Abstract: Nine French bean commercial varieties and over ninety breeding lines were assessed to ascertain their susceptibility to thrips (Megalurothrips sjostedti Trybom and Frankliniella occidentalis (Pergande). Samples of thrips infestation were collected after anthesis at weekly intervals for four weeks. Evidence from the study showed significant differences among the French bean cultivars and breeding lines in thrips infestation. The cultivars had higher thrips infestations compared to the breeding lines. Pod damage score also differed significantly among the varieties with cultivar Harricot 05 recording the highest thrips damage score (2.84).
INTRODUCTION
French bean (Phaseolus vulgaris L.) is the second major export crop in Kenya and accounts for up to 60% by value of all export vegetables and about 21% by value of the horticultural exports (Anonymous, 2001). It is grown as a monocrop mostly by small scale farmers on farm sizes of between 0.25 and 1 ha. (Nderitu et al., 1997). The year round production of French beans is hindered by nsect pests and diseases which are a major constraint. Thrips are second in importance among the French bean insect pests (Lohr and Michalik, 1995). Their feeding on the flowering parts causes abscission of flower buds, opens flower peduncles and causes curling of pods, leading to quality losses. The silvery lesions made on the pods render the product unfit for export market (Kibata and Anyango, 1996). Losses of 40-60% of pods in pre orting at the farm level and another 20% at collection points have been reported (Lohr, 1996). Various thrips control methods that include cultural, biological, chemical and host plant resistance have been evaluated and are in use in various parts of the world (Lewis, 1997). Host plant resistance against thrips pests of legume crops has been reported in several countries (Owusu et al., 1998; Salifu, 1992). Some cultivars of cowpea have been reported to register resistance mechanisms against thrips (Owusu et al., 1998). In Kenya, study by Anyango (1990) reported significant differences in the level of thrips infestation amongst various dry common bean varieties. This study was carried out to evaluate the commonly grown French bean cultivars and breeders lines for their resistance and tolerance to thrips infestation.
MATERIALS AND METHODS
The study was carried out from October 2001 at Kabete Campus Field Station farm, University of Nairobi. French bean cultivars (certified seeds) were obtained from the local market while breeder’s lines were sought from National Horticultural Research Centre, Thika and Moi University. All seeds were treated with Imidacloprid (Gaucho 350 FS) before planting to prevent bean stem maggot (Ophyiomyia sp.) and other soil borne pests’ infestation. Land was ploughed and harrowed to fine tilth using a disk harrow tractor. Furrows were made manually for seed sowing. At planting, diammonium phosphate (DAP) was applied at the rate of 494 kg ha-1 but was mixed well with the soil before placing any seed. Calcium Ammonium Nitrate (CAN) was applied at the same rate as DAP at second and fourth week after crop emergence. Sprinkler irrigation and weeding were done to provide enough water to the plant and keep the field clean of weeds, respectively.
Comparison of Cultivars and Breeding Lines
Nine cultivars (Amy, Samantha, Lexus, Impala, Paulista, J12, Jasmin, Julia
and Monel) and eleven breeding lines (Haricot 04, J4, Wacron, Piant, A21, Wacret,
Mae, Haricot 03, R0855, Haricot 05 and A17) were planted in October 2001 to
screen them for resistance to thrips under field conditions. The experiment
was laid in a Randomized Complete Block Design (RCBD) with three replicates.
Each cultivar and breeding line was sown in furrows of 3 m long within the blocks
to form a plot. Paths of 1 m between the plots and 3 m between the blocks were
maintained. Intra-row spacing was maintained at 15 cm. Sampling for thrips infestation
was done weekly after anthesis and continued for four weeks. Five open flowers
and five leaves were randomly picked from each of the cultivar and breeding
line. The samples were put in 70% alcohol in separate bottles in the field and
taken to the laboratory for thrips counting. The counting of both the adult
thrips and larvae was done under a dissecting microscope using a tally counter.
