INTRODUCTION

Tomato originated from the tropics of Central and South America, extending from Mexico, Ecuador through Chile. It was transported to Europe and improved further before reaching the United States and Asia (Kalloo, 1993). It was now the most widely grown vegetable crop in the world, giving growers income, expanding export potential and improving the supply of vitamins and minerals in human nutrition (Rawshan, 1996). Commercially, almost 70 million tones of tomato are grown in the world in more than 2 million hectares of land, but less than 20% of the yield comes from the tropics (Phene, 1989). The versatility of the tomato crop contributes greatly to its popularity as a food product; tomatoes can be processed and canned easily as a whole or as paste; Juice, sauce or powder, or eaten raw, alone or in combination with other foods.

Yield of tomatoes in the tropics is generally low when compared with the temperate regions. In 1994, the average yield was 9.9 t ha^-1 in Thailand, 15.6 in India, 25.3 in China, 8.8 in Philippines, 4.5 in Malaysia, 52.8 in Japan and 63.6 t ha^-1 in the U.S.A. (Anonymous, 1994). In Africa average yield of 8-25 t ha^-1 was recorded, with the highest yield from South Africa and the least from Benin and Nigeria (De Lannoy, 2001). In Nigeria, tomato is widely cultivated around Guinea savanna mostly in the wet season and Sudan savanna in the dry season through irrigation scheme (Adelana, 1977).

Inadequate application of improved cultural practices may be some of the factors that limit tomato production, farmers in Nigeria obtained very low yield compared with global yield. Tomato yield could be increased substantially through improved agronomic techniques like staking (a practice of supporting plant to prevent fruit clusters from touching the ground) and pruning (removal of side shoots and lower shoots). Ahmad and Singh (2005) reported a significant yield increase by staking tomato. Rafi (1996), Chen and Lal (1999) and Abdel-Al et al. (1962) also recommended pruning as a cultural practice that improves the yield and quality of tomato.

However, the benefits of staking and pruning according to Chen and Lal (1999) include; while staking improves fruit quality by keeping plants and fruits off the ground thus reduces rotting, incidence of soil borne diseases and providing a better spray coverage, pruning diverts nutrients to flower clusters and fruits on the main stem and allows more efficient air circulation. Wuster and Nganga (1970) stressed that, properly supported and pruned plants that are appropriately spaced produce larger, earlier and relatively reasonable fruit yield than non-pruned and non-stated plants of the same variety. Therefore, determining effects of staking and pruning on the performance of tomato was the objective of the study presented in this research.

MATERIALS AND METHODS

Two experiments were conducted during the 2004/05 and 2005/06 dry seasons at the Usmanu Danfodiyo University Fadama Teaching and Research Farm, Sokoto (latitude 13°9’N and longitude 5°15’E) (Kowal and Knabe, 1972). The climate of the area is semi-arid with rainfall range of 550-660 mm per annum, spread over a period of 4-5 months (May-September). A mean monthly temperature range of between 14-41°C was recorded between 2003-2006. The soil of the study area was clay loam (pH 5.7) and seasonally flooded (during rainy season). The physico-chemical analysis of the soil at the experimental site revealed that the soils were low in total N, available P and organic carbon and was slightly acidic in nature (pH = 5.61-6.35). The soil at the experimental site was loamy in 2004/05 and clay loam in 2005/06 cropping season (Table 1).

The treatments consisted of factorial combination of two training (staked and unstaked), three pruning levels (Three-stem, two-stem and unpruned) and three intra-row spacing (20, 40 and 60 cm) laid out in a split plot design replicated three times with staking allocated to the main plots and pruning and intra-row spacing allocated to the sub-plots. Results of training and pruning are presented in this research.

