Crop Science Grow Your Career with CSSA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (8)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Bell, P. F.
Right arrow Articles by McConnell, J. S.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Bell, P. F.
Right arrow Articles by McConnell, J. S.
Agricola
Right arrow Articles by Bell, P. F.
Right arrow Articles by McConnell, J. S.
Related Collections
Right arrow Cotton
Right arrow Plant Analysis
Right arrow Plant Nutrition
Crop Science 43:1367-1374 (2003)
© 2003 Crop Science Society of America

CROP ECOLOGY, MANAGEMENT & QUALITY

Relationships between Leaf-Blade Nitrogen and Relative Seedcotton Yields

Paul F. Bell*,a, D. J. Boquetb, E. Millhollonc, S. Moored, W. Ebelhare, C. C. Mitchellf, J. Varcog, E. R. Funderburgh, C. Kennedya, G. A. Breitenbecka, C. Craiga, M. Holmani, W. Bakerj and J. S. McConnellk

a Agronomy Dep., LSU AgCenter, Baton Rouge, LA 70803-2110
b Northeast Res. Stn. at St. Joseph and Winnsboro, LA, LSU AgCenter
c Red River Res. Stn., Bossier City, LA, LSU AgCenter
d Dean Lee Res. Stn., Alexandria, LA, LSU AgCenter
e MSU, Mississippi Agric. and Forestry Exp. Stn., Delta Research and Extension Center, Stoneville, MS
f Dep. Agronomy and Soils, Auburn Univ., Auburn University, AL
g Dep. Plant Soil Sciences, Mississippi State Univ., Mississippi State, MS
h Noble Foundation, Ardmore, OK
i Consultant and formerly of LSU AgCenter
j Agriculture Studies, Arkansas State Univ., State University, AR
k Univ. Arkansas–Southeast Res. Ext. Ctr, Monticello, AR

* Corresponding author (PBell{at}lsu.edu)


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
Nitrogen fertilization is a required production practice for cotton (Gossypium hirsutum L.) with risks arising from under- and over-fertilization. Tissue testing for diagnosing N deficiencies in crops can use leaf blades and the total N concentration, but this practice has not been rigorously examined in cotton. The primary objective of these experiments was to determine the leaf-N concentration of the uppermost, fully mature leaf blade below which yield loss could be expected. Nitrogen-rate field experiments were conducted at 12 research station and farm sites in the Midsouth USA in Louisiana, Arkansas, Mississippi, and Alabama in 1996 and 1997. Leaf-blade total N concentrations associated with yield loss were 4.3% N at early bloom (R2 = 0.50) and 4.1% N at mid-bloom (3 wk after early bloom, R2 = 0.32). The likelihood of applying N when not needed could be reduced by lowering the early bloom critical value to 3.9%. Only 4% of all samples sufficient in N would have been incorrectly diagnosed N deficient at that critical value, but 44% of all deficient samples would have been misidentified as N sufficient. Reduced yields due to over application of N were evident in some samples with leaf N between 4.6 and 4.8% at early bloom. These concentrations were also common for N-sufficient plants, making accurate diagnoses of the over application of N unlikely. Our leaf-N critical values probably differ from previously established values because earlier values were derived via survey techniques and because faster fruiting cultivars may require higher leaf N.


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
NITROGEN FERTILIZER is used on 90% of the U.S. cotton hectareage to optimize growth and profits (Fertilizer Inst., 1998). Fertilizer N is a small expense compared with land costs or insect control and, given the variation of N need across fields, some farmers prefer to over apply N to some areas of the field rather than risk under applying N. Nitrogen fertilization is an unregulated practice in almost all of the USA. The major contributor to in-stream N inputs to the Mississippi River and the most important cause of the seasonal hypoxia in the Gulf of Mexico (the Dead Zone) is thought to be N fertilizer (Goolsby et al., 1999). Nitrate-N may leach into ground water when fertilizer applications exceed plant needs or when fertilizer applications are not synchronized with plant uptake. Today, methods for predicting N status of cotton grown in the Midsouth are either inaccurate or require additional studies before the critical values used in tissue testing can be accepted and used by farmers and extension personnel (Bock and Adams, 1980; Sabbe and Zelinski, 1990). Critical values are used in tissue testing to separate N-deficient from N-sufficient plants. Plants with tissue samples that contain N levels above the critical value are considered sufficient in N and no N fertilizer would likely be needed while those plants with N concentrations below the critical value are considered deficient in N and N applications might be advisable. Inaccurate critical values can result in the over-application or under-application of N.

