Location and experimental design

This experiment was carried out in the second half of 2015 in the municipality of Puerto Gaitan, department of Meta, at the Taluma Research Center of the Colombian Corporation for Agricultural Research (AGROSAVIA), located at 4° 22’ 41” North latitude 72°13’25” west longitude and an altitude of 330 meters above sea level. A randomized complete block design (DBCA) was used with two treatments consisting of two distances between rows 40 and 80 cm and four repetitions. The 40 cm treatment is known as ‘ultra-narrow groove’ and the 80 cm treatment is the conventional planting distance. The experimental unit was constituted of 10 furrows of 6 m in length with an equal population density for the two treatments of 75 000 plants ha^-1.

Climate characterization associated with the phenology of cotton

A precipitation-evapotranspiration plot was constructed during the crop cycle (sowing to harvest) and the maximum, minimum and average temperature data were recorded. The time in days elapsed and the thermal index (IT) were determined as the units of heat accumulated in the stages of vegetative and reproductive development of cotton: 1) vegetative phase (stage of formation of roots, stem, leaves and branches); 2) youth phase (stage of reproductive growth with the appearance of flower buds; 3) early reproductive phase (appearance of buttons, flowers and capsules); 4) full reproductive phase; 5) final reproductive phase; and 6) maturity and harvest phase. The thermal index or heat units was calculated based on the formula: T_max + T_min/2-15.5.

Agronomic management

To know the level of fertility of the soil and derive the recommendation of fertilization, soil samples were taken at three depths in cm (0-10, 10-20, 20-30), in order to determine the physical characteristics: texture (Hydrometer or Bouyoucos method) and chemical: pH (potentiometer method); organic matter (%) (Walkley and Black method); phosphorus (mg kg^-1) (Bray II method); calcium (Ca), magnesium (Mg) and potassium (K), expressed in (cmol kg^-1) extraction with ammonium acetate pH 7 and quantification by atomic absorption) and base saturation (%) (calculated using the formula Ca + Mg + K/CIC). These analyzes were performed in the AGROSAVIA soil, water and plant chemistry laboratory with the technical quality standard NTC-2005.

A balanced fertilization in kg ha^-1 of N (140), P (80), K (80), S (1.2), B (0.5), Cu (0.5), Si (3.4) and Zn (3) was applied, applied at different times and fractionated: a) in planting, all phosphorus, one third of potassium and all minor elements were applied together with sulcamag; b) 15 days after the emergence (dde) of the culture, one third of the potassium and 40% of the N dose was applied; and c) a third application with 30% of the N and a third of the potassium was made at 30 days and a fourth fertilization was made at 45 days before flowering 30% of the remaining N.

Agronomic evaluations

The following agronomic, productive and fiber quality variables were determined: 1) plant height determined in 20 random plants, on the main stem, from the cotyledonal knot to the terminal bud of the plant and expressed in centimeters; 2) specks per plant, determined in 20 plants taken at random and expressed in number; 3) cotton-seed yield (RAS), determined by adding the weight of two passes of the eight-row crop, expressed in kg ha^-1 of seed cotton; 4) yield of cotton-fiber, obtained by ginning the cotton-seed and expressed in kg ha^-1; and 5) fiber extraction, determined by the difference in total speck weight and seed weight, expressed as a percentage.

In each experimental unit a sample of 50 grams of cotton fiber was taken to determine: 6) micronaire, as a measure of the fineness and maturity of the fiber with the following classification: extra fine (value less than 3), fine (3- 3.6), average (3.7-4.7), rough (4.8-5.9) and very rough (greater than 5.9); 7) fiber length, as the average length of half composed of longer fibers, in hundredths of an inch with the following classification: Extra short fiber, short, medium, long and extra-long; and 8) fiber length uniformity index, expressed as a percentage and defined with the following classification: very low uniformity (less than 77), low (77 to 79.9), average (80 to 82.9), high (83 to 85) , and very high (greater 85). The quality variables fiber was determined at the Laboratory Fiber Quality, Diagonal-Medellin adjusted to NTC Technical Standard 481-1.

Statistical analysis

With the data of the eight variables, the verification test of the assumptions of the variance analysis was carried out using the Proc Univariate procedure and the analysis of variance with the SAS Proc Anova procedure, version 9.3 (^{SAS, 2003}) and the comparison of means by Tukey test (p≤ 0.05).

Characterization of the environment: soil and climate

The floor of the oxisol order presented saturation greater bases 60% in the first 20 cm soil depth associated with a high content of organic matter and phosphorus. In soil depth greater than 20 cm, the base saturation did not exceed 20% with a low content of organic matter and phosphorus. The acidity condition of 5.8 and 5.6 in the 10 and 20 cm, respectively, of the soil were favorable for root development and cultivation, in accordance with the requirements established for cotton 6 to 7.5 (optimal 5.6) (^{FAO, 1994}) (Table 1).

This condition of soil preceded with a rotation of corn-soybean of three cycles to the sowing of cotton allowed a favorable condition for the proper nutrition of the plant, one of the indicators was the reported yield of corn and soybeans that ranged between 5.6 and 7.2 t ha^-1 and 1.9 to 2.1 t ha^-1, respectively (^{Campuzano et al., 2015}).

