| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a USDA-ARS, 810 Highway 12 East, Mississippi State, MS 39762
b USDA-ARS, 230 Bennett Lane, Bowling Green, KY 42104
* Corresponding author (htewolde{at}ars.usda.gov)
Received for publication January 12, 2005.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Abbreviations: NS, supplemental nutrient solution • NS-full, full Hoagland's nutrient solution • NS-macro, a solution of the macronutrients N, P, K, Ca, and Mg • NS-micro, a solution of the micronutrients B, Fe, Zn, Mn, Cu, and Mo • NS-none, water, no additional nutrients
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
The magnitude of accumulation of litter-supplied metal nutrients in the soil planted to row crops is also poorly understood when poultry litter is used repeatedly as the primary fertilizer. Application of fresh broiler litter of as much as 9 Mg ha–1 is needed to meet the N requirement of row crops such as cotton in the southeastern cotton production region of the United States. Based on average litter metal nutrient concentration reported by Jackson et al. (2003), the total metal nutrient input to the soil from 9 Mg ha–1 fresh broiler litter is 15.7 kg Fe ha–1, 3.4 kg Cu ha–1, 3.2 kg Mn ha–1, and 2.7 kg Zn ha–1. Grass and legume forage crops remove a small fraction of these amounts from soil. Annual ryegrass (Lolium multiflorum Lam.), for example, removed only about 200 g Zn ha–1 and 25 g Cu ha–1 (Brink et al., 2001). Certain legume forage species seem to be more efficient in extracting both Zn and Cu but the total extraction by these forages, relative to that applied, does not exceed 10 to 15% of applied (Brink et al., 2001). The efficiency of cotton and other row crops in extracting litter-derived metal nutrients, with benefits of reducing their buildup in the soil, is not quantified and therefore not well understood. The second objective of this research therefore was to determine the efficiency of cotton in extracting litter-supplied Fe, Cu, Mn, and Zn in the absence of other nutrient sources.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
|
Five cotton seeds were planted in each pot 21 d after litter application and thinned to two plants per pot after seedling establishment. Adequate tap water was applied to each pot to meet water needs of plants throughout the growing period. Drainage was prevented by applying just enough water to wet the soil volume. As a precaution, a clear plastic container was placed under each pot to collect drainage in case water applied exceeded the holding capacity of the growing mix.
Measurements
Plants were harvested 92 d after planting when they had produced squares, flowers, or small bolls. All plants were cut at soil level, lightly rinsed with a fine mist of tap water, and partitioned into leaves (blades and petioles), stems (branch and mainstem), and reproductive parts (squares, flowers, and bolls). Leaves were further divided into upper, middle, and lower mainstem leaves and branch leaves. After taking mainstem leaves from the upper three nodes, the remaining mainstem leaves were equally divided into lower and middle nodes. All leaves from branches were placed in a separate group. Roots of all plants were gently separated from the growing mix and thoroughly washed with tap water to remove adhering sand, vermiculite, or litter. Plant parts were dried in a forced-air oven at 80°C to constant weight, weighed, ground to pass 1-mm sieve, and analyzed for nutrient concentration.
Nutrient Analysis
Concentrations of Fe, Cu, Mn, and Zn in plant parts were determined by inductively coupled dual axial argon plasma spectrophotometer (ICP) (Model 1000; Thermo Jarrell-Ash, Franklin, MA) (Donohue and Aho, 1992). Approximately 0.2 g dried and ground samples were ashed in a muffle furnace at 500°C for 4 h. The ash was digested by adding 1.0 mL 6 M HCl for 1 h and 40 mL of a double-acid solution of 0.0125 M H2SO4 and 0.05 M HCl for an additional 1 h. The digested solution was then filtered using a 2V filter paper (Whatman, Maidstone, UK) and analyzed for concentration of the metal elements by the ICP. Concentration of metal nutrients in the litter was determined by the same method used for the plant tissues. The litter was oven-dried at 65°C, ground to pass 1-mm sieve, and analyzed for nutrient concentration.
Accumulations of the micronutrients Fe, Cu, Mn, and Zn in each plant part were calculated as the product of concentration and dry weight of each plant part. Total micronutrient extraction by plants in each pot was determined as the sum of micronutrient accumulation in leaves, stems, roots, and reproductive parts. Total micronutrient extraction as percent of total applied was considered as the efficiency by which cotton extracted these nutrients from the growing mix.