The principal method for assessing cotton-N status under irrigated conditions is the petiole nitrate test. The test is a snapshot of N movement to leaves because it is an analysis of a transportable form of N, and because petioles or leaf stems are the conduit for nitrate transport from roots. Petiole nitrate is a sensitive measure of N movement to leaves, but it is probably hypersensitive because of soil moisture effects on petiole nitrate. Researchers have found petiole nitrate testing a better measure of soil moisture status than cotton-N status (Bock and Adams, 1980; Touchton et al., 1981). Another study found "tremendous" variations in petiole nitrate concentrations across years at any week after first bloom (Phillips et al., 1987) and found little evidence for reliable critical values from this method. Thom and Spurgeon (1982) proposed that moisture effects on nitrate measurements can be minimized if P analyses are done concurrently with nitrate and their guidelines for interpreting the interplay between P and nitrate are used. A test less sensitive to moisture is the total concentration of nitrogen in leaf blades because the residence time of N in the leaf is longer than that of the continuously replenished petiole nitrate in stems.

Most tissue testing for diagnosing nutrient deficiencies use the leaf blade and not the petiole for analyses. Although cotton, sugar beet (Beta vulgaris L. subsp. vulgaris), and some other crops commonly use the petiole for N diagnoses, N diagnoses are made for most crops using the leaf blade (Mills and Jones, 1996). It would be simpler to have all nutrient analyses done on the same leaf-blade sample than to have stems for N analyses and leaf blades for other nutrients' diagnoses.

Total-N analyses of tissue samples used to be slow because it required the Kjeldahl method that took hours to conduct. Today, N analyses utilizing the Dumas combustion method and that is equivalent to Kjeldahl's total-N analyses, takes about 3 min per sample. Turnaround times for the analyses of total N of growers tissue samples at a large analytical laboratory in the USA is 1 d (J. Still, personal communication).

Numerous studies have established critical values for petiole nitrate testing (Maples et al., 1977; Touchton et al., 1981; Phillips et al., 1987). Less effort has been devoted to develop critical values for leaf-blade N tissue testing. The critical nitrogen value used today for cotton leaf-blade tissue testing was derived from surveys of cotton in Arkansas in the early 1970s (Sabbe et al., 1972). A coauthor of that study repeatedly called for experiments to develop new critical values (Sabbe and Mackenzie, 1973; Sabbe and Zelinski, 1990). Experimental verification of the survey determined critical values are needed.

The objectives of this study were to (i) determine leaf-blade N concentrations at four maturities below which yield losses would likely occur and (ii) assess the reliability of the critical values found by comparing with those found in the literature. Experiments were conducted across 2 yr at 12 sites on experiment stations and private farms in four states.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 REFERENCES
 
Sites and Plant Culture
Cotton was grown in 1996 and 1997 under cultural practices common to each experiment station or farm in Alabama (Brewton and Prattville), Arkansas (Marianna), Louisiana (Alexandria, Bossier City, Columbia, Newellton, St. Joseph, and Winnsboro), and Mississippi (Stoneville and two sites at Starkville). Two Louisiana sites were on private farms (Newellton and Columbia).

Nitrogen was applied near the time of planting at Marianna, AR, Alexandria LA, Bossier City LA, St. Joseph LA, Winnsboro LA, and Stoneville MS. Half of the N was applied near planting and half at about first pin-head square stage at the Starkville, MS, sites; and all the N was applied near first pin-head square at Brewton, AL, Prattville, AL, Newellton, LA, and Columbia, LA. All sites used a randomized complete block design with at least three replications. A description of N rates, soil types, and soil chemical properties are given in Tables 1 and 2. All sites were rain-fed except the Marianna, AR, site which was furrow irrigated. Cotton was picked and seedcotton yields determined. Yields were normalized as a percentage of maximum yields (relative yield) found at each site for each year.


View this table:
[in this window]
[in a new window]
  		Table 1. General description of sites used in nitrogen study including soil series and taxonomic classification.