Table 1:	Physico-chemical properties of the soil at the experimental site in 2004/05 and 2005/06 cropping season

Certified seed of tomato cultivar (Roma VFN) was obtained from Kebbi State Agricultural Supply Company (KASCOM) Birnin Kebbi. Seedlings were raised in nursery bed using nursery management techniques (Thinning out and hardening off was carried out before transplanting). Seedlings were transplanted at about 30-35 day after sowing (i.e., 4-5 leaf stage). Stakes of about 1m length were driven at 10 cm to the side of the plants in the staked treatments. A strong but soft thread was used to tie the plants to the stake at intervals as the plant grows. Irrigation was done at an interval of between 4-7 days at field capacity. Fertilizer was split-applied at transplanting and 4 weeks after transplanting at the rate of 300 kg NPK (15:15:15) ha^-1 and 140 kg Urea ha^-1, respectively.

Pruning was done starting from 4 WAT and continued 2-weekly up to 10WAT. Depending on the pruning level, one or two shoots just below the first flower cluster was left to grow as the second and third shoots, respectively, while the rest were removed. Weeds were controlled manually by weeding three times at 4 weeks interval. The plots were sprayed against insect pests at an interval of 3 weeks using Karate^® (Lambdacyhalothrin) at the rate of 4 mL liter^-1. Fruits were harvested at regular intervals at physiological maturity (skin turned yellowish-orange).

Data was collected on mean fruit length and diameter, mean fruit weight, total fresh fruit yield and percentage marketable yield. Data collected were subjected to Analysis of Variance (ANOVA) procedure and significant differences were further analyzed using Least Significant Difference (LSD) test.

RESULTS AND DISCUSSION

Mean Fruit Length and Diameter
Training showed a significant (p<0.05) effect on mean fruit length and mean fruit diameter only in the first trial (Table 1). The highest fruit length (6.30 cm) and diameter (4.35 cm) were recorded in staked tomato plants compared with the unstaked plants with 5.81 and 4.01 cm, respectively. Ahmad and Singh (2005) and Ariyarathne (1999) reported similar result for fruit length and fruit diameter, respectively. Both authors attributed the result to higher insolution (less mutual shading) advantage exhibited by the staked plants which result to higher photosynthetic rate.

Significantly higher mean fruit length was produced by three-stem (6.25 cm) and two-stem (6.37 cm) plants compared with unpruned (5.54 cm). Similarly, in terms of fruit diameter, higher mean fruit diameter was recorded in three-stem (4.47 cm) and two-stem (4.34 cm) compared with the unpruned (3.73 cm). This agrees with the findings of Hernandez and Sanchez (1992), Zhang (1999) and Myanmar (1999). However, the higher fruit size produced by the pruned plants could be because, pruned plants had a reduced vegetative sink (shoots) compared to unpruned plants. In that case, larger portion of the photosynthate would be partitioned to the reproductive sink (fruits) in the pruned plants while in unpruned, most of the photosynthate would be used by the shoots for respiration (Brown, 1984).

Mean Fruit Weight
Mean fruit weight in both trials and the combined (Table 2) was significantly (p<0.05) higher (48.74 g) in staked plants than in unstaked (45.52 g) plants. Ariyarathne (1999) and Ahmad and Singh (2005) reported similar results. However, the higher mean fruit weight by staked plants cold be because staking facilities good insolation (minimal shading effect) of leaves and enhances proper air circulation, which ultimately leads to more photosynthetic rate (Mckeen, 1984; Konsler, 1999).

In both trials significantly higher mean fruit weight was recorded in three-stem and two-stem compared with the unpruned. In the combined, two-stem plants produced the highest (52.19 g), followed by three-stem (48.83 g) and the least was unpruned (38.86 g) plants.

