Agronomy Journal Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 11 April 2006
Published in Agron J 98:562-568 (2006)
DOI: 10.2134/agronj2005.0181
© 2006 American Society of Agronomy
677 S. Segoe Rd., Madison, WI 53711 USA
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 (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Siebert, J. D.
Right arrow Articles by Leonard, B. R.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Siebert, J. D.
Right arrow Articles by Leonard, B. R.
Agricola
Right arrow Articles by Siebert, J. D.
Right arrow Articles by Leonard, B. R.
Related Collections
Right arrow Crop Growth and Development
Right arrow Cotton
Right arrow Seed Establishment

Production Papers

Comparative Growth and Yield of Cotton Planted at Various Densities and Configurations

Jonathan D. Sieberta,*, Alexander M. Stewartb and B. Rogers Leonardc

a Louisiana State Univ. AgCenter, Dep. of Agronomy and Environmental Management, 104 M.B. Sturgis Hall, Baton Rouge, LA 70803
b Louisiana State Univ. AgCenter, Dean Lee Research Station, 8105 Tom Bowman Drive, Alexandria, LA 71302
c Louisiana State Univ. AgCenter, Dep. of Entomology and Macon Ridge Research Station, 212 Macon Ridge Road, Winnsboro, LA 71295

* Corresponding author (jsiebert{at}agcenter.lsu.edu)

Received for publication June 15, 2005.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Cotton (Gossypium hirsutum L.) lint yield stability across a range of plant populations, coupled with expensive transgenic cotton seed, makes reduced seeding rates an attractive cost-saving option. Studies evaluated plant populations and seeding configurations in an effort to: (i) isolate a specific combination that minimizes seed use without sacrificing yield, and (ii) identify potential growth and development changes associated with cotton grown at these densities. Cotton planted in studies conducted during 2003 and 2004 on a Norwood silt loam (fine-silty, mixed, calcareous, thermic Typic Udifluvent) were hand thinned to densities ranging from 33 978 to 152 833 plants ha–1 in both hill-drop and drill-seeded configurations (96.5-cm row widths). Plant height, main-stem nodes per plant, maturity, boll retention by position, lint yield, and fiber quality were all evaluated. No seeding configuration x plant population interaction occurred for variables other than yield. A positive relationship existed between plant population and plant height; however, main-stem nodes per plant, days after planting to peak bloom, and boll retention were inversely related to plant density. Lint yield was highest for 152 883 plants ha–1 (1465 kg ha–1) planted in a hill-drop configuration with three plants per 20-cm hill spacing, and was not reduced until population was lowered to 33 975 (30.5-cm plant spacing, 1263 kg ha–1) and 50 958 (three plants per hill, 60-cm hill spacing, 1177 kg ha–1) plants ha–1. Treatments did not influence fiber properties. Reduced seeding rates appear to be a viable cost-saving option, given that a uniform stand is achieved and appropriate management practices employed.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
ESTABLISHMENT of an acceptable population of cotton seedlings is paramount to obtaining high yields (Christiansen and Rowland, 1981). The definition of an acceptable plant population, however, varies by location, environment, cultivar, and grower preference (Silvertooth et al., 1999). Current plant density recommendations in Louisiana are 10 to 13 plants m–1 of row for conventionally spaced cotton (96.5–101.6-cm row widths; Faircloth et al., 2002). These recommendations are based on research conducted by Boquet and Coco (1997) before the widespread acceptance of transgenic cotton cultivars.

Today seed premiums and technology-fees-associated transgenic cotton cultivars coupled with increased adoption of seed treatments in Louisiana for insect and disease control have resulted in increased at-planting variable costs. These variable costs may be offset through reduced seeding rates. In addition, seed-specific in-furrow application systems currently under development may result in pesticide savings of approximately 50% (Wilkerson et al., 2004). Reduced seeding rates may also have other management implications. For example, Leigh et al. (1974) reported higher numbers of Lygus spp. with increased plant populations. Formerly secondary pests of cotton, Lygus spp. now require multiple insecticide applications per season to achieve adequate control in the South and southeastern Cotton Belt (Williams, 2005). Thus, reduced seeding rates may have implications beyond simply saving seed.

Several researchers have concluded that seed cotton yield and plant density are unrelated (Ray et al., 1959; Hawkins and Peacock, 1973; Baker, 1976; Buxton et al., 1977; Jones and Wells, 1998; Bednarz et al., 2000; Franklin et al., 2000). Other researchers, however, have observed reduced yields with extremely high or low plant densities (Hawkins and Peacock, 1971; Bridge et al., 1973; Smith et al., 1979).