 

View this table:
[in this window]
[in a new window]
  		Table 2. Some chemical properties of soils and the application rates of N used in the experiments.{dagger}

 
Sampling Method
Leaves were sampled by personnel at each experiment station following a standardized protocol. Recently matured leaves were sampled from the main stem at the third or fourth node from the uppermost node containing a leaf about the size of a 2.5-cm coin. The sampled-leaf node corresponded to that used in conventional tissue testing and petiole-nitrate monitoring (Maples et al., 1977). Samples were oven-dried at 80°C overnight, and ground to pass a 40-mesh screen. Samples were taken at first or second pin-head square stage (about 500 degree day units after planting, degree day units were calculated for each day by averaging the daily high and low temperatures in degrees Fahrenheit and subtracting 60), early bloom (about 820 degree-days after planting or about one bloom for every 60 cm of row from 96-cm-wide rows), mid-bloom (3 wk after early bloom), and cut-out (blooms progressed up the plant and were at or above the fifth node from the top). Fertilizer application rates varied with sites, but most included low, optimum, and excessive levels (Table 2).

Chemical Analyses
Total-N concentration was determined with Leco FP-428 Dumas combustion-method total-N analyzer (Leco Corp., St. Joseph, MI; total Kjeldahl nitrogen equivalent, TKN). National Institutes of Standards and Technology peach or apple leaf standards were used to standardize the analyzer across the 2-yr study.

Statistical Analyses
Critical values for assessing cotton N status were determined by linear regression, a statistical procedure as described by Cate and Nelson (1971), and a method of Beverly (1993). The statistical procedure used SAS General Linear Models, where two treatments were composed of data below, and above or equal to the hypothesized critical value. Data included relative yields and leaf-N concentrations that corresponded to relative yields. One hundred percent relative yields were calculated from every site and year for that treatment that had the highest yield. The relative yields of other treatment means were calculated as the quotient of treatment average and treatment average from the highest yielding treatment and converting to percent. An iterative process was used to determine the critical value with the best fit (the highest R2) for the overall model. Beverly's (1993) method used the percentages of all sufficient samples correctly diagnosed (T-) and the percentages of all deficient samples correctly diagnosed (T+) to calculate an efficiency rating. The critical value selected had the highest efficiency rating. The efficiency rating was calculated as [(T+)/(T+ + T-)] x T+. Data were not included for any critical-value determinations from treatments where lower yields were caused by the over-application of N, i.e., treatments with <90% relative yields at N rates greater than that required for 100% relative yield. Means were used in calculating critical values.


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
REFERENCES
 
Relative yields and leaf-N concentrations from all field experiments are presented in Fig. 1. Each observation represents the average of leaf N concentrations at each N rate, growth stage, site, and year while the relative yields represent averages of each N rate, site, and year. The critical leaf-blade N concentration used to separate N deficient from N sufficient cotton for each growth stage was estimated using three techniques (Table 3). Critical values from these three methods are given in Table 3. One of the methods we use is equivalent to a graphical technique that utilizes figures like those of Fig. 1 (Cate and Nelson, 1965, 1971). Figure 1 shows a horizontal dashed line at 90% relative yield and is used to separate N-deficient (yields <90% relative yield) from N-sufficient plants (yields ≥90% relative yield). The 90% relative yield was selected because significant differences among relative yields in this study required about a 10% difference in yields using the LSD procedure. The leaf N% is given in the x axis and hypothetical critical values can be placed anywhere on that axis and a vertical line can be placed to intersect at the critical value that would separate diagnoses of N deficiency (those on the left of the line) from N sufficiency (those to the right of the vertical line). Correct diagnoses, therefore, of N–deficient and N–sufficient cotton are found in Fig. 1 in the bottom left and top right quadrants, respectively.



View larger version (32K):
[in this window]
[in a new window]
  		Fig. 1. Leaf N concentrations versus relative yields from N fertilization trials conducted in LA, MS, AR, and AL in 1996 and 1997. Results are presented for growth stages of first pin-head square (a), early bloom (b), mid-bloom (c), and cut-out (d). Our recommended critical values are presented as the vertical dotted line for early and mid-bloom growth stages. Correct diagnoses using these critical values are found in the bottom left and top right quadrants. At first pin-head square, some sites are represented as open squares (Brewton, AL), open triangles (Columbia, LA), and open diamonds (St. Joseph, LA).