The subregion ‘altillanura plana’ presented favorable edaphoclimatic conditions for the vegetative and reproductive development of cotton. The cultivation altitude of 330 meters above sea level was adjusted to the optimum cotton requirement of 0-600 meters above sea level (^{Robles, 1991}; ^{FAO, 1994}) and the rainfall during the crop cycle exceeded evapotranspiration favorably during the months of August, September, October and the middle of November, with synchrony of phenological phases with high water requirement (vegetative phase, juvenile, early and final reproductive phase) and evapotranspiration exceeded precipitation in the last two weeks of November and in the months of December and January, favorable condition for the maturation phase of the cotton and harvest speck (Figure 1).

Figure 1 Precipitation and evapotranspiration during the crop cycle of Corpoica Menegua cotton (sowing in August to harvest in January) in Puerto Gaitán (AGROSAVIA, 2015). 

During the crop cycle, maximum temperatures were presented in °C from 30.2 to 31.8 in the vegetative and harvest states, respectively, minimum temperature in °C from 20.8 to 22.6 in the juvenile and harvest states, respectively and average temperature in °C from 25.1 to 26.4 in vegetative state and harvest, respectively, values all appropriate for the vegetative development, opening and maturation of acorns (Figure 2) (^{Baradas, 1994}).

Figure 2 Maximum-minimum-average temperature during the vegetative and reproductive phenological stages of Corpoica Menegiua cotton (AGROSAVIA, 2015). 

The precipitation of 700 mm in all vegetative and reproductive states was satisfied according to the requirement for cotton from 700 to 1 300 mm (^{Baradas, 1994}) and there was no rainfall in the period of development and maturation of acorns. The number of days from sowing to harvest was 140 days with a thermal index of 1 845 appropriate for the growth and development of cotton (^{Young et al., 1980}) (Table 2).

Productive behavior: plant height, specks plant^-1, yield cotton seed and fiber and fiber extraction

The variables of productive behavior presented highly significant differences for specks/plant, yield cotton seed and fiber, significant for plant height and not significant for fiber extraction. The coefficients of variation in all cases did not exceed 10%, which indicates the good degree of reliability of the data. Plant height, number of spots per plant and seed and fiber yield showed statistically different averages (p≤ 0.05) due to the effect of planting distance. The lowest plant height of 98.7 cm was 40 cm apart between rows, statistically different (p≤ 0.05) from that obtained with the distance of 80 cm of 113.4 cm (Table 3).

The number of spots per plant was statistically different (p≤ 0.05) and higher with the planting distance of 80 cm (18.5) in relation to the narrow groove of 40 cm (13.1). The same situation was presented with the yield of cotton seed and fiber. The conventional planting distance of 40 cm showed a yield of cotton seed in kg ha^-1 of 2 832 and seed of 1 143 statistically higher (p≤ 0.05) than that observed with the narrow distance of 40 cm rows with a yield of 2 181 seed cotton and 869 fiber cotton. The percentage of fiber extraction did not show statistical differences due to the distance between rows. The 40 cm fiber extraction was 40.4% statistically equal (p≤ 0.05) to that obtained with the 80 cm distance between rows of 40.9% (Table 3).

Fiber quality behavior: micronaire, fiber length and uniformity index

The analysis of variance did not show statistical differences for micronaire, fiber length and uniformity index. The coefficients of variation in all cases did not exceed 5%, which indicates the good degree of reliability of the data. Micronaire presented statistically equal values of 4.21 and 4.20 for the planting distance of 40 and 80 cm. A similar situation was presented for fiber length with values of 1.12 and 1.14 and information index of 82.1 and 81.91 for planting distances of 40 and 80 cm, respectively in each variable (Table 4).

The change in the distance between furrows in the Menegua cotton variety, from 80 to 40 cm, presented an increase in percentage of plant height of 13.1% and a significant reduction in the number of specks (29.2%), in yield of cotton seed (23.1%) and in the yield of cotton seed (23.9%). The behavior of the height of the Menegua plant in conditions of acid soils in the Colombian altillanura plana is a desirable attribute that was reported by ^{Campuzano et al. (2015)}, which leads to the non-use of the growth regulator as a favorable factor in production costs and environmental sustainability.

This result does not correspond to that reported by ^{Nichols et al. (2004)} who found that plant height decreased as the distance between rows is shortened. It is possible that this situation is due to differences in the type of plant which, in the case of Menegua, is a low bearing variety with a semi-compact structure (^{Campuzano et al., 2015}).

The lower plant height is an important condition for cotton in the loft since it does not allow the use of hormonal control. Similar research in other latitudes shows that narrow furrows do have benefits in cotton production, different from that obtained in this study (^{Gerik et al., 1998}; ^{Palomo-Gil et al., 2003}) without affecting the quality of the fiber by ultra-narrow groove effect. The percentage of fiber extraction, as well as micronaire, fiber length and uniformity index were not affected by the ultra-narrow groove. However (Heitholt et al., 1993; Gaytán et al., 2004), they indicate otherwise, with a reduction in fiber length and uniformity due to the ultra-narrow groove.

The discrepancy of these results with the report of other authors regarding the higher performance of cotton with the conventional furrow and the absence of differences in micronaire variables, fiber length and uniformity index could be related to the study management conditions. in relation to nutrition and water. It is known that ultra-narrow grooves have a higher nutritional and water requirement (^{Ramírez-Seañez et al., 2012}), it is possible that these factors were not satisfied in the treatment of ultra-narrow groove during the crop cycle, in relation to the length of fiber, the maximum temperature as a climate factor that affects the length of the fiber, under the conditions of the study were appropriate.