IC0
NFREC, QUINCY RESEARCH REPORT 90-14
SOYBEAN IffD AND OIL-TEST
PHOSPHORUS, QTASSt AND MAGNESIUM
F. M. RHOADS AND R.
D. BARNET
Central Science
Library
r APR 16 199 1
University of Florida
Florida Agricultural Experiment Stations
Institute of Food and Agricultural Sciences
University of Florida, Gainesville
Ce.WACHcIW AMUCH EMPAiICm
The change to Mehlich-I extractant by the University of Florida extension soil testing
laboratory in 1977 created an immediate need for soil-test calibration data for crops grown on soils
of North Florida (Sartain, 1978). Soybeans are grown throughout the panhandle section of Florida,
however, much of the fertility research on soybeans was done before 1977 (Robertson et al., 1967).
Therefore, the need for yield versus soil-test phosphorus (P), potassium (K), and magnesium (Mg)
data for soybeans will continue into the future.
A cropping system with the possibility of increased income per acre is wheat in winter
followed by soybeans in summer. Since soybeans do not require fertilizer nitrogen, it appears
feasible to apply all of the fertilizer to the wheat crop and let the soybean crop recover the residual
P, K, and Mg not used by wheat. However, it has been shown that wheat does not respond to K
as much as to P and fertilizer efficie t be improved if K is applied to soybean instead of
wheat (Rhoads and Barnett, 1986).
The objective of this research is ne yield response of soybean following wheat to
soil-test levels of P, K, and Mg onlorfo amy fine sand (fine-loamy, siliceous, thermic, Typic
Kandiudult).
MATERIALS AND METHODS
Two soybean cultivars (Cobb and Davis) were planted with minimum-tillage 15 June 1984
following wheat harvested the first week of June. The second year replanting was required 1 Aug
1985 because of post-emergence herbicide damage, therefore, only the Cobb cultivar (maturity
group VIII) was replanted as it was too late for 'Davis' (maturity group VI).
Fertilizer P, K, and Mg were applied to the wheat each year and soybeans were produced
from residual P, K, and Mg with the exception of sidedressed application of K in 1985. Nutrient
levels applied to wheat each year were as follows: P1 = 0, P2 = 25 lbs P/acre, P3 = 50 lbs P/acre,
P4 = 100 lbs P/acre; K1 = 0, K2 = 190 lbs K/acre, K3 = 380 lbs K/acre; Mg1 = 0, Mg2 = 60 lbs
2
Mg/acre, Mg3 = 120 lbs Mg/acre. Sidedress applications of K in 1985 were K1 = 0, K2 = 95 lbs
K/acre, and K3 = 190 lbs K/acre. A standard level of all fertilizer elements except the element
being varied was applied to each treatment. The standard levels were P = 50 lbs/acre, K = 380
lbs/acre, and Mg = 60 lbs/acre. Sources of fertilizer elements were 46% triple superphosphate,
muriate of potash, and anhydrous magnesium sulfate. Calcitic lime was applied as necessary to
maintain soil pH above 5.5.
Row spacing was 30 inches with a plant population of approximately 80,000 plants per acre.
Plot size for each fertilizer treatment was 20 ft by 24 ft. Harvested area for each treatment was
approximately 5 ft by 17.4 ft. Yield was expressed in bu/acre at 12% moisture and 60 lbs/bu.
Leaf samples (most recent mature trifoliates) were collected at first bloom in 1984. Samples
were dried at 600C, ground to pass a 20-mesh screen, dry-ashed at 5000C, taken up in dilute HCI
and analyzed for P, K, and Mg. Phosphqis was determined by the molybdenum blue method and
K and Mg by flame emission spectrophoto Ijry.
Ten 1 in by 6 in soil cores were composite from each plot in July of each year. Soil
samples were air-dried and ground before extracting with Mehlich-I (double acid) extractant. Soil
analysis were carried out according to University of Florida Extension Soil-Testing Laboratory
procedures (Johnson et al., 1984).
Experimental design was a randomized complete block with four replications for each
cultivar in 1984 and eight replications in 1985. Analysis of variance procedures were used to
calculate error variance and single degree-of-freedom F-tests were used to compare response of
soybeans to different levels of soil-test fertilizer elements (Steel and Torrie, 1960).
RESULTS AND DISCUSSION
The soil pH of samples from individual plots ranged from 5.5 to 6.5 during each year of the
tests.
Yield response of soybean to soil-test P was similar between cultivars (Table-1.). Average
soil-test P for the lowest P treatment was 5 ppm for Cobb and 7 ppm for Davis, however, yields
were 27 and 25 bu/acre, respectively, for Cobb and Davis. The only significant yield increase
occurred between the lowest and next higher P levels for each cultivar. The increase amounted
to about 52% for Cobb and about 64% for Davis. Maximum yield for each cultivar occurred at
a soil-test P level near 30 ppm. Highest soil-test P levels were near 50 ppm, however, yields were
the same or less than for those near 30 ppm.
Table 1. Yield and phosphorus (P) concentration of two soybean
cultivars grown with different levels of soil-test P.
(1984).
Cobb Cultivar Davis Cultivar
Soil-test Pt P-Conc.it Yield Soil-test Pt P-Conc.it Yield
ppm % bu/acre ppm % bu/acre
5 0.26 27 a** 7 0.25 25 a**
18 0.35 41 b 13 0.31 41 b
28 0.36 46 b 27 0.32 43 b
45 0.37 43 b 48 0.34 43 b
tSoil samples taken in July. tPlant samples taken in September.