Each of the species was recorded separately for each of the varieties. Pod damage
by thrips was also assessed from samples of ten pods per plot harvested weekly
after their formation. The damage was rated on a scale of 1-5 where 1 = no damage,
2 = 1-25%, 3 = 26-50%, 4 = 51-75% and 5= over 76% damage.
Evaluation of Breeding Lines
Another Ninety three breeding lines were planted in March 2002 to study
their resistance and tolerance to thrips infestation under green house conditions.
Completely Randomized Design (CRD) was used to lay out the experiment. Three
seeds were sown in one pot for each line to represent a plot and were replicated
three times. Sampling for thrips infestation was done one week after the onset
of flowers and continued weekly for four weeks. Three open flowers were randomly
picked from each of the plot. The samples were put in 70% alcohol in separate
bottles and taken to the laboratory for thrips counting. The number of each
of the species and larvae was recorded separately.
Evaluation of Cultivars
Nine cultivars; Samantha, Julia, J12, Paulista, Impala, Jasmin, Monel, Amy
and Lexus were planted in the field and evaluated for thrips infestation in
January 2002. Planting was done twice with second planting done two weeks after
the first planting. The experiment was laid in a Randomized Complete Block Design
(RCBD) with four replicates. Plots of 2 by 2 m were established with 1 m paths
between plots and 2 m between blocks. Seeds were sown in furrows at a spacing
of 30 cm inter-row and 15 cm intra-row. Sampling of thrips was done as earliear.
All the data collected from the three experiments was subjected to analysis of variance (ANOVA) using GENSTAT Statistical package version 6.1. Means were compared and F-test was used to test significance while SED was used to separate means. Data was transformed where possible by use of square root method.
RESULTS
Comparison of Cultivars and Breeding Lines
There was significant difference among the varieties in the total thrips
count and separately for F. occidentalis and M. sjostedti (Table
1). Monel (cultivar) had the highest mean number of thrips but the pod damage
was just slightly above the average. Piant (breeding line) recorded the least
thrips population as well as pod damage. Cultivar J12 was the most tolerant
as it recorded high thrips number but low pod damage. Haricot 05 (breeding line)
recorded few thrips number but the damage was very high.
Thrips numbers varied significantly (p<0.05) over the sampling period. The general trend in all varieties and breeders’ lines showed increased numbers of thrips from first to the last sampling period across all the varieties (Fig. 1). The figure also shows that the density of F. occidentalis was higher than M. sjostedti during the sampling period.
Evaluation of Breeding Lines
There was significant difference (p<0.05) among the breeding lines in
the overall number of thrips counted and also separately for F. occidentalis
and larvae (Table 2). However many breeding lines were
not infested by M. sjostedti. Line 804.2 had the highest mean total thrips
number whereas lines 1169.1, 1182.2, 756.1, 486.1, 485.2, 769.2 and 825.2 had
the lowest mean thrips number. Lines 913.2, 804.2 and 950.1 had the highest
infestation levels of M. sjostedti, F. occidentalis and larvae,
respectively. Several lines had similar lower mean number for both thrips species
and larvae.
Evaluation of Cultivars
During the first planting, there were significant differences (p<0.05)
among the cultivars in the mean number of thrips counted and also separately
for M. sjostedti and F. occidentalis (Table 3).
Monel variety recorded the highest thrips numbers whereas Impala recorded the
least. During the second planting, there was no significant difference
in mean numbers of F. occidentalis though there was for M. sjostedti
and total thrips numbers. Monel recorded overall the highest thrips numbers
whereas Impala recorded the least. Pod damage score differed significantly (p<0.05)
among the varieties during the first planting. Julia and Lexus recorded the
highest score (2.46) whereas Impala recorded the least score (2.15). Pod damage
and number of thrips was positively correlated (r = 0.57).
Lower thrips density was recorded in the second planting (Table 4). The highest thrips count recorded was 28.19 compared to first planting of 80.40. Still monel and impala had the highest and lowest thrips count, respectively, just as in the first planting. However, pod damage score was not significant unlike the first planting.
