MATERIALS AND METHODS

Experimental site: The study was conducted in plastic tunnel at Egerton University, Kenya (at latitude of approximately 0° 23' South, Longitude of 35° 35' East and altitude of 2200 m above the Sea Level). The mean temperature, humidity, evaporation rate and rainfall over ten year period (1991 to 2000) are about 19.68 °C, 62.45%, 3.93 mm day^-1 and 907 mm year^-1, respectively (Egerton University Meteorological Station, 2000). The soil pH (H₂O) and (0.01M CaCl₂) at 0-30 cm depth was 6.26 and 5.37, respectively. The soil texture was clay loam (40% clay, 31% silt and 29% sand).

Experimental design and treatments: The experimental design was split-plot in randomized complete block replicated three times. The main plot factor was irrigation schedule, which consisted of irrigations on daily basis and after every two and three days, respectively. Sub-plots were made of mulching materials, which consisted of no mulch clear (transparent), plastic mulch (0.014 mm thick) and dry grass. The main grass was Hyperrhenia spp. Individual sub-plots measuring 1.6 m wide and 3.2 m long were planted with 'Money Maker' indeterminate tomato seedlings spaced at 50 cm between rows by 40 cm within rows. There were four rows per sub-plot and each row had nine plants (36 plants per plot). Paths measuring 0.5 m separated the sub-plots. Two outer rows formed guard rows while the two inner rows were used for data collection. A path measuring 1.0 m wide separated two main plots. Polyethylene sheet was buried at 60 cm to prevent water flow from other plots.

The plants were irrigated uniformly in the first 21 days after transplanting to ensure proper take-off of the transplants. Irrigation treatments then followed. The water application rate ranged from 1 liter for young (non-flowering) plants to 2 liters for older (flowering and fruiting) plants per day (Liu and Nyalala, 2002). The water discharge rate from each emitter was 4 l h^-1. The irrigation-time to supply two liters of water per plant for daily, two and three days schedule was 56 min, 1 h and 52 min and 2 h 48 min, respectively.

Nursery establishment and operations: The potting media was prepared by mixing topsoil with well rotten pig manure in the ratio of 2:1 (v/v). The mixture was filled to plastic pots measuring 9 cm and 5cm in diameter at the top and bottom, respectively with a hole at the bottom. Two seeds of tomato, 'Money Maker' were directly sown in pots at about 2 cm depth within the plastic tunnel. Two seeds were sown to guard against failure of seedling emergence. The seedlings were later thinned to one at two weeks from emergence. The pots were watered thoroughly using a watering can with a fine rose. Adequate water was provided on daily basis immediately after sunset for three weeks. Thereafter, water application frequency was gradually reduced to once every two days to harden the seedlings in readiness for transplanting. Weeds were spotted and carefully pulled out of the pots.

Cultural practices: Greenhouse soil was cultivated to a fine tilth and leveled. Well-decomposed pig manure was uniformly applied on the soil surface at the rate of 10 t ha^-1 (1 kg m^-2) and worked into the soil to a depth of 15 cm to allow exploitation by shallow roots of tomato seedlings at transplanting (Oikeh and Asiegbu, 1992). Pig manure was used because it has been shown to have adequate nitrogen content (Oikeh and Asiegbu, 1992) and was more available than poultry manure. Phosphorus in the form of Diammonium phosphate (DAP 18:46:0) was evenly broadcasted on soil surface at 123 kg P ha^-1 and incorporated into the soil to a depth of 15 cm to improve phosphorus uptake (Thompson and Doerge, 1995).

Planting beds, raised 0.15 m above the ground were made in each main plot prior to transplanting. The beds were leveled using a rake. Inflatable plastic drip tapes with emitters spaced 0.33 m apart were placed at the centre of each row so that each Dripline irrigated two rows. Mulching materials were randomly applied as per the treatment combinations. The clear plastic mulch (gauge 0.014 mm) was tightly spread on the bed and each edge firmly covered with 20 cm of soil after laying out the drip lines. Openings for planting holes 10 cm in diameter spaced at 0.5 m between the rows and 0.4 m within the rows (between plants) were made on the clear plastic mulch using a razor blade. The planting holes were then made using a wooden peg 10 cm in diameter. The dry grass mulch was applied at a thickness of 8 cm to cover 100% of the sub-plot immediately after transplanting.