Table 2:	Mean fruit length and diameter of tomato as influenced by training and pruning in 2004/05 and 2005/06 cropping seasons and the two years combined
Within a treatment group, means in a column followed by same letter(s) in superscript are not significantly different at 5% level using LSD; NS = Not Significant; s = Significant at 5% level of significance

Table 3:	Mean fruit weight and total fresh fruit yield (t ha-1) as influenced by training and pruning in 2004/05 and 2005/06 cropping seasons and the two years combined
Within a treatment group, means in a column followed by same letter(s) in superscript are not significantly different at 5% level using LSD; NS = Not Significant; S = Significant at 5% level of significance

Rafi (1996), Zhang (1999) and Myanmar (1999) independently reported that mean fruit weights produced by three-stem and two-stem pruned plants are the same, but was significantly higher than that produced by unpruned plants. The reason for higher mean fruit weight in pruned plants than unpruned could be because of former had less photosynthate-demanding shoots which results to more dry matter partitioning to its fruits. The least mean fruit weight in the trainingxpruning interaction in the first trial was obtained in unstaked and unpruned plants (Fig. 1).

Total Fresh Fruit Yield
Total fruit yield (Table 3) was significantly (p<0.05) affected by training in the first trial and in the combined. Staked plants produced higher (56.33) (55.06) fruit yield in (t ha^-1) than unstaked (50.43) (49.49) plants for the first trial and the combined respectively. This result is in line with the findings of Ahmad and Singh (2004) and Ariyarathne (1999). The high fruit yield obtained in staked plants could be reflected to the higher mean fruit weights recorded by staking in Table 3.

For pruning, two-stem and unpruned plants produce lower fresh fruit yield while three-stem plants produced the highest. The total fresh fruit yield in tow-stem plants was not significantly higher than in the unpruned, probably because the plants (two-stem) were heavily pruned such that the pruning advantages-i.e., increase fruit size and mean fruit weight (Maynard, 2000) could not outweigh the unpruned advantage-i.e., high number of fruits per plant.

Fig. 1:	Mean fruit weight of tomato as influenced by pruningxtraining interaction in the two years combined. Bars with same letter(s) are not significantly different at 5% level using DMRT

Fig. 2:	Total fruit fresh weight as influenced by training and interaction in the two years combined. Boxes with same letter(s) are not significantly different using DMRT at 5%

But the moderately pruned (three-stem) plants had higher fruit size, mean fruit weight and relatively comparable number of fruit to the unpruned, as a result, the three-stem plants out yielded both unpruned and two-stem plants significantly (Fig. 2). Rafi (1996) and Myanmar (1999) reported the same.

Percentage Marketable Yield
Significantly larger percentage of marketable fruits in the first trial (Table 4) were produced in staked (72.73%) plants compared with the unstaked (67.68%). More so, in the combined analysed result, staked treatment yield 77.17% fruits that were marketable compared with only 66.20% produced by the unstaked plants.

Fig. 3:	Percentage marketable yield of tomato as influenced by training and pruning interaction in 2004/05 cropping season Bars with same letter (s) are not significantly different at 5% level using DMRT

Table 4:	Percentage marketable yield as influenced by training and pruning in 2004/05 and 2005/06 cropping seasons and the two years combined
Within a treatment group, means in a column followed by same letter(s) in superscript are not significantly different at 5% level using LSD; NS = Not Significant; s = Significant at 5% level of significance

This result is in line with the recommendation of the advisory committee on vegetable crops of the United States (Ahmad and Singh, 2005) which stressed that one of the advantages of staking tomato plant is to increase the percentage of marketable fruits, by preventing fruits from touching the soil and thus reduces rotting and incidence of soil-borne diseases (Chen and Lal, 1999).

Pruning did not have significant effect on the percentage marketable yield in both trials and the combined. This result confirmed the report by Rafi (1996) and Rawshan (1996), but contradicts Zhang (1999) who reported that unpruned plants produced the highest percentage marketable yield than pruned plants. The interaction between training and pruning (Fig. 3) in the first trial, showed that highest percentage marketable yield was obtained in staked and pruned (both three and two-stem) plants (Fig. 3).

CONCLUSIONS

From the findings of this study particularly on yield (total fresh fruit yield) and yield components (fruit size and mean fruit weight), it could be concluded that staking, coupled with three-stem pruning increases tomato yield and quality.

ACKNOWLEDGMENTS

The authors wish to express their gratitude to the General Manager, Kebbi State Agricultural Supply Company (KASCOM) for providing the tomato seed used in the trials.