Interestingly, no one has ever investigated the interaction of plant population and seeding configuration on cotton lint yield. Many producers have adopted hill-drop seeding in an attempt to combat emergence problems in soils prone to crusting and achieve uniform plant spacing. Advances in crop planting equipment have improved the precision of seed placement and seeding rate. Some researchers have concluded that uniform seed placement is more important in maximizing yield potential than seeding rate (Lee, 1968; Wanjura, 1980). The objective of these studies was to evaluate plant population and seeding configurations on cotton growth, development, and yield in an effort to isolate a specific combination to minimize seed use without sacrificing yield and to identify any potential management variations that may exist with cotton grown at these divergent densities.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Two experiments evaluating plant population and seeding configuration were conducted at the Dean Lee Research Station near Alexandria, LA, during 2003 and 2004 on a nonirrigated Norwood silt loam soil. The seeding configuration study (Experiment 1) was a randomized complete block with four replications (Table 1). The plant population x seeding configuration study (Experiment 2) was a randomized complete block with an unbalanced factorial arrangement of treatments (Factor A: population; Factor B: seeding configuration) and four replications (Table 1). Plot size in both studies was four 96.5-cm-wide rows, 12.15 m long. All treatments were planted with cotton (cv. Paymaster 1218 BG/RR) seed using a four-row John Deere Max Emerge II vacuum planter (Moline, IL) with either 5.1 cm between seeds for drill seeded treatments or 20 cm between hills (four seeds per hill) for hill-drop seeded treatments. Plots were hand thinned 3 wk after emergence to their respective plant population. All data were recorded from the center two rows of the four-row plot. The entire experimental area was maintained using standard cultural practices based on extension recommendations by the Louisiana State University AgCenter.


View this table:
[in this window]
[in a new window]
  		Table 1. Seeding configurations and plant populations of studies conducted at Alexandria, LA, in 2003 and 2004.{dagger}

 
Plant height and number of main-stem nodes were recorded from five randomly selected plants within each plot weekly (after thinning). Cotyledons were counted as Node 0 and the uppermost node with a main-stem leaf >25 mm wide was considered the terminal node. Plant height and number of main-stem nodes per plant were used to calculate height/node ratio (average plant height/average no. of nodes). After anthesis, the number of white flowers per row and the number of main-stem nodes above the uppermost first-position white flower (NAWF) on five randomly selected plants per plot were recorded biweekly. White flower counts were terminated when NAWF = 5 (i.e., cutout). Several studies have indicated that a first-position white flower located five main-stem nodes below the terminal node (NAWF = 5) is the last boll likely to develop to maturity or contribute to yield; flowers set above this point contribute little to overall yield (Benson et al., 1999; Bourland et al., 1992; Jenkins et al., 1990). When approximately 60% of bolls were open, the experimental area was chemically defoliated with a tank mix of thidiazuron (N-phenyl-N'-1,2,3-thiadiazol-5-ylurea), tribufos (S,S,S-tributyl phosphorotrithioate), and ethephon (2-chloroethyl phosphonic acid). Following defoliation, 10 consecutive plants per plot were mapped to determine retention of fruiting forms by node and branch location. Fruiting positions were characterized as having either aborted or retained fruit.

The center two rows of each plot were harvested with a commercial two-row spindle picker fitted with a weigh cell capable of being tarred between plots. An approximate 0.9-kg subsample of seed cotton was retained from each plot and ginned on a 12-saw research gin to determine lint percentage. Treatment effects on the lint/seed ratio were not significant and all treatments were within 3% of the average for the entire experiment in both years (data not shown). Therefore an average lint percentage was used to calculate lint yields. Physical fiber properties were determined using a high-volume instrumentation method at the Louisiana State University AgCenter Fiber Laboratory, Department of Agronomy, Baton Rouge, LA (Sasser, 1981).

Plant height, number of main-stem nodes, and white flower counts were each plotted as a function of time (weeks after planting). The profiles of the lines generated from these data were each subjected to repeated measures analysis. Treatments were separated using pairwise comparisons with a Tukey–Kramer adjustment (PROC MIXED, SAS Institute, 1998). Plant mapping, lint/seed ratio, lint yield, and fiber data were subjected to analysis using the SAS MIXED procedure and means separated with Fisher's protected LSD ({alpha} = 0.05; PROC MIXED, SAS Institute, 1998). Significant interactions prevented combining data for plant height, main-stem nodes, and appearance of white flowers across years. Differences are attributed to a 25-cm increase in rainfall from 15 May to 31 Aug. in 2004 compared with 2003 (Fig. 1 ). Data for all other variables were combined across years.