Transplanting was done on the fourth week after sowing when the seedlings had attained 4 to 6 true leaves (Liu and Nyalala, 2002). Transplanting was carried out on 29^th/9/2001 and 25^th/4/2002 for experiments 1 and 2, respectively. Only healthy, vigorous and normal seedlings were selected and transplanted.

Poles 2.8 m were fixed 4 m apart and 2.2 m high supported the wires stretched horizontally at 0.3 m and 2.2 m above the ground. Sisal twines on which single stem of tomato plants were twisted joined the wires. Lateral shoots, senescing or diseased leaves were removed regularly by hand snapping as they appeared.

Tomato pests such as white fly, thrips, leaf miner, aphids, red spider and rousette mites found to be common in greenhouses were regularly controlled by application of broad spectrum pesticide, Brigade (Bifenthrin 25 g l^-1) at the rate of 2 ml l^-1 of water after every two weeks. Tomato early and late blights were controlled by application of Ridomil (40 g kg^-1 metaxyl-M + 640 g kg^-1 mancozeb) at 40 g l^-1 of water. Tomato plants that were severely attacked by bacterial wilt, were carefully uprooted and destroyed in both experiments.

Parameter assessments: To assess the effects of treatments on tomato crop the following parameters were determined:

Yield and its components: The number of trusses, flowers and fruits of plants that were selected for harvesting in each sub-plot were counted. The number of flowers and fruits of the first two trusses were counted. The percentage fruit set per truss of the first two trusses were calculated by dividing the number of fruits by that of flowers.

Tomato fruits with any externally detectable colour change (breaker stage) were considered mature and harvested. The tomato fruits were sorted into marketable and unmarketable (diseased, malformed, damaged by insect pests) categories. In experiment 1, harvesting was carried out from 12^th December 2001 to 3^rd February 2002. In experiment 2, harvesting commenced on 6^th April 2002 and ended on 10^th July 2002. Harvesting was carried out seven and eight times at weekly intervals in experiments 1 and 2, respectively. In experiment 1, harvesting was terminated earlier due to severe attack by tomato late blight disease.

Dry matter partitioning: Three randomly selected plants per sub-plot were assessed for dry matter partitioning at 68 days after transplanting when most trusses had formed fruits. The plant parts were separated into stem, leaves and fruits and weighed. The stem and leaf samples were put in brown paper bags and oven dried for four days at 70°C for four days. The fruit samples were cut into small pieces, spread out in single layers and dried at 70 °C for one week (Nyabundi and Hsiao, 1989). Oven dried samples were then weighed. Harvest index was calculated by dividing the fruit dry weight by total shoot (leaves, stems fruits and pruned plant parts) dry weight.

Fruit firmness: Ten tomato fruits per plant from four selected plants in each sub-plot were randomly sampled at the turning stage. The fruits were washed with calcium hypochlorite 20% (v/v) to eliminate surface pathogens and rinsed with distilled water for 3 min. Excess water was blotted-off using blotting paper. The fruits were placed in ventilated cartons at room temperature and tested for firmness using a hand pressure tester (Bishops instruments) at an interval of 0, 2, 3, 4 days through out the ripening period.

Total soluble solids: Total soluble solids of the juice from 50 randomly selected cherry-red ripe tomato fruits per sub-plot were measured using a hand refractometer (0-30° brix).

Data analysis: The data obtained was subjected to analysis of variance (ANOVA) and the treatment means separated by Least Significant Difference (LSD) or Duncan's Multiple Range Test (DMRT) at (P=0.05) using MSTATC computer software package (Michigan State University, 1993).

RESULTS

Significantly (P≤0.05) highest and lowest fresh fruit weight was attained with irrigation after every two days and no mulch and irrigation once every three days and clear plastic mulch, respectively (Table 1a).

