This consisted of the eight plant species Capsicum annuum L, Cleome viscosa L, Mytragina inermis (Wild) kuntze, Strophantus hispidus (A.CD), Ocimum basilicum L., Cassia nigricans (Vahl.), Cassia occidentalis L and Pseudocedrela kotchyi.

The choice of plant material was made on the basis of surveys carried out [6] (MAE, 1985) in previous years on plants with insecticidal properties and on the use of these plant substances in the control of other insects according to the literature ( Table 1 ).

One species of fruit fly was used in the study. This was Bactrocera dorsalis.

The study was conducted in the eco-toxicology laboratory of the former Plant Protection (PV) and in the biological control laboratory of the Centre National de Spécialisation en Fruits et Légume (CNS-FL) of the Institut de l’Environnement et de Recherche Agricole (INERA) Farako-Bâ in Bobo-Dioulasso. Fruit fly rearing conditions were: a temperature of 26 ̊C ± 1°C; a photoperiod of 12 h of light and 12h of darkness and a relative humidity of 70% ± 10.

The method used for extracting total extracts is successive depletion by percolation with solvents of increasing polarity according to the total extract preparation diagram (Kambou et al., 2008) ( Figure 1 ).

1) Biological activity of total extracts on eggs, pupation and emergence of Bactrocera dorsalis

Collection of Bactrocera dorsalis eggs

Bactrocera dorsalis eggs were obtained by introducing six egg-laying devices for 2 h into rearing cages containing Bactrocera dorsalis adults (200 males and 200 females) at least 15 days old. Using a graduated cylinder, 5,000 eggs were collected for the various tests at the biological control laboratory of the Institut de l'Environnement et de Recherche Agricole (INERA) in Farako-Bâ.

Bactrocera dorsalis larval nutrient medium is required to perform the biological activity test of total extracts on B. dorsalis eggs and development.

The nutrient medium for B. dorsalis larvae was prepared as follows: brewer’s yeast waste was heated to boiling, with a total heating time of 1 h 30 min, stirring frequently to prevent the yeast sticking to the bottom of the pot. After boiling, the resulting yeast was cooled completely for 24 hours. After cooling, 0.92 g of papain was added to 11.5 l of modern brewers’ yeast waste (in the ratio of 8g of papain to 100 liters of yeast waste). The mixture was thoroughly mixed and transferred to a container, then placed in an incubator set at 65° C for 24 h. A quantity of 69 g potassium sorbate was added to the mixture at a rate of 6 g/l, to preserve the product, which was then packaged in canisters. The yeast waste preparation technique is summarized in Figure 2 :

A protein-rich nutrient medium was then prepared and used for the development of Bactrocera dorsalis larvae using the following technique. Preparation of 1kg of nutrient medium requires:

To prepare ( Figure 3 ), add 295 ml distilled water to the brewer’s yeast waste, heat to boiling for 30 min and cool completely. Measure out the 295 ml of brewer’s yeast waste, add it to the 295 ml of distilled water, then add the various quantities of the above nutrient medium components while stirring, in the following order: sugar, rice bran, rice litter, corn flour, soy flour, nipagine. Measure out 2.95 ml of HCL and add to the medium, then stir. Finally, package the medium in a plastic jar and store in the fridge for various tests on B. dorsalis eggs and development.

Once the protein-rich nutrient medium has been prepared, the procedure for incorporating it into each of the 8 organic extracts is as follows:

- 15 mg of each of the eight organic extracts of natural substances (Capsicum annum L, Cleome viscosa L, Mytragina inermis (Wild) kuntze, Strophantus hispidus A. DC, Ocimum basilicum L, Cassia nigricans Vahl, Cassia occidentalis L., Pseudocedrela kotchyi) was first collected and then dissolved in 1 ml DiMethyl SulfOxide (DMSO) in a microtube. The mixture was vortexed to ensure complete dissolution;

- Next, 29 g of the local nutrient medium was taken and added to 1 ml of organic extract of each natural substance. The untreated control received 29 g of nutrient medium to which 1 ml of DMSO was added (i.e. 0.517 mg/g concentration). The resulting mixture of each natural substance was placed in Petri dishes (25 dishes in all) containing a piece of lotus paper.

Testing the biological activity of total extracts from eight plants on B. dorsalis eggs and development.

20 eggs were placed in each dish. This set-up was repeated 10 times for each of the eight extracts, for a total of 250 Petri dishes used in the experiment. Daily observations were made and the number of eggs hatched was counted every 24 h, 48 h, 72 h and 96 h. The experiment was monitored until the 25th day of the Bactrocera dorsalis cycle. This made it possible to determine the rates of egg hatching, pupation and fly emergence. In this test, we did not look separately at larval mortality, as Stage 1 larvae do not sink into the nutrient medium to feed, and are therefore not visible under a magnifying glass once dead.

These rates were calculated on the basis of the formulas.