Figure 1
View larger version (19K):
[in this window]
[in a new window]
  		Fig. 1. Weekly (bars) and total (lines) rainfall for May through August, 2003 and 2004. Weather data collected at the Dean Lee Research Station near Alexandria, LA.

 

   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Repeated measures analysis indicated that seeding configuration was not a significant effect across plant populations for all growth and development parameters measured in both years (Table 2). Limited information is available on the effect of intrarow seeding configuration at a given population on growth and development parameters. Hawkins and Peacock (1970) reported that plant population, as long as the stand is of uniform density, may be a more important factor than either spacing or number of plants per hill; and that boll and fiber characteristics were relatively stable across a wide range of planting patterns (Hawkins and Peacock, 1971).


View this table:
[in this window]
[in a new window]
  		Table 2. Response of cotton plants to seeding configuration (drilled vs. hill-drop seeded) within a given plant population in studies conducted at Alexandria, LA, in 2003 and 2004.

 
Plant Height and Main-Stem Nodes
In 2003, plant height was significantly lower for populations of 50 958 and 33 975 plants ha–1 than 152 883 plants ha–1 (Fig. 2 ). This trend in plant height was evident at the end of the season and was reflected in final plant height measurements. The plant population of 33 975 plants ha–1 produced plant heights significantly shorter than that of 152 883 plants ha–1. In 2004, these factors were not influenced by plant population. Differences between years were attributed to early season stresses from excessive rainfall and possibly poor root development. Contradictory effects of plant population on cotton height have been reported. Bridge et al. (1973) and York (1983) reported that populations in excess of 200 000 plants ha–1 decreased plant height. In Arizona, irrigated cotton grown in populations >300 000 plants ha–1 became tall, rank, and predisposed to lodging (Peebles and Hartog, 1956). Research with ultra-narrow row cotton (≤50.8 cm) at extremely high plant populations (≤620 000 plants ha–1) has shown to decrease plant height (Fowler and Ray, 1977). Our results support previous research that indicates cotton plant height increases with increasing populations only to a point, after which intraspecific competition between cotton plants for water, nutrients, and space presumably limits plant size. Total number of main-stem nodes was significantly lower for plant populations of 152 883 plants ha–1 than that of any other population evaluated in 2003 (Fig. 3 ). Buxton et al. (1977) and Kerby et al. (1990a, 1990b) noted that increasing plant density decreased the number of main-stem nodes per plant.


Figure 2
View larger version (18K):
[in this window]
[in a new window]
  		Fig. 2. Effect of plant population on cotton plant height, Alexandria, LA, 2003. Plant populations (plants ha–1) are averaged across drill and hill-drop seeding configurations. The p-value represents repeated measures analysis of plant height over the duration of data collection with the table representing treatment separation for repeated measures analysis, populations followed by the same letter are not significantly different ({alpha} = 0.05). Bars from left to right represent populations of 152 883, 76 466, 50 958, and 33 975 plants ha–1, respectively; final plant height of treatments followed by the same letter are not significantly different ({alpha} = 0.05).

 

Figure 3
View larger version (11K):
[in this window]
[in a new window]
  		Fig. 3. Effect of plant population on total number of main stem nodes per plant, Alexandria, LA, 2003. Plant populations are averaged across drill and hill-drop seeding configurations. The P value represents repeated measures analysis of total main-stem nodes; 33 975, 50 958, and 76 466 plants ha–1 do not significantly differ and are significantly greater than 152 833 plants ha–1 ({alpha} = 0.05).

 
Shorter plant height and increased number of nodes per plant resulted in a reduced height/node ratio and could reduce plant growth regulator use requirements and relate to less intensive crop management. In a study addressing cotton response to mepiquat (1,1-dimethylpiperidinium) chloride application, yields progressively decreased with increasing plant populations due to excessive vegetative growth (York, 1983); however, variations in plant response to mepiquat chloride at different plant densities with respect to plant height was not investigated.