During the first planting, there was significant difference (p<0.05) among the varieties in the total number of thrips and larvae counted from the leaves (Table 5). There was however no significant difference for either F. occidentalis or M. sjostedti on their own. Monel recorded the highest mean total thrips numbers whereas Impala had the least.
During the second planting, there was no significant difference among the cultivars in the overall thrips numbers and also for the different species separately (Table 6). However, Lexus had the highest thrips count while Julia had the lowest.
Generally, the population trend of mean thrips numbers increased significantly from 1st to 3rd sampling and then dropped during the 4th sampling (Fig. 2). The trend was the same for F. occidentalis and M. sjostedti separately during the first and second planting.
| Table 1: | Mean numbers of thrips infesting flowers and pod damage score among the cultivars and breeding lines of French beans at Kabete, Kenya in 2001 |
| v-cultivar; The unlabelled are all breeders line | |
| Table 2: | Mean number of thrips infesting flowers of French bean breeding lines at Kabete, Kenya in 2002 |
| Data transformed by (x+0.5)0.5 | |
| Table 3: | Mean numbers of thrips infesting flowers and pod damage score among French bean varieties during first planting at Kabete, Kenya in 2002 |
| *Significance at 95%; *** Significance at 99% level | |
| Table 4: | Mean numbers of thrips infesting flowers and pod damage among French bean cultivars during second planting at Kabete, Kenya in 2002 |
| -Not significant; ***Significance at 99% level | |
| Table 5: | Mean numbers of thrips infesting five French bean leaves during the first planting at Kabete, Kenya in 2002 |
| -Not significant; *Significance at 95%; ***Significance at 99% level | |
| Table 6: | Mean numbers of thrips infesting French bean leaves during the second planting at Kabete, Kenya in 2002 |
| -Not significant; **Significant at 95% level | |
| Fig. 1: | Mean thrips numbers on flowers of French bean cultivars and breeding lines over the sampling period at Kabete, Kenya in 2001 |
| Fig. 2: | Mean number of thrips on flowers of French bean cultivars over the sampling period at Kabete, Kenya |
DISCUSSION
Significant differences in thrips infestation were observed in both French bean cultivars and breeding lines. Among the breeding lines, 1169.1, 1182.2, 756.1, 486.1, 769.2 and 825.2 were found to have lower thrips infestations and could be considered for use in breeding programes to develop thrips resistant cultivars. However, they need to be screened further for their sensitivity to pod damage. Piant was the most resistant breeding line when compared to the cultivars as it had both low infestation and lower pod damage. It can be considered cultivar development. Among the cultivars, Impala had the lowest thrips infestation level. Awareness on Impala as a less susceptible variety to thrips should be created especially to the exporters who in most cases determine the varieties grown by the farmers. Cultivar J12 can also be encouraged to plant as it was the most tolerant cultivar, with high count of thrips but low pos damage score. The differences in thrips resistance observed among the cultivars and breeding lines is in agreement with earlier work done on cowpeas and dry common bean varieties that were found to differ in thrips infestation (Owusu et al., 1998; Anyango, 1990). The study also established that flowers had higher adult thrips numbers than the leaves in all cases. This is similar to findings by Cho et al. (2001) who found that adult thrips were more dorminant on flowers than on leaves in cucumber. The larvae were more than adults on leaves. Frankliniella occidentalis was more abundant than Megalurothrips sjostedti among the commercial varieties. This was attributed to the fact that sampling coincided with the warm periods of the crop season, a time when this species thrives best. Farmers can use this information in planning their production to coincide with the cooler months of the year in order to reduce infestation by F. occidentalis, a difficult species to control. If production must be done during the hot periods, then profitability of the enterprise must be critically assessed. Use of resistant or torelant cultivars is advantageous given the stringent quality requirements in the export market. Furthermore, a resistant/tolerant cultivar poses no technical difficulties to the farmer and hence they are user friendly. Such varieties should however be selected guided by market and consumer preference.
ACKNOWLEDGMENTS
The authors wish to thank Rockfeller Foundation through FORUM for supporting this study financially.