**Means within columns followed by different letters are signifi-
cantly different (P < 0.01).
Soybean-tissue P concentration ranged from 0.25 to 037% for both cultivars combined
(Table-1). The critical level of tissue-P in recently mature trifoliates at first bloom appears to be
about 0.36% for Cobb and about 0.32% for Davis.
Cobb soybean did not respond with a yield increase to soil-test K in the first year of the tests
(Table-2.), however, Davis responded with a 25% yield increase between soil-test K levels of 40
and 49 ppm. Both cultivars responded to increased soil-test K with an increase of tissue-K
concentration. The data suggest a tissue-K level in recently mature trifoliates at first bloom of
4
Table 2. Yield and potassium (K) concentration of two soybean
cultivars grown with different levels of soil-test K.
(1984).
Cobb Cultivar
Soil-test Kt K-Conc.t Yield
ppm % bu/acre
40 0.76 43 a**
62 1.26 45 a
90 1.42 46 a
Davis Cultivar
Soil-test Kt K-Conc.4 Yield
ppm % bu/acre
40 0.72 36 a**
49 1.21 45 b
70 1.33 43 b
tSoil samples taken in July. 4Plant samples taken in September.
**Means within columns followed by different letters are signifi-
cantly different (P < 0.01).
Yield response to soil-test Mg ranged from 16 to 24% for both cultivars combined (Table-
3.). Since the highest rate of Mg fertilization did not increase the soil-test Mg above that of the
middle rate, a critical level of Mg in the soil or tissue cannot be determined from these tests.
Table 3. Yield and magnesium (Mg) concentration of two soybean
cultivars grown with different levels of soil-test Mg.
(1984).
Cobb Cultivar
Soil-test Mgt Mg-Conc.t Yield
ppm % bu/acre
31 0.25 37 a*
50 0.32 46 b
51 0.34 43 b
Davis Cultivar
Soil-test Mgt Mg-Conc.t Yield
ppm % bu/acre
25 0.24 37 a*
48 0.28 43 b
37 0.35 42 b
tSoil samples taken in July. tPlant samples taken in September.
*Means within columns followed by the same letter are not signifi-
cantly different (P > 0.05).
The late planting reduced the yield potential of soybeans in the second year of the tests by
more than 50% (Table-4.). However, the increased number of replications for the single cultivar
increased the sensitivity of the tests by a considerable amount. For example, a significant yield
increase in response to soil-test P occurred between 14 and 30 ppm P. Therefore, maximum yield
occurred each year at a soil-test P level of near 30 ppm. Yield response to K was significant
between soil-test K levels of 43 and 69 ppm. Lowest soil-test K was 40 ppm in the first year of the
tests but Cobb did not respond. Perhaps the second year response of Cobb was due to sidedressed
K. The second year, Cobb responded with a 68% yield increase between soil-test Mg levels of 18
and 43 ppm. Mean soil-test Mg ranged from 31 to 51 ppm in Cobb soybean plots in the first year
of the tests.
Table 4. Yield of Cobb soybean planted in August with different
levels of soil-test P, K, and Mg. (1985).
Soil-test Soil-test Soil-test
P Yield K Yield Mg Yield
ppm t bu/acre ppm t bu/acre ppm t bu/acre
6 8.2 a** 43 12.5 a* 18 11.1 a**
14 11.8 b 69 15.3 b 43 18.6 b
30 18.6 c 91 18.6 b 45 15.8 b
70 14.9 c
tSoil samples taken in July. **,* Means within columns followed
by different letters are significantly different (P < 0.01 and
0.05, respectively).
CONCLUSIONS
Since consistent yield response to soil-test P occurred at levels below 16 ppm and maximum
yield occurred near 30 ppm P, the data agree with the medium range of 16 to 30 ppm P found in
the University of Florida extension soil-test recommendations (ESTR) (Johnson et al., 1984). Yield
response to soil-test K occurred on two out of three occasions in the range defined as medium in
the ESTR. Therefore, data from these tests do not disagree with ESTR for soil-test K on soybean.
However, a yield response in these tests to Mg in the soil-test range of 31 to 50 ppm suggests that
the upper boundary of the medium range (15-30 ppm) for soil-test Mg in the ESTR may be low.
Furthermore, the reduced yield due to late planting did not change the relative response of Cobb
soybean to soil-test P and Mg.
ACKNOWLEDGEMENT
This research was supported by funds from the Florida Agricultural Experiment Stations and
the Potash and Phosphate Institute.
REFERENCES
1. Johnson, G. V., R. A. Isaac, S. J. Donohue, M. R. Tucker, and J. R. Woodruff. 1984.
Procedures used by state soil-testing laboratories in the southern region of the United States.
Southern Cooperative Series Bulletin No. 190.
2. Rhoads, F. M., and R. D. Barnett. 1986. Response of 'Florida 301' and 'Florida 302' wheat
cultivars to N, P, K, and Mg fertilization. Southern Branch Amer. Soc. Agron. Abstracts p.
2.
3. Robertson, W. K., C. E. Hutton, L. G. Thompson, R. W. Lipscomb, and H. W. Lundy. 1967.
Soybeans in Florida, soils and soil management. Univ. of Fla. (IFAS) Agric. Exp. Stn.
Bulletin 716 p. 34-46.
4. Sartain, J. B. 1978. Adaptability of double acid extractant to Florida soils. Soil Crop Sci.
Soc. Fla. Proc. 37:204-208.
5. Steel, R. G. D., and J. H. Torrie. 1960. Principles and procedures of statistics. McGraw-
Hill, New York.
|