The fraction of a micronutrient partitioned to a plant part as a percent of total accumulated in all plant parts was calculated as follows: % partitioning = (100 x NWtissue)/NWtotal, where NWtissue = nutrient weight partitioned to plant part and NWtotal = sum of nutrient weights in all plant parts. Partitioning to leaves was calculated as a sum of the nutrient weights partitioned to the different leaf positions.
Statistical Analysis
Evaluation of nutrient concentration data for variance homogeneity revealed that treatment variances were not independent from treatment means. The data were therefore log10–transformed before statistical analysis; other data required no transformation. All data were analyzed using the MIXED model analysis on SAS (Littell et al., 2002). The analysis was performed as a simple RCB design with a 4 x 4 factorial treatment structure (four NS and four litter rate treatments) when plant parts was not included in the model as in the analysis of nutrient concentration in upper leaves. When plant parts was included in the model, the analysis was based on a split plot treatment design where the main plot, with a 4 x 4 factorial treatment structure, was a RCB design and the seven plant parts were included as the sub plot. When included in the model, the sub plot plant parts was treated as a repeated measures factor because plant parts cannot be randomized as the sub plot and the data for the seven plant parts were derived from the same plant.
Treatment means were compared using LSD when the analysis of variance and mean comparisons were made on data with the same scale. For example, if the analysis was performed with data that needed to be transformed to a log-scale, LSD was used to compare means on a log scale before transforming back to normal scale. The least significant ratio (LSR) was used to compare means when it was necessary to analyze the data on a log scale but the means are presented on a normal scale. All means are least significant means generated by the LSMEANS statement of SAS. Unless specified otherwise, all declared significant differences are at the P 
0.05 level.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
|
Supplementing litter with NS-full or the NS-micro treatments, both of which contained Zn, increased Zn concentration in the upper leaves (Table 2). Supplementing with the NS-full treatment, which increased Zn concentration by an average of 176%, brought Zn concentration to levels considered sufficient for cotton at early bloom stage (20 to 200 mg kg–1) but not at late bloom or maturity stage (50 to 300 mg kg–1) (Mitchell and Baker, 2000). Supplementing with the NS-micro treatment increased Zn concentration in the upper leaves when applied litter was 
90 g pot–1 but not when applied litter was 60 g pot–1. Unlike its increasing effect on Fe and Cu concentration, the NS-macro treatment did not significantly affect Zn concentration in upper leaves.
Litter alone, at each rate, resulted in Mn concentration in the upper leaves well within the published sufficiency range (Table 2). Supplementing litter with any of the NS treatments did not significantly affect upper leaf Mn concentration at any rate of applied litter. Despite an average water solubility of Mn of only 2% (Jackson et al., 2003), our results suggest that litter-derived Mn is readily available for uptake by cotton roots and that the Mn requirement of cotton may be satisfied by litter rates much less than the rates needed to satisfy the requirement of the other micronutrients.
Micronutrient Concentration in Plant Parts
Plant part main effect, plant part by NS interaction effect, and plant part by broiler litter interaction effect were statistically significant for all four micronutrient concentrations (Table 3). The NS by broiler litter interaction was also significant for Cu and Zn but not for Fe and Mn concentration. The three-way interaction effect (plant part by NS by broiler litter) was significant for Fe and Mn concentrations only.
|
|
|
|
Increasing rate of applied litter did not always significantly affect Fe concentration in the different plant parts. Despite a decreasing trend of Fe concentration in stems and reproductive parts and an increasing trend in branch and upper leaves with increasing litter rate under the NS-none treatment, litter did not consistently affect Fe concentration in most plant parts under the different NS treatments (Fig. 1, Table 4).
Copper
Similar to that of Fe, the concentration of Cu was several orders of magnitude greater in roots than in any other plant part (Fig. 1). When averaged across the nutrient solutions, Cu concentration was 6 to 10 times (depending on litter rate) greater in roots than in upper mainstem or branch leaves. Differences among the other plant parts in Cu concentration were smaller but statistically significant. In contrast to Fe, Cu concentration within the leaf parts was greatest in upper mainstem leaves and branch leaves with the following order: upper leaves = branch leaves > middle leaves > lower leaves. Copper concentration in reproductive parts exceeded Cu concentration in any leaf parts. The greater concentration of Cu in younger leaves probably is an indication that Cu, unlike Fe, is moving from older leaves to reproductive parts and younger leaves through the phloem. Copper has been reported to remobilize through the phloem as a complex of N compounds following the hydrolysis of proteins and senescence of older leaves (Kochian, 1991).