 

View this table:
[in this window]
[in a new window]
  		Table 3. Leaf-N critical values used to diagnose N status of cotton at four growth stages using data from all sites and years. Critical values determined using three methods: statistical techniques (Cate and Nelson, 1971), linear/quadratic regression, and efficiency ratings (Beverly, 1993) that used the percentages of correctly diagnosed deficient and sufficient samples.

 
The first technique we used to estimate the critical values and listed in Table 3 was Beverly's (1993) efficiency rating method. It proved ambiguous because no single critical value was clearly the best because of the algorithm used in determining the efficiency rating. An efficiency rating of 53 at a critical value of 4.0% at early bloom was essentially the same as the 54 efficiency rating at the 4.6% critical value. This documents the tradeoffs made in selecting one critical value over another. On the basis of a critical value of 4.0%, 95% of all sufficient samples were diagnosed correctly at early bloom, but only 58% of all deficient samples were correctly diagnosed. This relationship was reversed at a critical value of 4.6 where 47% of all sufficient samples were diagnosed correctly and 87% of deficient samples were diagnosed correctly. Similar effects were seen at other growth stages and helps explain the unusual critical values listed in Table 3 for Beverly's method.

The other two techniques for calculating critical values used forms of regression analyses. The Cate-Nelson method used ANOVA and produced R2 less than that from simple linear/quadratic regression. We concluded that critical values selected with the latter method were most accurate and are used in Fig. 1.

First Pin-Head Square
Differences in N–application rates did not often result in differences in leaf N% at first pin-head square. This was partially understandable because fertilizer N was applied near the time of the first sampling at some sites (open squares in Fig. 1 for the Brewton, AL, site and open triangles for Columbia, LA). At other sites, similar effects were found even though fertilizer N was broadcast applied and incorporated near planting (St. Joseph, LA, site with open diamonds). The latter effect may have been caused by the high leaf N and a difficulty in increasing leaf-N beyond luxury levels. Plant uptake of N may also be less efficient at first pin-head square because of lower root density and the practice of applying N banded near the row (10 cm to the side and 12 cm deep) far from the smaller root systems.

The leaf-N critical value at pin-head stage of 5.4% determined by regression (Table 3) had a low R2 of 0.13, but the intercept and slope were statistically significant (P = 0.0001). Given the methods of applying N, even high applications of N near planting did not increase leaf N by first pin-head square in this study. Therefore, we consider this a tentative critical value and with limited utility. It may be more appropriate to use histogram data to determine another critical value at this growth state. For example, in our data set, only 25% of samples had leaf N <4.5%, and leaf N from 10% of the samples were <3.9%. Either leaf-N value, however, is much higher than the widely cited value of 3.5% from Mills and Jones (1996). Only treatments from one site in one year had leaf-N concentrations as low as 3.5% although cotton at several sites had large yield losses from N deficiencies. High yields were found at sites with low leaf N at first pin-head square and this indicated that adequate leaf N could be achieved by early bloom stage with N fertilization at the time of first pin-head square stage. Unresolved is whether the 100% relative yields calculated from these sites could have been increased had leaf N at first pin-head square been higher. However, cotton is not a crop that benefits from rapid growth or encouragement to grow rapidly during emergence to first bloom. If so encouraged, it will shed fruit in favor of vegetative growth that penalizes early yield production (Boquet et al., 1993).

Early Bloom
Critical values calculated for early bloom are listed in Table 3 with our recommendation (4.3% leaf N) for the value calculated with linear regression. Our recommended critical value can also be found in Fig. 1b and corresponds to the leaf N concentration along the x axis at the vertical dashed line. Incorrect diagnoses occur in the upper left and lower right quadrants. Figure 1b can be used, also, to determine the effect of using other critical values on the amount and type of error produced. Other critical values may be desired if growers or regulators are more interested in minimizing one type of error more than another (under or over application of N). For example, if one wanted to minimize the over application of fertilizer-N to plants already adequately supplied, perhaps fertilizer-N should be applied only if leaf N was less than 3.9% resulting in the over application of N to only 4% of all N-sufficient samples (i.e., 4% of all samples that had ≥90% relative yields were <3.9% leaf N). However, this would also result in no application of N to 44% of the samples that were N-deficient (relative yields <90% in Fig. 1b and leaf N ≥3.9%). If one wanted a more balanced approach, then our recommended critical value of 4.3% leaf N would be used. This would result in fewer errors of not applying N where it was needed (31% of samples that needed N would not have received N), but more errors where N would be applied where it was not needed (application of N to 26% of all N-sufficient samples).

