RESPONSE OF ONION (Allium cepa L.) TO SOWING DATE AND PLANT POPULATION

The response of onion cv. Jaquar (Allium cepa L.) to sowing date and plant population was investigated in Bloemfontein (Free State Province, South Africa) during the year 2010.  Jaquar seeds were sown on 11 and 25 May; and 8 June at five different populations (95, 83, 74, 67 and 61 plants m-2) in a randomized complete block design with each treatment combination replicated three times.  Plant height and leaf number were measured 18 weeks after emergence.  Bulb yield and quality (bulb mass, diameter, firmness and neck diameter) were recorded at harvesting.  Significantly taller plants were obtained with the early sowing date (11 May) and a population between 61 and 74 plants m-2.  Leaf production was not significantly influenced by any treatment.  Bulb mass, yield and firmness were significantly higher with early sowing dates (11 or 25 May).   Plant population significantly influenced bulb mass, but not yield.  Bulb diameter increased as population decreased, irrespective of planting date.  The three lowest populations (74, 67 and 61 plants m-2) yielded significantly harder bulbs than denser plants (83 and 95 plants m-2).  The best sowing date for Jaquar in Bloemfontein was from 11 to 25 May at a population of between 61 and 74 plants m-2. 
 
   
 
 Key words: Bulb diameter, bulb firmness, bulb fresh mass, bulb shape, leaf number, plant height, yield.


INTRODUCTION
The performance of onion (Allium cepa L.) is greatly influenced by environmental factors (photoperiod and temperature) and agronomic practices.The primary objective of any producer is to attain high yields and for onions the final yield depends on the amount of vegetative growth before bulb initiation (Adjei-Twum, 1980 cited by Ibrahim, 2010).Sowing date and plant population are the two major factors that influence the vegetative growth of onions.Early sowing of onions results in more vegetative growth (leaf length and quantity) (Mulungu et al., 1998), which ultimately reflects in the leaf area index (LAI) (Brewster, 1994).The yield of late sown plants will be lower than earlier sown plants because leaf blade production switches to bulb initiation while the LAI and light interception is still low (Pakyϋrek et al., 1994;Brewster, 2008).Therefore, sufficient time is needed for leaf production before bulb initiation.Both early sowing and an increase in plant population will increase the LAI (Mondal et al., 1986).
As plant population increases, onion bulb yield also increases because the leaf canopy intercepts a higher percentage of light (Brewster, 2008).In a field experiment done in Skierniewice, onion bulb yields of between 31 and 59 t ha -1 were recorded with a plant population ranging between 80 and 100 plants m -2 when grown from seed (Rumpel and Felczyński, 2000).However, with a plant population lower than 60 plants m -2 or more than 100 plants m -2 onion yield decreased.Kanton et al. 2002 also reported a yield increase with higher onion plant populations in Northern Ghana.Plant populations ranged from 37.04 to 156.25 plants ha -1 and the highest yield was recorded for a population of 76.92 plants ha -1 .
Bulb size is an important quality characteristic and different markets require different bulb sizes.According to Brewster (1994), a plant population between 50-100 plants m -2 will produce bulbs that are between 50-70 mm in diameter, which are preferred in most markets (Lai et al., 1994;Rumpel and Felczyński, 2000;Bosch Serra and Currah, 2002), while at a plant population of 25-50 plants m -2 large bulbs (>70 mm) suitable for the processing industry will be produced (McGeary, 1985;Shock et al., 2005).
Split bulbs downgrade onion bulb quality and are common in large bulbs, normally resulting from larger plants, and this is also associated with a lower plant population and a too early planting date.In Bangladesh, Khan et al. (2003) reported that at a low plant population (48 plants m -2 ), 24.34% of the total bulb yield split, whereas only 14.38% split when bulbs were planted more densely with a population of 163 plants m -2 .Larger plants easily form split bulbs because of the development of more leaf bases further away from the apical meristem that eventually produce split bulbs (Rabinowitch, 1979).Plant population also influences the shape of onion bulbs, which also influences the marketability of onions.Round bulbs are preferred in South Africa (Eksteen et al., 1997).An increase in plant population leads to elongated bulbs (Hygrotech, 2010).
In the central Free State province of South Africa, mostly intermediate cultivars are sown.In the past few years, the onion market underwent a major turn around with the release of a number of new short, early and midintermediate cultivars.The new onion cultivars were introduced so that the sowing season would be lengthened from February until the end of June in the central areas of South Africa (Northern Cape and Free State provinces).Onion is one of the vegetable crops that are very sensitive to sowing and transplanting dates.Each cultivar, soil type and micro climate combination can result in different growth and yield responses and it is therefore important that all plantings start with the correct sowing date.
In a field experiment conducted in 2009 and repeated in 2010, four onion cultivars best suited for central South Africa (Southwood, 2009) (Bosekeng and Coetzer, 2013).
The aim of this study was therefore to investigate the effect of delayed planting dates on Jaquar and to determine whether plant population could compensate for yield losses that may have occurred with late planting dates in Bloemfontein (central Free State Province, South Africa).