All NS treatments that contained micro- or macronutrients increased Cu in leaves and stems relative to the NS-none treatment. The largest increase of 126% was in branch and upper mainstem leaves due to the NS-full treatment. Only the NS-full increased root Cu concentration above the NS-none treatment. The supply of Cu as the NS-micro treatment did not increase root Cu concentration when averaged across litter rates.
Root Cu concentration, unlike Fe concentration, increased with increasing litter rate for every NS treatment (Fig. 1, Table 4). Averaged across the NS treatments, root Cu concentration increased by 0.55 mg Cu kg–1 dry wt. g–1 applied litter. Increasing applied litter rate did not affect Cu concentration in reproductive parts but increased Cu concentration in stems and all leaf positions when the NS treatment did not contain the macronutrients N, P, K, Ca, and Mg. In the presence of these macronutrients in the NS treatments (NS-macro and NS-full), Cu concentration in all leaf positions and stems remained the same across all litter rates. Because root Cu concentration increased with increasing litter rate at each NS treatment, the lack of increase of Cu concentration in leaves and stems with increasing litter rate when the NS treatment contained macronutrients may be an indication that one or any combination of the macronutrients was antagonistic to the translocation of absorbed Cu from roots to leaves and stems.
Manganese
Unlike Fe and Cu, the concentration of Mn, independent of the amount of applied litter, was greatest in the oldest and middle leaves and lowest in stems (Fig. 2)
. When averaged across litter rates and NS treatments, the relative order of Mn concentration in leaves was similar to that of Fe: lower leaves > middle leaves > branch leaves = upper leaves (84, 75, 59, 56 mg Mn kg–1 dry wt., respectively). This order suggests that the redistribution of Mn from older plant parts to new growth is similar to that of Fe, which is limited. These results support the observation of Kochian (1991) who believes that Mn is mobilized easily from roots to shoots through the xylem but not so easily remobilized from older leaves to other plant parts through the phloem. The concentration of Mn in reproductive parts was similar to that in branch and upper leaves under the NS-none and NS-micro treatments but not under the other NS treatments.
Relative to the NS-none treatment, supplementing litter with NS treatments that contained Mn had an overall decreasing effect on Mn concentration of most plant parts (Fig. 2). This effect was largest on reproductive parts, stems, and the oldest leaves. When averaged across litter rates, Mn concentration in reproductive parts was reduced from 60 mg Mn kg–1 under the NS-none treatment to 48 and 37 mg Mn kg–1 dry wt. under the NS-micro and NS-full treatments, respectively. It appears Mn uptake or translocation was competitively suppressed by one or more of the other micronutrients, most likely by Fe (Reisenhauer, 1994). The NS-macro treatment also reduced Mn concentration in reproductive parts and stems but increased it in the older lower and middle leaves with little or no effect on upper and branch leaves. This may be due to the divalent cations Ca and Mg competitively suppressing tissue Mn concentration (Reisenhauer, 1994).
Increasing amount of applied litter resulted in a gradual increase of Mn concentration in roots but the change in other plant parts was not consistent (Fig. 2). When averaged across the NS treatments, root Mn concentration increased from a low of 26 mg kg–1 when 30 g litter pot–1 was applied to 61 mg kg–1 when 120 g litter pot–1 was applied. Manganese concentrations, in general, are higher in roots than in leaves when plants are supplied with adequate Mn (Kochian, 1991). When Mn is not supplied in adequate amounts, however, Mn concentration declines in roots and stems and remains high in older leaves. The gradual increase in root Mn concentration in our study agrees well with the description in Kochian's (1991) review. It appears that the translocation of absorbed Mn out of roots to other plant parts was curtailed by some mechanism as the amount of applied litter and therefore Mn increased. It is possible that cotton plants have a mechanism that regulates the translocation of excess Mn out of roots and prevents its accumulation to toxic levels in younger aboveground tissues such as upper leaves and developing reproductive parts. Roots and older leaves may have greater tolerance for higher Mn concentration than younger tissues. Unlike Sudax grass [Sorghum bicolor (L.) Moench] which accumulated Mn to toxic concentration (>400 mg/kg) in a similar study by van der Watt et al. (1994), our results show that cotton may not absorb and accumulate Mn in leaves and reproductive parts in excess of its need regardless of its availability in the growing medium.