$\text{Hatching}\text{rate}=\frac{\text{Number}\text{of}\text{eggs}\text{hatched}}{\text{Total}\text{number}\text{of}\text{eggs}}\ast 100$

$\text{Pupaison}\text{rate}=\frac{\text{Number}\text{of}\text{pupae}\text{recorded}}{\text{Number}\text{of}\text{eggs}\text{incubated}}\ast \text{100}$

$\text{Emergence}\text{rate}=\frac{\text{Number}\text{of}\text{pupae}\text{emerged}}{\text{Number}\text{of}\text{pupae}\text{put}\text{in}\text{emergence}}\ast 100$

Breeding Bactrocera dorsalis and Ceratitis cosyra in the laboratory

The B. dorsalis adults used in this test represented the 60th and 40th generation of flies reared in the laboratory, respectively. Cotton soaked in water and a mixture of sugar and a protein source were placed in the cages for fly nutrition.

Phytochemical screening of the total active extracts was carried out using the chemical tube characterization test method described by [16] Ciulei (1982) and adapted by the chemistry laboratory of the Institut de Recherche en Sciences de la Santé (IRSS). These are tests of general and/or specific chemical reactions, coloration or complexation in a liquid medium. The characterization tests are listed in Table 2 .

The collected data were entered and grouped using Microsoft Excel 2010. The database was then imported into R software version 3.6.0 for various statistical analyses. The data were first tested for normality using the Shapiro-Wilk test.

This was followed by a comparison of means test at the 5% threshold (effects of total extracts from eight plants on eggs, pupation and emergence of Bactrocera dorsalis in the laboratory).

Table 3 shows the test samples, mass and yield of the various organic extracts. A 400 g test portion of each sample was successively macerated for 24 h with 1500 ml of n-hexane, ethyl acetate and a methanol-water mixture (80:20, v/v). Yields of apolar extracts (hexane) ranged from around 0.66 to 5.76%. The Mytragina inermis sample gave the highest yield (5.76%) of aqueous macerated extracts, while the lowest yield (0.66%) was observed with the Ocimum basilicum sample. The mass of apolar lipophilic extracts (n-hexane) varied from 2.49 g to 21.71 g. The Mytragina inermis sample gave the highest mass of lipophilic extracts, and the Ocimum basilicum sample the lowest. The proportion of medium-polar extracts (ethyl acetate) ranged from 1.64% to 5.91%. The highest level of medium-polar extracts was found in the Ocimum basilicum sample, and the lowest in the Cassia occidentalis sample. The level of polar extracts (hydro-methanolic) varied from 1.45% to 18.13%. The highest content was obtained with the M. inermis sample, and the lowest with the Cassia occidentalis sample ( Table 3 ).

Biological activity tests on the 8 total extracts enabled us to select Cleome viscosa hexane (with high insecticidal activity on B. dorsalis adults), Capsicum annuum (with hexane and ethyl acetate) and Strophantus hispidus (with methanol), which showed high insecticidal activity on B. dorsalis eggs and development. Chemical characterization tests in solution in test tubes revealed groups of potentially bioactive phytochemical compounds in total active extracts of Cleome viscosa (in hexane), Capsicum annuum (in hexane and ethyl acetate) and Strophantus hispidus (in methanol). The results of phytochemical characterization tests on C. viscosa are given in Table 4 , as are those on C. annuum and S. hispidus. Analysis of hexanic C. viscosa revealed the presence of sterols, terpenes and carotenoids. Analysis of C. annuum and S. hispidus also revealed groups of potentially bioactive chemical compounds: sterols and triterpenes, coumarins and derivatives, anthraquinones, alkaloids, flavonoids, tannins and cardenolides. The presence of the various constituents in the plant is revealed by extractions using different solvents, which indicate a greater or lesser number of compounds depending on their solvent affinity.

Legend: += present; (-) = absent; ± = traces.

No hatching of B. dorsalis eggs was observed after 24 h of observation.

Figure 4 shows the effect of organic extracts on Bactrocera dorsalis egg hatchability after 48h of observation. Statistical analysis shows a highly significant difference. Among the organic extracts of natural substances tested, the hexanic extract of O. basilicum produced the lowest rate of egg hatching (57.37%), down 30.42% on the untreated control. In contrast, methanolic extracts of S. hispidus produced a higher hatching rate of 99%, an increase of 20.07% on the untreated control. This was followed by the methanolic and ethyl acetate extracts of M. inermis (87.42%, an increase of 18.25%), the methanol extracts of C. viscosa (97%, an increase of 17.65%) and C. nigricans (96%, an increase of 16.43%).

Statistical analysis shows a highly significant difference ( Figure 5 ). The lowest hatching rate was obtained with the methanol extract of O. basilicum (86.5%), a reduction of 11.29% compared with the untreated control. On the other hand, the hexane, ethyl acetate and methanol extracts of C. annuum and the methanol extract of S. hispidus resulted in a higher hatching rate of 100% after 72 h, i.e. an increase in hatchability of 2.56% compared with the untreated control. A similar hatching rate was obtained in the presence of ethyl acetate extract of M. inermis (99.5%), hexanolic and methanol extracts of C. nigricans (99%) and hexanolic extract of C. viscosa (99%).