MATERIALS AND METHODS
A field experiment was conducted at the University of the Free State in Bloemfontein (South Africa) in the year 2010.Bloemfontein (29°06'S and 26°18'E, 1395 m above sea level) is located in the Central Free State on the southern edge of the Highveld within the semi-arid climate of the revised Köppen Climate Classification (Kruger, 2004).Summers are hot (annual average maximum temperature 24.4°C), and winters cold and dry, often with severe frost (annual average minimum temperature of 7.5°C).It is a summer rainfall area with an annual average rainfall of 559 mm.Table 1 shows a summary of weather conditions during the production season.
The onion cultivar Jaquar, was selected based on performance in a previous field experiment conducted during 2009 as was explained earlier where different cultivars were sown on different planting dates in Bloemfontein (Bosekeng and Coetzer, 2013).Seeds of the cv.Jaquar were sown on three sowing dates (11 May, 25 May and 8 June) at five different plant populations (95,83,74,67 and 61 plants m -2 ) (Table 2).The experiment was laid out as a randomized complete block design, with each treatment combination replicated 3 times.
Soil samples were collected according to the guidelines of the Fertilizer Society of South Africa (Fertilizer Society of South Africa, 2007), from a depth of 15 cm prior to land preparation, dried and sent for analysis.Analyses were carried out according to standard methods of The Non-Affiliated Soil Analysis Work Committee (1990).Some chemical and physical properties of the soil are indicated in Table 3.The soil is a sandy loam of the Bainsvlei form (Soil Classification Working Group, 1991), or Plinthustalf according to the Soil Survey Staff (2003).
Results of the analyses were used to determine nutrient requirements according to the nutrient withdrawal amounts for onions (Fertilizer Society of South Africa, 2007) at a targeted yield of 70 t ha -1 .These withdrawal amounts were 5, 0.5 and 2.4 kg ha -1 for N, P and K, respectively.According to the soil analysis, the phosphorus level of the soil was below the optimum for vegetable production and an additional 11 kg P ha -1 was applied.Phosphorus (46 kg ha -1 ) and nitrogen (138 kg N ha -1 ) was broadcast and incorporated into the soil for the total experimental area before sowing using 3.1.0(28).All the potassium (280 kg K ha -1 ) was applied before sowing using potassium chloride.A split application of the residual nitrogen was done using urea at 6 (106 kg ha -1 ) and 12 weeks (106 kg ha -1 ) after sowing.This was calculated and applied per plot to assure that each plot received the same amount of fertilizer.
Before sowing, the soil was watered to field capacity.Seeds were sown by hand and covered with soil to a depth of 2 cm.A light irrigation was applied after sowing, using surface drip system with emitters spaced 30 cm in the line and with a delivery capacity Table 1.Monthly average maximum and minimum temperatures, photoperiod and total rainfall from the weather station for 2010 (South African Weather, 2010;South African Astronomical Observatory, 2010).