Earliness
No significant differences were detected for number of white flowers per hectare during the bloom period for the populations evaluated; however, the date of peak bloom demonstrated that a significant delay in maturity was associated with the lower plant populations (Table 3). Using the period of days after planting to peak bloom, a 4- and 5-d (2003) and 13- and 14-d (2004) delay in peak bloom was associated with 50 958 and 33 975 plants ha–1, respectively, when compared with 152 883 plants ha–1. Heitholt (1995) reported that plant density had little effect on flower numbers (20 000–200 000 plants ha–1) and Jones and Wells (1997) supported this by stating there were no differences in total flowers per meter or flower retention (20 000–120 000 plants ha–1); however, that study showed that plants in low populations had more bolls on monopodia and more distal sympodial positions, more late-season flowers, and greater retention of these bolls, which contributed to delayed crop maturity (an average 16-d delay to peak bloom for 2 plants m–1). A delay in crop maturity associated with a low stand density (33 969 plants ha–1) was also noted in Arkansas under irrigated conditions (Smith et al., 1979).


View this table:
[in this window]
[in a new window]
  		Table 3. Plant population effect on peak bloom and boll distribution at Alexandria, LA, in 2003 and 2004.

 
Boll Distribution
The effects of plant population on delaying maturity can be explained by variations in fruiting patterns (Table 3). As plant population decreased, total bolls per plant, first-position bolls per plant, and second, third, and monopodial bolls per plant increased. Averaged across both years, there was a 2.25-fold increase in the total number of bolls per plant at 33 975 plants ha–1 compared with 152 883 plants ha–1. This increase in total bolls is mainly attributed to a 3.0-fold increase in the number of second- and third-position sympodial and monopodial bolls per plant, but the number of first-position sympodial bolls increased only 1.5-fold. These results are similar to those of Jones and Wells (1998). The higher number of bolls at distal locations from the main-stem may result in an appreciable delay in maturity. During the end of the season, bolls require more time to accumulate heat units as average daily temperatures decline.

Guinn et al. (1981) reported fewer flowers and higher boll retention in low plant populations with no appreciable delay in maturity. Staggenborg and Krieg (1993) found plant population had little or no effect on boll retention. In our study, the higher number of bolls per plant associated with lower plant densities may be related to LAI (leaf area index), PPFD (photosynthetic photon flux density), or efficiency of solar radiation utilization. Buxton et al. (1977) noted that increases in cotton plant density increased LAI; however, those plants also exhibited a lower photosynthetic rate per unit leaf area due to mutual shading (Pegelow et al., 1977). Previous work showed LAI required to maximize PPFD interception (>90%) was obtained by cotton canopies by 83 d after planting regardless of row spacing or plant density, but the efficiency of PPFD interception per unit leaf area was greater at low plant densities (Heitholt, 1994). Cotton leaves on plants in high populations have lower total available carbohydrate levels than leaves of plants in low densities (Saleem and Buxton, 1976). This effect may be a result of poor assimilate partitioning due to photomorphogenic responses and the greater relative partitioning of photosynthate into leaf biomass (Heitholt, 1994). Therefore, cotton plants in low densities could potentially maintain a higher boll retention per plant than plants in higher densities.

Lint Yield and Fiber Properties
No significant year x treatment interactions were present and data were combined across years. In the seeding configuration study, significant yield reductions were not observed with a population of 67 952 plants ha–1 (four plants per hill, 60-cm hill spacing) compared with 101 929 plants ha–1 planted in drill or hill-drop seeding configurations (Table 4).


View this table:
[in this window]
[in a new window]
  		Table 4. Plant population and seeding configuration effect on lint yield and physical fiber properties at Alexandria, LA, in 2003 and 2004 (seeding configuration study).{dagger}

 
In the population x seeding configuration study, the highest lint yield (numerically) was obtained with 152 883 plants ha–1 (three plants per hill, 20-cm hill spacing) at 1465 kg ha–1 (Table 5). Lint yields were not significantly different among drill-seeded treatments regardless of population. Yield of the lowest population (33 975 plants ha–1, drilled, 30.5 cm between plants) was significantly below that of the highest yielding treatment (152 883 plants ha–1, three plants per hill, 20-cm hill spacing) — 1264 and 1465 kg ha–1, respectively — but was not different than yield from the equivalent density (152 883 plants ha–1) in a drill-seeded configuration (1399 kg ha–1). Yield was similar for hill spacings of 20 and 40 cm, but was significantly reduced at 60 cm. Hawkins and Peacock (1970, 1971) also reported higher yields with 20- and 40-cm hill spacing than with plants on 60-cm hills. Although not significant, yields tend to increase as the number of seeds per hill increased from three to five seeds when hills were spaced at 60 cm (Hawkins and Peacock, 1970). This may explain why 60-cm hill spacing with four plants per hill was not significantly different from other treatments in the seeding configuration study, but a significant yield reduction did occur with 60-cm hill spacing with only three plants per hill in the population x seeding configuration study.