Mid-Bloom
Tissue testing at mid-bloom stage or later may be too late for N to be applied to correct severe N deficiencies (Craig, 2002). Nonetheless, if accurate, mid-blossom testing could identify N-fertilization practices requiring modification for the next year's crop or could provide assurances to the grower or regulator that adequate N had been applied. Accuracy at this growth stage (R2 = 0.32) was poorer than at early bloom (R2 = 0.50) but better than at first pin-head square (R2 = 0.13). The decrease in accuracy from early bloom to mid-bloom may have been caused by artifacts of sampling. The sampling date was arbitrarily defined as 3 wk after early bloom at each site and, although this assisted in uniform sampling, mid-bloom may not have occurred in all treatments at that site. For example, at Winnsboro LA, in 1997 cotton from the zero and the lower N rates were already cutting out at 3 wk after early bloom while cotton at the highest N rate was not.

Cut-Out (Blooming Cessation)
A trend toward lower leaf N concentrations with advancing maturity was obvious from results of sampling because calculated critical values dropped from 5.4% at first pin-head square to 3.8% at cut-out. There was difficulty in determining cut-out since some cotton entered cut-out and then reentered the vegetative-growth phase after rainfall. Also, lower-N treated cotton usually entered cut-out earlier than higher-N treated cotton. The "critical value" is presented in Fig. 1d and Table 3 only to help document the decreasing trend in leaf N with increasing age.

Explanation of Differences in Critical Values across Sites and Years
At most sites each year, as more N was applied with the treatments, leaf N% and yields increased to a plateau. This relationship between yield and leaf N% could be used to select a critical value at each site corresponding to leaf N% at 90% of the maximum yield. Using this site and year specific method, we found critical values at early bloom to vary among sites between 3.9 and 4.6% in 1996 and from 3.4 to 4.7% in 1997. Others have also found variation in leaf-N critical values across years and sites (Cope, 1984). This range of values and the reasons some sites differed so much from earlier presented critical values calculated for all sites may be explained as follows.

  1. Explanation for unusually low critical values: Craig (2002) compared leaf N concentrations of cotton grown at various N rates at the St. Joseph, LA, site in 1999 and found that an extended period of drought could lower critical values. Soil-moisture effects also affect petiole nitrate measurements, but the moisture deficit required in the case of petiole N compared with leaf N is likely hours rather than days.
  2. Explanation for unusually low or high critical values: Although we describe N-deficient cotton as that with leaf N below a critical value, actual N deficiency may not have occurred at that level. It is more likely if leaf N drops below critical concentrations, soil-N reserves will soon be depleted, and then real N deficiency with a decrease in chlorophyll will occur and yields will decrease. Therefore, the rate of decrease of available soil-N interacts with the draw-down rate of soil N and, thus, may explain differences among critical values among sites. Sometimes our critical values are not diagnostic for N deficiencies but prognostic for future yield losses.
  3. Unusually high critical values: Application of N above what is normally required may cause rank growth and lower yields, but higher yields may result in some years. If so, critical leaf-N values would be greater in those years when plants require more N.
  4. Unusually high critical values and peculiarities associated with irrigated cotton: The one irrigated site in this study had the highest critical value at early bloom, 5.7% (treatments with relative yields of 63, 82, 93, and 100% had leaf-N values of 4.9, 5.6, 5.9 and 5.9%, respectively). Yields were not unusually high, however, with maximum lint yields of 1045 and 1280 kg/ha for 1996 and 1997, respectively. The approximate critical value for this site in 1996 was 4.2% leaf N.
  5. Other factors: As discussed below, cultivar differences, especially obsolete versus modern cultivars, could also affect critical values, but did not in this study because all cultivars were of the modern type.