Month
Temperatures of 2.1 L h -1 .A drip irrigation system was used to keep the soil as close as possible to field capacity and to assure each plot received the same amount of water.Irrigation (1.78 mm) was done on a daily basis, except on days that it rained, to prevent soil from forming a crust that would delay emergence (Mondal et al., 1986).
After emergence, irrigation (3.25 mm) was done every second day during the vegetative phase and daily (2.33 mm) from the start of bulb development until maturity.The seedlings were thinned to the required plant population (Table 2) three weeks after emergence.Weeds were controlled manually every week throughout the season.Sodium fluosilicate (Cutworm Bait-Efekto, Private Bag 215, Bryanston, 2021 South Africa) was broadcast at a rate of 10 g m -2 over the planted area as soon as the first seedlings emerged.Mercapthothin (Malasol-Efekto, Private Bag 215, Bryanston, 2021 South Africa) was applied at a rate of 17.5 ml 10 L -1 H 2 O using a knapsack sprayer whenever aphids or thrips were noticed on the onion plants.
Growth parameters were measured on 11 randomly selected plants located in the center of each plot.Growth data were collected every three weeks, starting from the third week after emergence until 10% of the foliage collapsed.Only final plant height and leaf number 18 weeks after emergence will be reported on in this study.Plant height was measured from the ground level to the tip of the highest leaf using a standard ruler.The numbers of fully developed active green leaves were counted.The focus of this study was mainly on the sellable product and therefore only yield (fresh mass) and external quality parameters were measured.External parameters that were measured such as bulb diameter (size) and shape play a major role in marketing, as well as neck diameter and bulb firmness in the storage life of the bulbs.These yield and quality parameters were assessed on bulbs from net plot size (Table 2) of the middle rows in each plot.Onions were lifted by hand when 100% of the foliage top per plot had collapsed.If necessary, the soil was first loosened using garden fork.Onion bulbs were transported immediately to the laboratory and not left to cure.
After harvesting, leaves and roots were cut from the bulbs, and bulbs cleaned by washing with water and dried with tissue paper.Bulbs were weighed separately to obtain fresh mass per bulb of 11 randomly selected bulbs harvested from the net plot.The fresh mass of all bulbs harvested from net plot size was expressed as g m -2 , and then converted to t ha -1 in order to indicate onion yield.The same 11 bulbs were also used to measure the bulb quality parameters (diameter, firmness, neck diameter, defects and shape).Bulb diameter was measured at right angles to the longitudinal axis at the widest circumference of the bulb using a digital caliper.Thereafter, the bulbs were graded into extra small (10 to 34 mm), small (35 to 39 mm), medium (40 to 69 mm), large (70 to 89 mm) or extra-large (≥ 90 mm) bulbs (Joubert et al., 1997).The firmness of bulbs was measured using a Seta-Matic penetrometer controller (Model 1720), automatically controlled with a constant load penotrometer (Model 1719 of Stanhope seta Ltd., England).Bulb defects such as decay, split and mechanical damage were observed to aid in determining marketable bulbs.Bulb shape was determined by observation using the norms of Boyhan and Kelley (2008).
Analysis of variance was calculated on all measured parameters to determine the significance of differences between means of treatments using the NCSS 2000 statistical program (Hintze, 1999), and Tukey's test for the LSD ≤ 0.05, except where stated.

RESULTS
The growth period for Jaquar (early intermediate cultivar) for the different sowing dates and plant populations are indicated in Table 4.

Plant height and leaf number
The interaction between plant population and sowing date did not significantly affect plant height or leaf number (Table 5).Although not always significant, plant height decreased as plant population increased.Plants planted at 95 plants m -2 were significantly shorter (46.40 cm) than plants from the three lowest populated plantings of 61, 67 and 74 plants m -2 (55.07, 51.13 and 52.21 cm, respectively).However, plants planted at lower densities of 61, 67 or 74 plants m -2 , did not significantly differ in height from each other.Onion plants sown early (11 May) were significantly taller than plants sown on 25 May or 8 June.The total number of onion leaves produced was not significantly influenced by either plant population or sowing dates (Table 5).

Bulb fresh mass and yield
The interaction between plant population and sowing date did not significantly influence onion bulb fresh mass and yield.
Plant population, however, significantly influenced bulb fresh mass but not yield (Table 6).Bulb fresh mass decreased as plant population increased with the bulbs from plants at a density of 61 plants m -2 being significantly heavier than those planted at 83 or 95 plants m -2 .Bulbs harvested from those planted at the lowest population (61 plants m -2 ) were 38.46 and 28.29% heavier than bulbs of plants planted at 95 or 83 plants m - 2 , respectively.Fresh mass of bulbs harvested from plants sown at a density of between 61 and 74 plants m -2 did not differ significantly from each other.Even though plant population did not significantly influence yield, the highest yield was obtained from a plant population of 74 plants m -2 (38.68 t ha -1 ) and the lowest from a plant population of 95 plants m -2 (36.00 t ha -1 ).Yield increased as plant population decreased from 95 to 74 plants per m - 2 , but with a further decrease in plant population bulb yield tend to decrease again, although not significantly.

Bulb diameter (size)
The interaction between plant population and sowing date significantly affected onion bulb diameter (Table 7).As plant population increased from 61 to 95 plants m -2 the size of bulbs decreased, irrespective of sowing date.All the bulbs were graded as medium size bulbs (40-69 cm), with the exception of bulbs harvested from plants sown on 11 or 25 May at the highest population planting of 95 plants m -2 .Bulb diameter responded differently to sowing date and plant population.Significantly larger bulbs were produced by plants at a density of between 61 and 67 plants m -2 when sown on 11 May; and those sown on the 8 June were only significantly larger when sown at 61 plants m -2 compared to that of bulbs from plants sown more densely between 83 and 95 plants m -2 .Bulbs harvested from the two lowest densities (61 or 67 plants m -2 ) were significantly larger than bulbs at the three highest populated plantings (74, 83 or 95 plants m -2 ) when sown on 25 May.

