Assessment of Relationships Between Genotypic Variation and Growth and Yield of Spider Plant in Kenya

Spider plant (Cleome gynandra L.) is an important African leafy vegetable (ALV) that has been used by local African communities as a source of nutrition in their diets for many years. The plant has recently attracted an increasing demand for its highly nutritive and health promoting bioactive compounds important in combating malnutrition and reducing human degenerative diseases. Despite the great value of spider plant, its supply and cultivation remain low, a factor attributed to unavailability of superior genotypes. This study carried out at Ruiru sub county, Kiambu county of Kenya sought to establish the influence of genotypic variation on growth and yield of spider plant. Experimental plots were set up in the field in Ruiru and greenhouse in Juja. Analysis of variance (ANOVA) was used to assess the significance of variables. Results indicated that genotypes MLSF17, UGSF14, P6, UGSF9 and UGSF36 yielded outstanding agronomic performance. However, there was no significant difference among growth parameters of genotypes in greenhouse compared to the field experiments.

receives average annual rainfall of 1,025 mm. Temperature range is 10-26°C with altitude of 1,795 m above sea level. The soils are typically red on undulating topography. Main human activities include coffee farming, dairy, and horticulture (MoA, 2008). The experimental factors tested consisted of three nitrogen levels. The nitrogen levels were manure, 2.6 g N/plant and 5.2 g N/plant. 1 bucket, each weighing 10kg of fine and well decomposed cattle manure were put in each sub-plot measuring 1.2 m by 3 m. The experiment was laid out as a complete randomized design (CRD) with three replications. Analyses of variance (ANOVA) were done using SAS (SAS 9.1.3) for dry weight, leaf area, height and number of leaves. The level of significance was at p<5% and mean separation was done using LSD.

Influence of spider plant genotype on the number of leaves across different harvesting periods in Ruiru season one
There was no significant difference among genotypes in Ruiru trials. Genotypes significantly influenced (P≤ 0.05) the number of leaves across different harvesting period's long rain season (table 3.10). The control variety (P6) resulted into more leaves per plant when compared with UGSF9 and UGSF25 in the first harvesting period. In the seventh week of harvesting, MLSF17 significantly produced more leaves when compared to P6 (P≤ 0.05). UGSF14 significantly produced more leaves than IP3, UGSF25, UGSF12, MLSF3 and MLSF17 in the eighth week of harvesting. There was a significant difference in the number of leaves produced between the various genotypes in the ninth week of harvesting. UGSF35 produced more leaves than IP3 and UGSF9 at significant level of r (P≤ 0.05). Similarly, the amount of leaves produced by UGSF14 was significantly higher than that of UGSF9 in the long rains (Table 3.10). In the second season, results indicated that genotype significantly influenced (P≤ 0.05) the number of leaves produced (table 3.11). MLSF3 produced significantly more leaves in the second season than IP3 at the sixth week of harvesting (table 3.11). There were no significant differences in the number of leaves produced by various genotypes in the fifth, seventh, eighth and ninth weeks of harvesting in the second season.

Influence of genotype on the number of leaves across different harvesting periods in Greenhouse season one
Genotypic variation in spider plants significantly influenced (P≤ 0.05) the number of leaves across different harvesting periods in the greenhouse (Table 3.12). Results indicated that UGSF9 produced more leaves per plant than UGSF36 and MLSF17 in the fifth week of harvesting (table 3.12). There were no significant differences in the number of leaves produced per plant as a result of genotypic variation beyond five weeks. For greenhouse, above-ground diurnal temperature ranged from 15-37 o C for the first season and 11-31 o C for the second season. The greenhouse plants began to flower five weeks after planting compared to outdoor that started flowering later in week six. Similar observations were made in Ruiru first season, indicating a positive correlation between temperature and the time to flowering

Effect of genotypic variation of spider plant on yields across different harvesting periods in Ruiru season two
Genotypes of spider plant significantly influenced (P≤ 0.05) yields across different harvesting periods in Ruiru season two (Table 3.13). MLSF3 resulted in significantly higher yields than IP3 in the seventh and eighth week of harvesting. However, in the ninth week, there was no significant different in yields between MLSF3 and IP3 but MLSF3 significantly exceeded yields of UGSF36 (Table 3.13).

DISCUSSION
Plant selection for superior traits is an old practice among most researchers either through breeding or phenotypic observation. Desirable characteristics such as biomass, yield, resistance to pests and diseases are considered. The level of management is critical factor influencing spider plant fresh leaf yields. Besides the genetic influence, growing conditions and management practices undertaken during growth have important bearing on crop nutritional status (Hutchinson et al., 2006). Application of manure and/or fertilizer and the stage of maturity of spider plant are critical in determining the phytochemical, nutritional and sensory characteristics of the vegetable (Kebwaro, 2013). The improvement in yields for the second season in the field was also attributed to better management coupled with lessons learnt from the previous season. The trial was set under similar conditions to minimize the effect of environmental variations such as storms and pests. Greenhouse above-ground diurnal temperature ranged from 15-37 o C for the first season and 11-31 o C for the second season. The plants in greenhouse began to flower five weeks after planting compared to outdoor that started flowering later in week six. Similar observations were made in Ruiru season I. This may be attributed to high temperature stress that induce early flowering. Yields are also being improved through selection of genotypes of spider plant, which has intensified in the recent past (Onim and Mwaniki, 2008;Masinde, 2011), since commercial varieties have shortfalls such as yield, nutrient, and geographical diversity. Limited access to quality seed and shortage of suitable cultivars has been key cause of low spider plant productivity (Abukutsa-Onyango, 2010b) In terms of seed weight, heavier Rumex acetosella seeds have higher relative growth rate (RGR) in the first 7 weeks after germination in the field experiment than greenhouse. Thereafter, lighter seeds had twice RGR greater than those from heavier seeds. Nonetheless, after 10 weeks, there was no significant difference in RGR. This suggests that a trade-off between allocation to sexual and vegetative reproduction occurs over successional time (Houssard and Escarré, 1991). Studies have shown that seed weight factor has greatest importance during the early stages of plant growth. According to Ocxcmann (1942), tomatoes, cucumbers and soybean plants grown from lighter seeds had slower initial growth rate, which persisted until the end of the sixth week. Therefore, there is a positive correlation for this period of growth. Since spider plant takes relatively short growth period of about twelve weeks, the light seeded genotypes are less likely to catch pace with the heavier seeded accessions after the period. This is consistent with the findings of this study where lighter seeded genotypes such as IP3 and MLSF3 that generally had poor agronomic performance and vice versa. The study also documented higher mortality rate among seedlings grown from seedlings of lighter weight than among seedlings grown from heavier seeds. This difference in mortality rate is probably due to differences in plant vigour.

CONCLUSION AND RECOMMENDATIONS
Appearance, colour, aroma and/or taste and medicinal properties influence acceptability of vegetable lines. Two methods were used to measure consumer choices: the vegetables were cooked and a panel invited to taste and fill in questionnaire; and market survey was done to determine purchase preference. Food colour is an indicator of available phytochemicals present e.g. carotenoid, lycopene and anthocyanin. The main challenge with this study was its wide scope. For instance, the highest yielding genotypes were not necessarily the ones most preferred by neither the growers nor the consumers, nor will it be nutritionally endowed variety. Consequently, informed compromise must therefore be reached in order to make invaluable and sound recommendations. The wider crop