Comparison of Critical Values from this Paper with Others
Critical values from this and other studies are listed in Table 4. Our critical values are greater than most of those listed in the literature and differ from some widely cited values. Our critical value at mid-bloom of 4.1% is much greater than a survey-derived value of 3.0% (Sabbe et al., 1972). The Sabbe et al. (1972) critical value was probably the source for the critical values cited by another widely cited reference material for critical values for all crops, Mills and Jones (1996), for their mid-bloom reference. Mills and Jones (1996) also listed a 3.5% critical value for vegetative-stem samples for first pin-head square and early bloom stages. The "vegetative stem" description is likely an error from this and an earlier edition (Jones et al., 1991) and should read instead "uppermost, fully developed leaf blade from the main stem." Our critical value at early bloom of 4.3% is also much greater than the earlier 3.5% of Mills and Jones (1996). Our critical values are high, too, compared with other agronomic or plantation crops with only three out of 22 crops listed in the Mills and Jones (1996) handbook having the same or higher leaf-N critical values than ours: sugar beet, 4.3% N (Beta vulgaris); faba bean, 4.8% N (Vicia faba L.); and cassava, 5.0% N (Manihot esculenta Crantz).


View this table:
[in this window]
[in a new window]
  		Table 4. Comparison of our recommended critical values with others. Leaf blades were sampled from the youngest, fully developed leaf on the main stem, unless otherwise noted.

 
Causes of Differences in Critical Values
The differences among investigators' critical values in Table 4 can be explained, first, by the location of the sampled tissue. Malavolta and Gomes (1961) used mature leaves from sympodia located mid-plant, rather than the latest mature leaf on the main stem as we used, and recommended a critical value 0.4% leaf-N less than we do. It is not known if the ontogeny of these leaves differs, although the data of Zhu and Oosterhuis (1992) would suggest that N concentration of sympodial and main stem leaves diverge with time. We also found lower leaf N% at lower positions of the plant with values of upper and lowest leaves differing by 1.9% leaf N in 1996 and 0.5% leaf N in 1997 at St. Joseph, LA (data not shown).

Second, a higher concentration of N in mainstem leaves may supply more N to a more rapidly developing fruit load in modern cultivars than leaves with lower N%. Wells and Meredith (1984b) indicated a significant trend toward greater partitioning of dry matter into reproductive structures with more recent cultivar releases. This partitioning occurred over the same or shorter time period found for obsolete cultivars. It is believed that the trend toward a shorter period has continued to present day cultivar releases. Increased partitioning of dry matter into reproductive growth could shift the temporal and, to some extent, the overall requirement for N by the cotton plant. Meredith et al. (1997) found modern cultivars responded with modest but significant yield increases to increased rate of N fertilizer while obsolete cultivars did not.

Leaf N concentration is related to increased leaf area and boll number (Gerik et al., 1994). Although boll number and/or weight generally increases with more recent cultivars (Wells and Meredith 1984b), leaf area does not consistently change compared with older cultivars (Wells and Meredith 1984a) even with increasing N rate (Heitholt et al., 1998). If increasing N does not differentially increase leaf area among different cultivars, then the benefit noted by Meredith et al. (1997) must come from some other mechanism, perhaps, from increased leaf N%.

Hearn (1969) indicated improved yield occurs through improved sink strength rather than source capacity. The greater partitioning of dry matter into reproductive structures earlier and faster suggests efficient assimilation of N would be necessary. The data from Zhu and Oosterhuis (1992) indicated the mainstem-leaf, subtending-node 10 exported 60% of assimilated N within 3 wk after reaching maximum dry weight. Corresponding sympodial leaves also exported some N, but not to the same extent. This redistribution of N from older mainstem leaves may be an important component in the N nutrition of modern cotton cultivars especially when root growth declines as a result of increased boll sink strength. This latter phenomenon was noted to occur for what would now be considered obsolete cultivars (Eaton, 1931) and would likely occur for modern cultivars resulting in a greater impact. A higher N concentration in uppermost mature mainstem leaves (our sample leaf) would be indicative of more N available for export to bolls later in the season. Certainly other factors could come into play to determine potential yield, but N availability in mainstem leaves (and to some extent sympodial leaves) for subsequent export to developing bolls would play a role in productivity.