Neck diameter
The interaction between plant population and planting

Bulb firmness
Bulb firmness was significantly affected by both plant population and sowing date.The interaction between these two factors was, however, not significant (

DISCUSSION AND CONCLUSION
Plant population significantly influenced plant height, bulb fresh mass, bulb diameter, bulb neck diameter and bulb firmness.As the plant population increased, plant height and bulb fresh mass decreased (Tables 5 and 6).Bulb quality (neck diameter and firmness) also showed a decrease with an increase in the plant population (Table 7).The reason for higher plants in less dense populations may be associated with competition between plants for available water and nutrients.Jilani et al. (2009)   (2007) investigated three plant populations (40, 60 and 80 plants 4 m -2 ) using three onion cultivars under Peshawar conditions.
They reported that as plant population increased from 40 to 80 plants 4 m -2 , plant height (leaf length) increased significantly from 47.79 to 52.51 cm.This was ascribed to competition for sunlight making the plants grow taller to exploit light to a maximum in denser plant populations.Conversely, the current study found that plant height decreased as the population increased from 61 to 95 plants m -2 .
The increase in competition for water and nutrients between plants under denser populations also reflected in bulb fresh mass, yield and diameter.The higher bulbfresh mass and larger bulbs in less dense populations may have been due to sufficient leaf growth and space for bulb development.In Northern Ghana, Kanton et al. (2002) transplanted onions (cv.Bawku Red) in the field using eight different plant populations (37.04 -156.25 plants m -2 ).They reported a decrease in bulb fresh mass as plant population increased from 37.04 to 156.25 plants m -2 .Bulbs planted at 37.04 plants m -2 were 128% heavier than bulbs at 156.25 plants m -2 .This phenomenon was also associated with sufficient leaf growth and space for bulb development (Kanton et al., 2002).Although yield was not significantly influenced by plant population in the current study (Table 6), yield increased as plant population increased from 61 to 74 plants m -2 (37.24 to 38.68 t ha -1 ), but with a further increase to 95 plants m -2 , yield decreased to 36.00 t ha -1 .The yield of onions sown directly at six different plant populations (20,40,60,80,100 and 140 plant ha -1 ) increased from 29.8 to 32.6 t ha -1 when plant population increased from 60 to 100 plants m -2 ; but with an increase to 140 plants m -2 , a decreased yield of 31.1 t ha -1 was reported (Rumpel and Felczyński, 2000).With a plant population lower than 60 plants m -2 , onion yield also decreased from 26.4 to 20.5 t ha -1 .According to these researchers, the decrease in bulb yield as plant population increases may be due to early maturation of bulbs.As bulbs mature earlier, the period for leaf growth is also short resulting in a smaller LAI with less efficient light interception (De Visser, 1994).This was not the case in the current study because all bulbs were harvested on 8 November irrespective of the plant population, except the 61 and 67 plant m -2 plantings, which were harvested a week later on 15 November (Table 4).Hatridge-Esh and Bennett (1980) planted onions at four different plant populations (7, 40, 80 and 100 plants m -2 ) in California and reported that bulb diameter decreased from 99 to 63 mm as plant population increased from 7 to 100 plants m -2 .This was caused by competition for nutrients and insufficient space restricting bulb enlargement when plant population increased.In the current study, the highest percentage of medium bulbs (81.8-93.9%)occurred at a plant population of 74 plants m -2 for all three planting dates.
Leaf number was, however, not significantly influenced by either plant population or sowing date and thus corresponded with the results of both Farooq-Ch et al. (1990) and Kanton et al. (2002).Farooq-Ch et al. (1990) stated that number of leaves produced by an onion plant is a genetic character, and is not influenced by plant population.Similarly, Bosekeng and Coetzer (2013) found that sowing dates did not significantly influence leaf number of Jaquar sown at a plant population of 61 plants m -2 .Sowing date significantly influenced all other parameters.As sowing date was delayed from 11 May to 8 June, plant height, bulb fresh mass, bulb diameter and neck diameter decreased, and bulbs were harder.In a separate field experiment during the same season where Jaquar was sown on the same dates at a population of 61 plants m - 2 , the same trends for the all parameters were recorded (Bosekeng and Coetzer, 2013).Onions will switch from leaf blade production to bulb initiation when the minimum day length requirements are met.For the same cultivar sown on different dates, bulb initiation will occur at more or less the same time.However, other factors such as temperature and plant size may also play a role (Brewster, 1994).When considering sowing date,day length and temperature data (Table 1 and 4) of the current study, the early sown plants (11 May) had a longer growth period of 118 days compared to 97 days (25 May) and 76 days (8 June) for the later sown plants.The average maximum temperatures during the different growth periods for the different sowing dates did not differ much, and ranged between 20.1 and 21.6°C (Table 4).These results are in agreement with those of Cramer (2003) who reported that early sown plants were taller than late sown plants due to a longer vegetative growth period.The longer growth period for early sown plants resulted in longer leaves (Table 5) that intercepted more light (Brewster 1994) and this eventually increased bulb fresh mass (Table 6), larger bulbs with thinner necks (Table 7) and softer bulbs.
In Eqypt, Leilah et al. (2003) reported that a yield of 46.95 t ha -1 was obtained with the December planting, followed by 38.10 t ha -1 for January and 31.80 t ha -1 for February.Jaquar is known for its globe shaped bulbs (Hygrotech, 2009).From the current study, none of the plant populations or sowing dates produced globe shaped bulbs.Bulb shape changed with plant population.At a population of 95 or 83 plants m -2 bulbs were spindle shaped (epllitic) and changed to egg shaped pointed (oval) as plant population decreased to 74 or 67 plants m -2 .Bulb shape changed from egg shaped pointed, to a broad reversed (top) shape with a further decrease in plant population to 61 plants m -2 .The reason for the change in bulb shape was the limited space available for each bulb to develop.Elongated bulbs are the result of squeezing against each other (Hygrotech, 2010).McGeary (1985) studied planted onion cv.White Spanish under six different populations (178, 400, 6325, 816, 111and 1600 plants m -2 and reported a decline in number of round bulbs (13.3%) when population increased from 178 to 1 600 plants m -2 .Current results showed that all the bulbs (100%) changed shape as plant population changed.The manipulation of plant population is a critical agronomic principle, which cannot be ignored as it plays a role in growth, yield and bulb quality (bulb mass, size and firmness) of onions.Jaquar produced the best growth, highest bulb fresh mass and yield; and best quality when sown from 11 to 25 May at a plant population of between 61 and 74 plants m -2 in Bloemfontein, Central Free State Province of South Africa.When sowing date is extended (8 June), a lower plant population must be used (61 plants m -2 ) to compensate for losses and to harvest a relatively marketable yield.
conducted a study using three planting populations (20, 30 and 40 plants ha -1 ) and five onion cultivars in Pakistan and recorded higher plants from a less dense population (20 plants m -2 ) than a denser population (40 plants m -2 ).However,Dawar et al.
were selected.These cultivars included one short, two early intermediate and one midintermediate onion cultivar.Jaquar (a mid-intermediate cultivar) performed the best in 2009 and this was confirmed in 2010