View this table:
[in this window]
[in a new window]
  		Table 5. Plant population and seeding configuration effect on lint yield and physical fiber properties at Alexandria, LA, in 2003 and 2004 (population x seeding configuration study).{dagger}

 
Recent studies in Texas (Franklin et al., 2000), Georgia (Bednarz et al., 2000), and North Carolina (Jones and Wells, 1998) have shown no differences in yield due to plant population (64 531–129 111 plants ha–1, 38 623–276 983 plants ha–1, and 21 518–129 111 plants ha–1, respectively). Our studies have evaluated "stacked" gene transgenic cotton varieties, which, although very similar to their recurrent parent cultivars, are not genetically identical, and yield "drag" or decreased performance of cultivars after gene introgression was an initial concern (York et al., 2004). Recent studies by Nichols et al. (2004) have shown that glyphosate [N-(phosphonomethyl)glycine]-tolerant transgenic cultivars grown in ultra-narrow row spacing (<38 cm) had lint yields equal to or higher than conventional (non-transformed) cultivars in 2 of 3 yr. Our studies have addressed the response of transgenic cultivars grown at various plant densities in conventional row spacing (96.5 cm) in the environment of the lower Mississippi Delta.

Fiber properties including micronaire, staple length, fiber strength, and uniformity were not significantly influenced by plant population or seeding configuration in either study. Baker (1976), Bridge et al. (1973), and Hawkins and Peacock (1971) reported that fiber length, strength, and elongation were unaffected by plant population; however, micronaire tended to increase as population decreased (Bridge et al., 1973; Jones and Wells, 1998).


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
Considerable research efforts have been ongoing for >100 yr to determine the optimum plant population for maximum yield and quality in upland cotton. Many studies report that the highest yields occur in plant populations ranging from 49 000 to 256 000 plants ha–1 (Kittock et al., 1986). Our results show that maximum yields can be obtained with plant densities between 33 975 and 152 883 plants ha–1 if planted in a drill-seeded configuration or hill-drop configuration with hill spacing not to exceed 40 cm. Yield stability across populations is attributed to a greater number of bolls per plant at lower populations, the majority of which are monopodial and outer position sympodial bolls. Shorter plant height and a greater number of main-stem nodes associated with lower plant densities may reduce plant growth regulator requirements, resulting in less intensive crop management. A greater number of outer position bolls, however, can result in delayed maturity and may cause problems with heat unit accumulation in a short growing season. A compromise between easier crop management and the delay in maturity must be made when selecting an appropriate seeding rate to achieve a desired final plant population.

As seed and technology costs continue to rise, it is likely that producer interest in reducing seeding rates will also increase. Adverse weather conditions coupled with reduced seeding rates will inevitably result in extremely low plant populations and increase the likelihood of replanting to obtain an acceptable plant density. These data indicate that cotton plant populations can be lowered, given planting and environmental conditions conducive to achieving uniform plant distribution, to 50 958 plants ha–1 drilled or 76 466 plants ha–1 hill-dropped (3 plants hill–1, 40-cm hill spacing) with no adverse effects on yield. Future investigations need to address the importance of plant distribution in suboptimal stands.


   ACKNOWLEDGMENTS
 
We would like to thank the faculty and staff at the Dean Lee Research Station for assistance in conducting these studies, Dr. David Blouin for assistance with statistical analyses, and Cotton Incorporated for financial assistance. This manuscript was approved for publication by the director of the Louisiana Agricultural Experiment Station, manuscript no. 05-52-0302.


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




This article has been cited by other articles:


Home page
 Agron. J.Home page
J. D. Siebert and A. M. Stewart
Influence of Plant Density on Cotton Response to Mepiquat Chloride Application
Agron. J., October 31, 2006; 98(6): 1634 - 1639.
[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 (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Siebert, J. D.
Right arrow Articles by Leonard, B. R.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Siebert, J. D.
Right arrow Articles by Leonard, B. R.
Agricola
Right arrow Articles by Siebert, J. D.
Right arrow Articles by Leonard, B. R.
Related Collections
Right arrow Crop Growth and Development
Right arrow Cotton
Right arrow Seed Establishment

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Crop Science 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