Zinc
Unlike the other three micronutrients, Zn concentration was greatest in reproductive parts at every applied litter rate when no additional Zn was applied in the nutrient solutions (Fig. 2). When litter was supplemented with the NS-full treatment, Zn concentration was greatest in upper (39 mg kg–1) and middle (41 mg kg–1) mainstem leaves. Zinc concentration in stems was distinctly below the concentration in all other plant parts when litter was supplemented with the NS-micro or the NS-full treatments, both of which contain Zn. The concentration of Zn in stems was similar to the concentration in branch and the older lower and middle mainstem leaves when litter was not supplemented with additional Zn. It appears cotton meets the Zn demand of its reproductive parts as a priority followed by the need of upper leaves and roots when faced with Zn deficiency as in the NS-none treatment. When adequate Zn is supplied, absorbed Zn in excess of that required for reproductive parts is allocated to the upper and middle mainstem leaves and roots. Interestingly, the Zn concentration of branch leaves was more similar to that of older leaves than to the younger upper leaves. In contrast to Zn, the concentration of Fe, Mn, and Cu in branch leaves was similar to the concentration of these same micronutrients in upper leaves.
Overall, Zn concentration in all plant parts, other than reproductive parts, increased with increasing rate of applied litter when the litter was supplemented with either of the two solutions that contained Zn (NS-micro and NS-full) (Fig. 2). Zinc concentration in reproductive parts remained virtually constant at approximately 25 mg kg–1 dry wt. when applied litter increased from 30 to 120 g pot–1 and the litter was supplemented with the NS-micro or NS-full treatments. The concentration of Zn in all plant parts changed little with increasing litter rate when litter was supplemented with the NS-macro treatment which did not contain Zn.
Relative to the NS-none treatment, NS treatments with Zn generally increased Zn concentration in all plant parts other than reproductive parts. The NS-full treatment increased Zn concentration in the older lower and middle mainstem leaves by four- to fivefold relative to the NS-none treatment. The Zn concentration in all plant parts that received the NS-macro treatment without Zn was virtually the same as the Zn concentration in the respective plant parts of the NS-none treatment. These results show that soluble Zn supplied in nutrient solutions is readily absorbed by cotton roots. The results also suggest that litter-supplied Zn was not readily plant available probably because of insolubility.
Partitioning and Accumulation of Micronutrients in Different Plant Parts
Expressing micronutrient accumulation in the different plant parts as a percent of the total absorbed showed distinct differences in micronutrient partitioning. Both Fe and Cu were partitioned to roots more than to any other plant part at all NS treatments (Table 5). Averaged across litter rates, 58% of total absorbed Fe and 64% of total absorbed Cu were partitioned to roots in the absence of supplemental nutrient solution, but the dry matter partitioned to roots did not exceed 19% (Tewolde et al., 2005). The corresponding partitioning to leaves was 32% for Fe and only 13% for Cu.
|
Increasing litter rate affected the percent partitioning of Cu to the different plant parts differently, but partitioning of Fe to any plant part was little affected. Percent Cu partitioned to roots significantly increased and that partitioned to leaves and stems significantly decreased with increasing litter rate between 30 and 120 g pot–1 when litter was supplemented with macronutrients (NS-macro) only (Table 5). The trends in Cu partitioning were similar when litter was supplemented with the other NS treatments but the decreases or increases were not significant (P 0.05). According to the review by Kochian (1991), Cu binds to and forms strong complexes with N compounds such as amino acids, amides, and ureides. It is possible litter supplies these N compounds to aid in the uptake of Cu to roots without contributing to the translocation of Cu out of roots to leaves and stems through the xylem. Percent Cu partitioned to reproductive parts seems to have peaked around 60 g litter pot–1 and decreased beyond this rate.
Unlike Fe and Cu, a greater fraction of absorbed Mn was partitioned to leaves than to any other plant part (Table 5). An average across litter rates and NS treatments of 65% of the total Mn uptake was partitioned to leaves compared with only 17% to roots. This partitioning pattern can largely be attributed to the greater fraction of total dry matter partitioning to leaves (approximately 42%) than to any other plant part (Tewolde et al., 2005). As in Cu, percent Mn partitioned to roots increased and that to leaves decreased with increasing rate of applied litter. There was no clear increasing or decreasing effect of litter rate on percent Mn partitioned to reproductive parts or stems. The percent partitioning of Mn to the different plant parts averaged across litter rates remained the same across all NS treatments although the Mn-containing NS treatments (NS-micro and NS-full) decreased Mn concentration in virtually all plant parts (Fig. 2). However, the partitioning of Mn to stems was slightly but significantly decreased by the supply of macronutrients (NS-macro or NS-full).