Table 2 .
Different plant population and sowing date treatments, plant spacing, plot size and plants per row used.

Table 3 .
Measured chemical and physical properties of the soil samples and optimum soil analysis values for vegetables according to the Fertilizer Society of SouthAfrica (2007)

Table 4 .
A summary of the growth period of Jaquar (early intermediate cultivar) as influenced by sowing date and plant population during 2010.

Table 5 .
Influence of sowing date and plant population on plant height (cm) and leaf number of the onion cv.Jaquar 18 weeks after emergence.: plant population; SD: sowing date; PP x SD: plant population interaction with sowing date; Avg: average; LSD: least significant difference; ns: not significant.Means with small cap (a, b and c) in the same column differ significantly from each other.Means with different cap letters (A, B and C) in the same row differ significantly from each other; P ≤ 0.05 PP

Table 6 .
Influence of sowing date and plant population on bulb fresh mass and yield of the onion cv.Jaquar.
PP: plant population; SD: sowing date; PP x SD: plant population interaction with sowing date; Avg: average; LSD: least significant difference; ns: not significant.Means with small cap (a, b and c) in the same column differ significantly from each other.Means with different cap letters (A, B and C) in the same row differ significantly from each other; P ≤ 0.05.

Table 7 .
Influence of sowing date and plant population on bulb diameter, neck diameter and bulb firmness of the onion cv.Jaquar., plant population; SD, sowing date; PP x SD, plant population interaction with sowing date; Avg, average; LSD, least significant difference; ns; not significant.Means with small cap (a, b and c) in the same column differ significantly from each other.Means with different cap letters (A, B and C) in the same row differ significantly from each other; P≤0.05. PP