
[1/4(63+88+47+72-47-70-22-30)=25%] available to watermelon in the
second crop (residual effect) than in the first crop. Also, there
was a small overall increase (10%) between crops in availability of
Unlike dry-matter response, there was no interaction between
lime treatments and P sources on P uptake of watermelon. Liming 4
wk before P application increased P uptake by 17% (LSD0g05= 8.3%) in
the first crop and by 22% (LSD0.05= 11.3%) in the second compared to
no lime. Response to liming 4 wk after P application was different
between the first and second crop. Liming after P application in
the first crop reduced P uptake 12% (LSD.05= 8.3%), however,
residual (second crop) lime after P increased P uptake 11% (LSDo.05=
11.3%) compared to no lime, although the increase was not
Availability of TSP was 23% greater than DAP in the first crop
and 14% greater (significant at P 0.01, Table 1.) in the second
crop, averaged over P rates and lime treatments. The data indicate
that liming 4 wk before P application increased the availability
of P from both P sources. Liming after P application decreased P
availability in the first crop and increased it in the second crop.
Furthermore, liming after P application in the first crop decreased
P availability more with DAP than TSP. The difference between TSP
and DAP in P availability did not change between the first and
second crop with no lime. However, liming either before or after
P application had a tendency to reduce the difference in
availability of residual P between TSP and DAP.
If cost per unit of P is the same, TSP appears to be superior
Sto DAP on acid soils. However, if DAP has cost advantage and soil
pH is below 6.0, lime application several weeks before P
application will make P availability from DAP about equal to that
of TSP. Cost of DAP, TSP, and lime should be considered when
deciding which P source to use.
Freie, R. L., and H. V. Young, Jr. 1993. Vegetable summary 1991-
1992, Florida Agricultural Statistics. Florida Department of
Agriculture and Consumer Services, Orlando.
Hanlon, E. A., G. Kidder, and B. L. McNeal. 1990. Soil, container
media, and water testing; interpretations and standardized
fertilization recommendations. Florida. Coop. Extension
Serv., Univ. of Florida, Gainesville. Circular 817. 49 pp.
Locascio, S. J., P. H. Everett, and J. G. A. Fiskell. 1968. Effects
of phosphorus sources and copper rates on watermelons. Proc.
Am.Soc. Hort. Sci. 92:583-589.
Rhoads, F. M. 1991. Phosphorus source, soil-test P, snap bean
growth, and P uptake. Soil Crop Sci. Soc. Florida Proc.
Rhoads, F. M., E. A. Hanlon, and S. M. Olson. 1993. Phosphorus
availability to snap beans as affected by lime sequence and
phosphorus source. Soil Crop Sci. Soc. Florida Proc. 52:90-
Singh, R. N. and L. F. Seatz. 1961. Alfalfa yield and composition
after different times and rates of lime and phosphorus
application Soil Sci. Soc. Am. Proc. 25:307-309.
Soltanpour, P. E., F. Adams, and A. C. Bennett. 1974. Soil
phosphorus availability as measured by displaced soil
solution, calcium-chloride extracts, dilute-acid extracts, and
labile phosphorus. Soil Sci. Soc. Am. Proc. 38:225-228.
Steel, R. G. D., and J. H. Torrie. 1960. Principles and Procedures
of Statistics. McGraw-Hill, New York. 481 pp.
Wang, H. D. and T. L. Yuan. 1989. Lime effects on phosphorus
retention and release in four Ultisols. Soil Crop Sci. Soc.
Florida Proc. 48:131-137.
* Table 1. Calculated F values from analysis of variance procedures
for shoot dry-weight and phosphorus uptake of watermelon plants.
Shoot dry-weight P uptake
Source df First crop Second crop First crop Second crop
Replication 3 0.57 3.52* 2.49 0.49
P source (A) 1 56.95** 10.58** 47.53** 10.44**
P rate (B) 1 84.94** 42.05** 69.76** 51.84**
AB 1 8.76** 0.09 6.33* 0.12
Lime (C) 2 24.69** 12.60** 25.10** 7.78**
AC 2 5.73** 4.28* 0.47 1.33
BC 2 1.42 1.67 1.68 .0.15
ABC 2 0.73 0.86 0.74 0.88
* *, ** = Significance at 0.05 and 0.01, respectively.
Table 2. Dry-matter yield of watermelon shoots for two crops grown
in the same soil with application of phosphorus and lime to the
first crop only. Sources of phosphorus were triple superphosphate
(TSP) and diammonium phosphate (DAP)
P rate Rate Timet
mg kg'' g kg'1
0 0 -
40 1 L-P
80 1 L-P
40 1 P-L
80 1 P-L
First crop Second crop
TSP DAP TSP DAP
% of maximum
11 11 6 6
62 41 63 28
100 68 91 73
69 67 75 81
98 84 100 99
52 36 72 58
87 46 94 77
13.9 13.9 20.1 20.1
t L-P = Lime incubated in soil 4 wk before P application and P-L =
P incubated in soil 4 wk before lime application.
$ The treatment in each crop that produced the highest yield was
given the value of 100%.
S Table 3. Total phosphorus uptake of watermelon plants for two
crops grown in the same soil with application of P and lime to the
first crop only. Sources of P were triple superphosphate (TSP) and
diammonium phosphate (DAP).
First crop Second crop
TSP DAP TSP DAP
% of maximum
3 3 6 6
47 26 53 23
80 53 82 66
54 47 62 63
100 70 100 88
37 22 57 47
67 30 92 72
16.7 16.7 22.5 22.5
t L-P = Lime incubated in soil 4 wk before P application and P-L
= P incubated in soil 4 wk before lime application.
$ The treatment in each crop that had the highest P uptake was
given the value of 100%.
TSP DAP TSP DA