The partitioning pattern of Zn was similar to that of Mn—largely to leaves (Table 5); however, the percent Zn partitioned to leaves was substantially less than that of Mn—average across litter rates of 37% for Zn vs. 65% for Mn when the NS treatment contained no Zn and Mn. Increasing the rate of litter affected percent Zn partitioning in the same way as Mn partitioning but to a lesser extent. Percent Zn partitioned to leaves increased from an average across litter rates of 37% when no Zn was supplied to 55% when Zn was supplied in the NS treatments. The Zn-containing NS treatments decreased the percent Zn partitioned to reproductive parts and stems. The partitioning of Zn to reproductive parts was 22% when no additional Zn was supplied but decreased to 11% when Zn was supplied in the NS treatments. Zinc partitioning to roots, with an average across litter rates and NS treatments of 23%, was not affected by the NS treatments.
Nutrient Extraction Efficiency
Cotton extracted by far more Fe from the growing mix than any other micronutrient at any litter rate (Table 6). This was particularly true when litter was supplemented with the NS-full treatment which contained all micronutrients. In the absence of supplemental nutrients, extraction of Cu, Mn, and Zn from the growing mix increased linearly with increasing rates of applied litter. The extraction of these nutrients increased with increasing litter rate when supplemented with the other NS treatments but the increases were not always linear. The extraction of Fe was not significantly affected by litter rate at any of the NS treatments.
|
The litter used in this study contained 684 mg Fe kg–1, 424 mg Zn kg–1, 522 mg Mn kg–1, and 556 mg Cu kg–1. All micronutrient concentrations in this litter are within the range reported by Jackson et al. (2003) for 40 litter samples. Annual application of 9 Mg litter ha–1, which would meet the typical N need of cotton production in the southeastern United States, would add 6.2 kg Fe ha–1, 3.8 kg Zn ha–1, 4.7 kg Mn ha–1, and 5.0 kg Cu ha–1 to the soil. According to Jackson et al. (2003) the average water solubility of Fe, Zn, Mn, and Cu in poultry litter, respectively, is 5, 6, 2, and 41% suggesting that, other than Cu, the extraction of these nutrients by plant roots would be low. Our research suggests that up to 9% of Fe and 7% of Mn supplied by the litter can be extracted by cotton which suggests that the plant availability of these micronutrients was better than the one-time solubility reported by Jackson et al. (2003). Interestingly, despite the reported 41% water solubility of Cu, a maximum of only 1.7% of litter-supplied Cu was extracted by cotton in this study and thus the percent water solubility of these metals in litter is not a good indicator of phytoavailability.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Related articles in JEQ:
This article has been cited by other articles:
|
|
 |
|
  H. Tewolde, M. W. Shankle, K. R. Sistani, A. Adeli, and D. E. Rowe No-Till and Conventional-Till Cotton Response to Broiler Litter Fertilization in an Upland Soil: Lint Yield Agron. J., May 7, 2008; 100(3): 502 - 509. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Adeli, M. W. Shankle, H. Tewolde, K. R. Sistani, and D. E. Rowe Nutrient Dynamics from Broiler Litter Applied to No-Till Cotton in an Upland Soil Agron. J., May 7, 2008; 100(3): 564 - 570. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Adeli, K.R. Sistani, D.E. Rowe, and H. Tewolde Effects of Broiler Litter Applied to No-Till and Tillage Cotton on Selected Soil Properties Soil Sci. Soc. Am. J., May 16, 2007; 71(3): 974 - 983. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Tewolde, K. R. Sistani, D. E. Rowe, A. Adeli, and J. R. Johnson Lint Yield and Fiber Quality of Cotton Fertilized with Broiler Litter Agron. J., January 1, 2007; 99(1): 184 - 194. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Vadose Zone Journal | ||||
| Soil Science Society of America Journal | Journal of Plant Registrations | The Plant Genome | |||
) treatments as applied to the growing mix. The stock solutions of each nutrient source were prepared according to 