Statistical analysis shows a highly significant difference ( Figure 6 ). The hexane, ethyl acetate and methanol extracts of C. annuum and the methanol extracts of S. hispidus, C. nigricans, C. occidentalis, M. inermis and the hexane, ethyl acetate extracts of C. viscosa, P. kotschyi, and the hexane extract of O. basilicum showed a total hatching rate of 100%. In contrast, the ethyl acetate and methanol extracts of O. basilicum (99.5%) and the methanolic extract of P. kotschyi (99.5%) were identical. The methanol extract of C. viscosa showed an egg hatching rate of 99%, the lowest of all extracts.

Statistical analysis shows a highly significant difference ( Figure 7 ). The hexane extract of C. annuum obtained the lowest pupation rate (5.7%), i.e. a reduction of 90.42% compared with the untreated control, followed by the ethyl acetate extract (7%), i.e. a reduction of 88.23%, and the methanol extract of C. annuum (10.8%). Next came the hexane extract of C. nigricans at 16.4%, a reduction of 75.44%, and the methanol extract of Strophantus hispidus at 17%, a reduction of 71.43%. The least effective extracts were those of Cleome viscosa, particularly the methanol extract, which facilitated pupation of Bactrocera dorsalis at 36.9%, i.e. a reduction of 37.8% compared with the untreated control.

Statistical analysis shows a highly significant difference ( Figure 8 ). Only the hexane and methanol extracts of Capsicum annuum and Strophantus hispidus inhibited the emergence of Bactrocera dorsalis. The methanol extract of Pseudocedrela kotschyi also inhibited the emergence of B. dorsalis at 100%, i.e. zero B. dorsalis emerged, statistically indistinguishable from the two extracts of C. annuum and S. hispidus mentioned above. The other treatments stimulated B. dorsalis emergence to a greater or lesser extent than the untreated control.

Legend: - = no effect; +++++ = total effect; +++ = moderate effect; ++ = low effect.

All the results obtained on the development of B. dorsaslis are summarized in Table 5 . Based on these results, C. annuum and S. hispidus were selected for further work.

Chemical composition of total extracts

Phytochemical analysis of the three active extracts of C. viscosa, C. annuum and S. hispidus was carried out in the course of this study. This analysis revealed the presence of triterpenes, carotenoids, coumarins, anthocyanosides, alkaloids, flavonoids, saponosides and cardelonides in these organic extracts, with the exception of tannins, which are absent in C. annuum and present in C. viscosa and S. hispidus. These results are similar to those of [17] Ahouansinkpo et al. (2016), [5] Liljana et al. (2013) and [18] Gurnani et al. (2016) who also revealed the presence of these same constituents in the fruits of C. annuum and the leaves of C. viscosa and S. hispidus.

Biological activity of total extracts on B. dorsalis eggs and development

The 8 total plant extracts had no inhibitory effect on B. dorsalis egg hatching. This could be explained by the fact that the egg wall was not permeable to the penetration of the total extracts. Hence the high hatching rate of B. dorsalis eggs after 72 hours of observation. This means that the presence of chemical compounds with insecticidal properties in the various plant extracts had no effect on the hatching of B. dorsalis eggs. This result is in line with that obtained by [19] Boutchelta et al. (2005) who showed that the presence of compounds such as flavonoids in Capsicum frutescens extract had a weak effect on the hatching of Bemisia tabaci eggs.

Once the B. dorsalis eggs had hatched, the stage 1 larvae were unable to feed on the nutrient medium containing certain total extracts in order to continue their development. This means that some of the 8 extracts had an effect on the development of B. dorsalis larvae. The hexane extract of C. annuum was the most inhibitory, followed by the extract of Strophantus hispidus. This result could be explained by the presence of phytochemical compounds that had anti-appetant effects on the subsequent development of B. dorsalis. These effects prevented their evolution into the adult stage. As pointed out by [20] Koul (2008), alkaloids, terpenoids and phenols have anti-appetent properties on the larvae of certain insect species. The toxicity of alkaloids by contact and/or ingestion on B. tabaci larvae was demonstrated by [19] Bouchelta et al. (2005). This could explain the similar results obtained in this experiment.

Also, the presence of tannins and saponins in nutrient media from total C. annuum and S. hispidus extracts could be toxic to larvae. Saponin toxicity has been demonstrated in Callosobruchus chinensis larvae ( [21] Applebaum et al., 1969).

Furthermore, B. dorsalis larvae that were able to feed on these nutrient media containing total extracts of C. annuum and S. hispidus were unable to develop into adults. This could be explained by the fact that the larvae may have paralyzed the peristaltic movements of their intestines, thus inhibiting their development. This result is similar to that of [22] Fortin et al. (2000), who demonstrated this paralysis of peristaltic movements in the gut of Helicoverpa armigera and tomato thrips due to the presence of secondary metabolites.