Mulching material significantly (P≤0.05) influenced fruit dry weight. Dry grass mulch resulted in significant (P≤0.05) lowest fruit dry weight compared to bare soil and clear plastic mulch (Table 1b). Irrigation after every two days and grass mulch resulted into significantly (P≤0.05) lowest dry fruit weight while irrigation after every two days and no mulch produced the highest dry fruit weight (Table 1b).

Significantly (P≤0.05) highest harvest index was obtained from no mulch (Table 2). Irrigation schedule and mulching material significantly (P≤0.05) interacted to influence harvest index (Table 2). Significantly (P≤0.05) highest harvest index was obtained from daily irrigation and no mulch (Table 2). Clear plastic mulch significantly (P≤0.05) produced the highest fruit number (Tables 3). Clear plastic mulch and daily irrigation significantly (P≤0.05) produced the highest fruit number (Table 3).

Dry grass mulch had significantly (P≤0.05) highest individual fruit weight (Table 4). There was no significant (P≤0.05) interaction between drip irrigation and mulching material on mean fruit weight (Table 4).

Table 1a:	The effects of drip irrigation schedule and mulching materials on the fruit fresh weight per plant

Table 1b:	The effects of drip irrigation schedule and mulching materials on tomato fruit dry weight

Table 2:	The influence of drip irrigation schedule and mulching materials on harvest indices of tomato

Table 3:	Effects of drip irrigation schedule and mulch materials on tomato fruit numbers

Table 4:	Effects of drip irrigation schedule and mulching materials on marketable individual tomato fruit weight (g fruit-1)

Table 5:	Effects of drip irrigation schedule and mulch materials on marketable tomato fruit yields (t ha-1)
The means followed by same letters are not significantly different by row and columns for interactions (n = 9), by column for mulching effects (n = 3) and by row for irrigation effects (n = 3), Duncan Multiple Range Effects (p<0.05)

Table 6:	Effects of drip irrigation schedule and mulch materials on tomato total soluble solids (°brix)
The means followed by same letters are not significantly different by row and columns for interactions (n = 9), by column for mulching effects (n = 3) and by row for irrigation effects (n = 3), Duncan Multiple Range Effects (p<0.05). L.S.D. value for interaction effect is 0.18

Table 7:	Effects of drip irrigation schedule and mulch materials on tomato fruit firmness (N) at day 2 after harvesting
Mulch and irrigation effects were separated by columns and rows, respectively n=9. Interaction effects were separated by columns and rows using, n = 3. Means followed by the same letter are not significantly different according to Duncan's Multiple Range Test at P≤0.05)

Fig. 1:	The influence of days from harvest on the tomato firmness

Daily irrigation and clear plastic mulch significantly (P≤0.05) produced the highest yield (Table 5).

Total soluble solids were highest in tomatoes mulched with clear plastic compared to dry grass mulch in both experiments (Table 6). There was no significant (P≤0.05) interaction between drip irrigation schedule and mulch type on total soluble solids in the two experiments (Table 6).

Daily irrigation and dry grass mulch had significantly (P≤0.05) highest fruit firmness at day two after harvesting (Table 7). The fruit firmness was significantly (P≤0.05) lowest with dry grass mulch with day 2 after harvesting (Table 7). Fruit firmness was generally highest and lowest at 0 and 4 days, respectively after harvesting (Fig. 1).

DISCUSSION

Effects of drip irrigation and mulching material on tomato yield and yield components: Higher tomato fruit yields under grass or clear plastic mulches may be partly due to low weed population, which resulted in reduced competition for nutrients and water (Shrivastava et al., 1994). The higher flower number with daily irrigation is due adequate moisture available to the flowers, which reduced flower abortion. This is in line with findings of Ramalan and Nwokeocha, (2000).