Lastly, irrigation inputs, early-season insect control and modern cultivars that emphasize early boll set would certainly increase the yield potential of the crop, thereby resulting in increased demand for N and, perhaps, a greater critical value for leaf N. This increased N demand could increase the draw-down rate of soil N as discussed before and transfer some of the N requirement for seed from soil N to leaf N for subsequent reassimilation and transport.

Ability to Diagnose Over Application of N at Early Bloom
Some treatments produced low yields because of the over application of N (i.e., treatments with <90% relative yields at N rates greater than that required for 100% relative yield). We did not include these data in calculating critical values because their effect on linear regression analyses resulted in higher critical leaf-N values by about 0.1% leaf N when it was obvious it should have no effect on critical values used to differentiate between N-deficient and N-sufficient cotton.

In crop studies with other nutrients, data that included tissue samples with toxic levels of nutrients could be used to help identify the leaf nutrient concentrations associated with toxicity, but this was not the case here. Leaf-N concentrations from treatments where N was over-applied varied from 4.6 to 4.8% at early bloom. These concentrations, however, were not unusual and, in fact, have been considered critical values by others (Table 4). Therefore, in this study, leaf N was not useful in identifying cotton that had excessively high applications of N. This failure of the test may be related to cotton's ability to rapidly increase vegetative growth under high N conditions (Boquet et al., 1993). This may inhibit high concentrations of N in leaves in favor of a dilution of N among new growth.


   ACKNOWLEDGMENTS
 
We thank LWRRI (Louisiana Water Resources and Research Institute), USGS (US Geological Survey), and Cotton Incorporated for their support. The Arkansas portion of the study received some funding from the in-state fertilizer tonnage fee research program.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
Approved for publication by the director of the Louisiana Agric. Exp. Stn. as manuscript no. 02-09-0132.

Received for publication April 22, 2002.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 




This article has been cited by other articles:


Home page
 Crop Sci.Home page
H. Tewolde, K. R. Sistani, D. E. Rowe, A. Adeli, and D. L. Boykin
Nitrogen Extraction by Cotton Fertilized with Broiler Litter
Crop Sci., May 31, 2007; 47(3): 1131 - 1142.
[Abstract] [Full Text] [PDF]

Home page
 Agron. J.Home page
T. W. Katsvairo, D. L. Wright, J. J. Marois, D. L. Hartzog, J. R. Rich, and P. J. Wiatrak
Sod-Livestock Integration into the Peanut-Cotton Rotation: A Systems Farming Approach
Agron. J., June 27, 2006; 98(4): 1156 - 1171.
[Abstract] [Full Text] [PDF]

Home page
 Agron. J.Home page
D. Zhao, K. R. Reddy, V. G. Kakani, J. J. Read, and S. Koti
Selection of Optimum Reflectance Ratios for Estimating Leaf Nitrogen and Chlorophyll Concentrations of Field-Grown Cotton
Agron. J., January 1, 2005; 97(1): 89 - 98.
[Abstract] [Full Text] [PDF]

Home page
 Crop Sci.Home page
J. J. Marois, D. L. Wright, P. J. Wiatrak, and M. A. Vargas
Effect of Row Width and Nitrogen on Cotton Morphology and Canopy Microclimate
Crop Sci., May 1, 2004; 44(3): 870 - 877.
[Abstract] [Full Text] [PDF]

Home page
 Crop Sci.Home page
F. B. Fritschi, B. A. Roberts, R. L. Travis, D. W. Rains, and R. B. Hutmacher
Seasonal Nitrogen Concentration, Uptake, and Partitioning Pattern of Irrigated Acala and Pima Cotton as Influenced by Nitrogen Fertility Level
Crop Sci., March 1, 2004; 44(2): 516 - 527.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (8)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Bell, P. F.
Right arrow Articles by McConnell, J. S.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Bell, P. F.
Right arrow Articles by McConnell, J. S.
Agricola
Right arrow Articles by Bell, P. F.
Right arrow Articles by McConnell, J. S.
Related Collections
Right arrow Cotton
Right arrow Plant Analysis
Right arrow Plant Nutrition

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Vadose Zone Journal
Journal of Natural Resources
and Life Sciences Education		 Soil Science Society of America Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome