1. Introduction
French bean (Phaseolus vulgaris L.) is among the most important horticultural crops grown and consumed worldwide. It is grown for its tender pods and shelled green or dry beans. The crop is a rich source of important nutritional elements such as flavonoids, vitamin A, dietary fibres, potassium, folate, iron, magnesium, thiamine, riboflavin, copper, calcium, phosphorus, Omega-3 fatty acids and niacin [1] . French bean production in Kenya is mainly undertaken by smallholder farmers who constitute approximately 80% of all growers. Production of this crop is however often undermined by major pests like spider mites, bean fly, white flies and aphids [2] .
The two-spotted spider mite, Tetranychus urticae Koch (Acari; Tetranychidae), is one of the most important pests that attack French beans and other crops worldwide [3] . The pest feeds on the plants through piercing and sucking cell contents. The resulting symptoms include tiny yellow or white speckles and bronzing of leaves [4] [5] . Severe mite infestations cause premature defoliation leading to reduced sugar content and drastic reduction in the crop yield [6] [7] and quality due to increased risk of pod damage from ultraviolet rays. The rapid developmental rate, short generation time and high net reproductive rate of T. urticae allow them to achieve damaging population levels very quickly when growth conditions are good, resulting in an equally rapid decline of host plant quality [8] .
Most of the smallholder farmers heavily rely on synthetic pesticides to manage the two spotted spider mite pest on French beans. Chemical pesticides used for spider mite suppression are usually weak acaricides and often do not perform well. Some of the active ingredients that have been used by smallholder farmers with little success include abamectin, spiromesifen, dicofol and chlorfenapyr [9] [10] . Synthetic acaricides have also caused serious problems such as pesticide resistance, environmental contamination, unacceptable pesticide residues in food and lethal effects on non-target organisms [11] [12] . Results from a study by [13] demonstrated variable levels of abamectin resistance in T. urticae populations. These negative effects have resulted in the increasing interest for natural plant-based pesticides which might be safer, biodegradable and have shown low pest resistance [14] .
Several studies have evaluated the potential of natural plant extracts to protect crops from insect and mite pest species such as whiteflies and spider mites [15] . Plant species have been found to contain natural deterrents which are toxic to various insect and mite pests [16] but safe to mammals. The extracts of Satoreja hortensis L. (Lamiaceae) were found to be toxic to TSSM [17] . Similarly the extracts from neem (Meliaceae), some species of solanaceae, Capparis aegyptia (Capparaceae), Nerium orleander L. (Apocynaceae) and Alianthus altissima L. (Simaroubaceae) have also been found to be effective against TSSM [18] [19] . Antifungal, antibacterial and antioxidant activities of L. nepetifolia and O. gratissimum have also been confirmed [20] [21] .
Further studies by [22] have demonstrated insecticidal, antifungal and antibacterial activity of L. nepetifolia and O. gratissimum extracts. Phytochemical examination of these plants indicated the presence of different diterpenoids and other bioactive compounds [23] [24] . Water extracts of O. gratissimum have been reported to inhibit egg hatch and resulted in juvenile mortality of root-knot nematode (Meloidogyne incognita) while improving grain yields of cowpea [25] . [26] reported highest mortality and repellence of Callosobruchus maculatus treated with ethanolic extracts from O. gratissimum leaves under laboratory conditions. They attributed the insecticidal activity to the presence of secondary metabolites such as saponins, flavonoids, alkaloids, phenolics and terpenes. Similar results with ethanolic extracts of O. gratissimum leaves against bean weevil have been documented [27] [28] .
This study therefore sought to determine the miticidal activity of L. nepetifolia and O. gratissimum plants extracts against T. urticae and their subsequent influence on the quality and yield of French bean under field conditions.
2. Materials and Methods
2.1. Study Site
Field experiments to evaluate the miticidal activity of L. nepetifolia and O. gratissimum methanolic crude plant extracts on the two spotted spider mite (T. urticae) were conducted at Horticulture Research and Teaching Farm of Egerton University situated in Nakuru county Kenya during the 2014 and 2015 growing seasons. The farm lies at a latitude of 0˚30'S, longitude 36˚30'E and an altitude of 2238m.The experimental site receives an annual rainfall of 1013 mm and the dominant soil type is mollic andosols.
2.2. Preparation of t Plant Extracts
Composite fresh leaves and tender stems of L. nepetifolia L. O. gratissimum were collected from fallow fields at Egerton University and the surrounding. The plant materials were dried in well ventilated room at 18˚C - 28˚C for two weeks. The dried leaves were ground into fine powder using an electric laboratory hammer mill, and subjected to methanol (100% AR) extraction at a rate of 200 gL−1. The extracts were kept in air tight containers refrigerated at 4˚C for use in the bioassays.
2.3. Mass Rearing of Two-Spotted Spider Mites
The two spotted mites were obtained from infested leaves of French bean plants which had not been sprayed with any acaricide. Rearing was done on 2 - 3 week old beans which were maintained in the greenhouse (25˚C ± 1˚C) and RH 65 ± 5 for use in the bioassays
2.4. Field Bioassay
The plant extracts were evaluated at different concentrations (1.5%, 3%, 6%, 12% w/v) in a Randomised Complete Block Design (RCBD) replicated three times. A synthetic acaricide (Abamectin 0.6 ml/L DW) was used as positive control and water as negative control. Experimental plots measuring 2 m × 2 m with a spacing of 0.5 m between and 0.2 within the rows were planted with certified French bean seeds (variety Teresa) according to recommendations for commercial French bean production. To avoid natural infestation, all plants in each plot were covered with nylon mesh size 0.4 by 0.5 mm and thread thickness of 0.1 mm at the primary leaf stage.
Twenty adult spider mites from greenhouse cultures were randomly introduced onto each bean plant at 21 days after planting using a fine hair brush. The plant extracts were applied as spray solutions using a hand held sprayer 14 days after mite infestation. A repeat treatment application was done after another 14 days. Six plants from the two middle rows in each plot were randomly selected and tagged for data collection.
2.5. TSSM Population Reduction
The population of the mites was assessed three days before treatment application by counting the number of adult mites from the underside of leaves from the six tagged plants in each plot. A second mite population count was done at 72 h after the second treatment application. The corrected percent efficacy of the plant extracts was then calculated according to Sun-shepard formula [29]
(1)
2.6. Percent Leaf Reduction
This was done by counting the number of leaves on the six tagged plants in each plot before treatment and after the second treatment application. The change in number of leaves constituted the damage by TSSM which was calculated as follows
(2)
2.7. Number of Pods, Pod Length, Diameter and Yield
This was done by hand plucking all immature green pods from the two middle rows. The number of pods was counted before measuring their pod length and diameter using a ruler and a veneer calliper respectively. The pods were then weighed when still fresh using an electric weighing balance.
2.8. Data Analysis
All data collected was subjected to analysis of variance (ANOVA) at P ≤ 0.05 using SAS statistical package [30] .
3. Results
3.1. Two Spotted Spider Mite Population Reduction
Results of the field bioassays indicated a dose dependent percent mite population reduction expressed as corrected percent efficacy during the two seasons (Table 1). The highest efficacy of 82.75% for L. nepetifolia and 69.06% for O. gratissimum plant extracts occurred at 12% w/v concentration in 2014 and 78.0% for L. nepetifolia and 77.56% for O. gratissimum in 2015. The synthetic acaricide (Abamectin 0.6 ml/L) showed a corrected percent efficacy of 65.76% and 69.05% in 2014 and 2015 respectively. Abamectin however showed higher percentage efficacy when compared with the lower plant extract concentrations of 0.0%, 1.5%, 3.0% and 6.0% w/v during both growing seasons.
3.2. Percent Leaf Reduction
The plant extracts concentrations 12% w/v, 6% w/v and 3% w/v showed significantly lower percent leaf reduction 1.71, 4.88% and 5.92% respectively for L. nepetifolia and 0.39%, 3.5% and 19.86% for O. gratissimum compared to abamectin (20.46%) in 2014 (Table 2). The trend was the same in 2015 with L. nepetifolia showing 0.58%, 5.09% and 21.72% leaf reduction at concertation levels of 12% w/v, 6% w/v and 3% w/v respectively and O. gratissimum showed 12.49%, 13.86% and 33.58%. Abamectin on the other hand showed a percent leaf reduction of 23.94% in 2015.
3.3. Number of Pods, Pod Diameter, Pod Length and Pod Yield
A higher number of pods was recorded in the plots that were treated with plant extracts compared to the synthetic acaricide (abamectin). L. nepetifolia plant extracted recorded 61.00, 48.33 pods at concentrations levels of 12% and 6% w/v followed by O. gratissimum plant extract which recorded 48.67 and 35.00 pods while abamectin recorded 28.33 pods during 2014 growing season (Table 3). The trend was similar in 2015 where L. nepetifolia plant extract recorded 41.00
Table 1. Mite population reduction (corrected % efficacy) at different plant extracts concentrations in season 1 and 2.
*Means in a column whose standard error values do not overlap are significantly different at P ≤ 0.05.
Table 2. Percent French bean leaf reduction at different plant extracts concentrations in season 1 and season 2.
*Means in a column whose standard error values do not overlap are significantly different at P ≤ 0.05.
Table 3. Effects of L. nepetifolia (LN) and O. gratissimum (OG) plant extracts on number of pods, pod diameter (cm) pod length (cm) and pod yield (kg).
*Means in a column whose standard error values do not overlap are significantly different at P ≤ 0.05.
and 27.67 pods at concentrations levels of 12% and 6% w/v followed by O. gratissimum which recorded 40.67 and 31.00 pods while abamectin recorded 18.00 pods.
Plant extracts applied at 12% concentration level also indicated the widest pod diameter of 0.80 cm and 0.77 cm for L. nepetifolia and O. gratissimum respectively in 2014. This was however not significantly different from abamectin which recorded a pod diameter of 0.70 cm. No significant differences in pod diameter was observed between the plant extracts and abamectin in 2015.
Although L. nepetifolia plant extracts caused a significant increase in in pod length compared to abamectin and the untreated control, there were no significant differences in pod length across the plant extracts applied either at low or a high concentrations in 2014 and 2015. O. gratissimum however recorded significantly longer pods at plant extract concentration levels of 6% and 12% w/v in 2014.
Results also showed significant differences in pod yields at different plant extract concentrations. The highest pod yield (0.88) was recorded at 12% w/v for L. nepetifolia and 0.90 for O. gratissimum in 2014 and 0.76 L. nepetifolia and 0.86 for O. gratissimum compared to 0.36 kg and 0.25 for abamectin in 2014 and 2015 respectively.
4. Discussion
The plant extracts from L. nepetifolia and O. gratissimum demonstrated biological potency against the two spotted spider mites. Leaves, stems and roots extract of O. gratissimum have been found to contain potent bioactive components (essential oils) made up of eugenol and other compounds such as diterpenes, coumarins, iridoids, saponins, condensed tannins, flavonoids, alkaloids and steroids that have antioxidant and insecticidal properties [31] [32] . Dried O. gratissimum leaves have also been reported to have insecticidal activity against cowpea bruchids (Callosobruchus maculatus F.) [33] . Phytochemical studies by [34] also identified and quantified fixed oil components from the leaves of L. nepetifolia. There current research did not however observe significant differences in the activity of L. nepetifolia and O. gratissimum on the two spotted spider mite population and subsequent pod yield. The results therefore suggesting that there may be similar compounds with similar bioactivities in the plant extracts.
The highest efficacy of the plant extracts in reducing TSSM populations was demonstrated at the highest plant extract concentrations of 12% w/v. This is consistent with studies by [35] who reported a concentration and exposure time-dependent increase in the efficacy of L. nepetifolia and O. gratissimum plant extracts against the TSSM under laboratory conditions. Similarly, this study observed the highest bean leaf reduction, pod numbers and pod yield in plots applied with the highest plant extract concentrations of 12%. Reduced mite infestations corresponded to lower mite damage and hence reduced percentage leaf reduction. Reduced leaf loss meant that more leaves were retained by the plant for photosynthesis and hence increased pod numbers and yield. A full canopy is required for French bean crop to fully intercept sunlight to produce sugars, fill pods and maximize yield [36] [37] [38] . The plant extracts did not however strongly influence the pod diameter and length which are important French bean quality parameters. All the treated plots recorded a pod length of about 9 - 12 cm representing size code number 4 according to the Asean standard for French beans [39] .
This study has thus demonstrated the efficacy of L. nepetifolia and O. gratissimum in managing two-spotted spider mite and subsequent increase French bean yield under field conditions.
5. Conclusion
This study illustrates the potential biological potency of L. nepetifolia and O. gratissimum plant extracts against the two-spotted spider mite on French beans. Yields obtained from the fields where the plant extracts had been applied were comparable to or even better than yields from the fields where the Abamectin acaride had been applied. This research has however shown that although French bean pod yield increased with increase in plant extract concentrations, pod quality parameters (diameter and length) do not appear to significantly improve with increase in plant extract concentrations. The study also suggested that L. nepetifolia and O. gratissimum may contain similar compounds with similar bioactivities in their plant extracts.