The high fruit number under clear plastic mulch led to increased fruit competition for assimilates and probably moisture, which reduced fruit expansion hence, low individual fruit weight. This shows that clear plastic mulch only increases tomato yield by increasing tomato fruit number but not individual tomato fruit weight. Clear plastic mulch probably increased salinity build up around the plant rhizosphere due to maintained higher soil temperature (Gupta and Achrya, 1993), which increased transpiration rate compared to dry grass mulch. The high soil electrical conductivity might have reduced the rate of fruit growth by reducing water flow into the fruit and consequent reduced fruit expansion under clear plastic mulch (Mitchell et al., 1991; Dorais et al., 2001). However, Sweeney et al. (1987) reported higher tomato yield with black plastic mulch under field conditions. The variation of the findings can be associated with plant protection system used, the colour of polyethylene mulch and soil type. The high relative humidity within the greenhouse reduces consumptive water use as evaporation demand is decreased (Sweeney et al., 1987). Humidity within the greenhouse might have confounded the effect of mulching on tomato yield in the greenhouse (Perry and Sanders, 1986).

Non-significant influence by irrigation schedules on tomato yield parameters found in the experiment is in agreement with findings of Meek et al. (1983) and Kadam and Magar (1993). This suggests that drip irrigation once after every three days is not detrimental to tomatoes grown in soils with good water retention under greenhouse conditions. These findings are at variance with that of Dalvi et al. (1995) who reported that drip irrigation after every two days significantly increased tomato yield. In addition, Russo (1983) and Sanders et al. (1989) found that daily drip irrigation schedule increases tomato yield compared to longer drip irrigation schedule. In other experiments, Pasternak and De-Malach (1995) reported that drip irrigation after every two or three day significantly lowered yields of tomatoes in comparison to daily irrigation. These conflicting results may be due to variations in soil type and climatic factors (Shrivastava et al., 1994), tomato growth rate (Sanders, 2000), water quality and plant protection system (Perry and Sanders, 1986). Thus, the influence of drip irrigation schedules should be reported under prescribed set of tomato production conditions.

Effects of drip irrigation schedule and mulching materials on quality of tomato fruit: Lowest soluble solids obtained with dry grass mulch can be attributed to low soil water tension maintained by dry grass mulch, which led to higher water uptake and hence dilution of the concentration of the total soluble solids (Muller, 1993; Tan 1995). Delayed first 50 % flowering by dry grass mulch made fruiting to coincide with increased competition between vegetative and fruit growth for photosynthates. This might have lowered the amount of photosynthates partitioned to the fruit hence reduced concentration of total soluble solids in the ripe fruits (Mitchell et al., 1991).

The highest total soluble solids obtained from clear plastic can be associated with increased photosynthesis due to the plastic reflective properties, which increased light reception onto the leaves and fruits of tomatoes (Mathieu and Aure, 2000). The reflection of more light onto the tomato shoot by clear plastic is known to increase transpiration rate, amount of photosynthesis available to fruits and sugar: acid ratio (Dorais et al., 2001) and hence the higher brix value of tomato fruits. Also the high rate of transpiration and high temperature below plastic mulch may cause localized soil salinity build-up (Dorais et al., 2001). This stimulated starch accumulation in the mature tomato fruits leading to sucrose unloading into the developing fruits, which were hydrolyzed on maturation into soluble hexoses thereby increasing total soluble solids contents (Dorais et al., 2001).

Plastic mulches concentrate carbon dioxide around the plant canopy as the planting holes acts as vents for carbon dioxide escaping from beneath the mulch (Sanders, 2000). This relatively elevated carbon dioxide concentration might have accounted for the increased total soluble solids.

Higher fruit firmness realised with clear plastic mulch may be due to thicker flesh of the fruits as reflected by high dry matter content. Falluji et al. (1982) reported that fruit firmness is strongly influenced by tomato fruit flesh tissues. Also, it can be suggested that the high firmness of tomato fruit immediately after harvesting is due to skin strength, which weakens upon ripening and senescence.

In conclusion, irrigation of up to three days interval does not affect yield and quality of greenhouse fresh market tomato if the water application rate is uniform. Clear plastic mulch enhances tomato fruit total soluble solids while dry grass mulch reduces it. Irrigation schedule interacts with mulch type to influence yield and quality of greenhouse tomato.

ACKNOWLEDGMENTS

The authors thank The Department of Horticulture, Egerton University, Kenya for supporting the experiment.