Fruit fly surveillance in Togo (West Africa): state of diversity and prevalence of species

The production and marketing of fruits, especially mango, in sub-Saharan Africa are threatened by fruit flies (Diptera: Tephritidae). This baseline study analysed the biodiversity of fruit fly species, in mango orchards, in different ecological zones of Togo. Traps used to monitor the flies in the orchards consisted of dry baits, made from four types of parapheromones specific to the males of different species. Forty species of fruit flies were identified in the mango orchards in Togo. The most common species were Bactrocera dorsalis (Hendel), Ceratitis cosyra (Walker), Ceratitis fasciventris (Bezzi), Ceratitis capitata (Wiedemann), Ceratitis bremii Guérin-Méneville, Dacus bivittatus (Bigot), Dacus humeralis (Bezzi), Dacus punctatifrons Karsch and Zeugodacus cucurbitae (Coquillett). The invasive B. dorsalis and the endogenous species, C. cosyra were dominant in the mango producing areas of Togo because their prevalence were very high (B. dorsalis: 2.1 ≤ flies per trap per day (FTD) ≤ 472.2; C. cosyra: 0.34 ≤ FTD ≤ 97.28). There was no area free from fruit flies in Togo during the study. These results constitute an essential reference in the future evaluation of the effectiveness of any control activities initiated in Togo against fruit flies.


Introduction
The consumption of fruits and vegetables is at the core of a healthy diet (Amiot-Carlin et al. 2007;Griep et al. 2010;Zhang et al. 2011). Fruits and legumes are an important source of water, fiber, vitamins (A, B9, C, E), minerals (Calcium, phosphorus, Zinc, Iron, Selenium, Magnesium) and antioxidants necessary for the proper functioning of the body (Amiot-Carlin et al. 2007). The demand for healthy vegetable and fruit products is increasing in West Africa where consumers are increasingly purchasing expensive, good quality fruits and vegetables because of their nutritional importance (FAO and BAD 2015). Moreover, several exploratory and epidemiological studies have shown that high consumption of vegetables and fruits reduces the risk of cardiovascular diseases and the occurrence of some cancerous and other chronic diseases (Zhang et al. 2011;He et al. 2007;CIRAD 2009). To respond in part to the growing urban demand for fruits and vegetables, especially the diversification of diet as a source of welfare, West African countries are developing their horticultural sectors and production has more than doubled in 26 years, increasing from 14,403,034 tons in 1980 to 32,668,682 tons in 2008, with average growth rates of 1% and 1.7% for fruits and vegetables, respectively (FARM 2008). The production and trade in fruits and vegetables is an important income source for countries in general and those of sub-Saharan Africa in particular. In Togo, fruit and vegetable production is estimated at around 560,000 tons in 2017 (InterFaxPress 2019). Apart from their importance in food security, the production and trade in fruits and vegetables is an important income source for sub-Saharan African countries. Indeed, in 2017, the horticultural sector contributed an income of 4.5 billion FCFA to the national economy of Togo, with 30,265 tons exported (Vert Togo 2019). Given its importance, the fruit and vegetable sector is one of the key agricultural sectors targeted for promotion in Togo's National Development Plan (PND). Of those targeted, mango, Mangifera indica L. (Anacardiaceae) is one of the important commercial fruit tree crops in the world (Akotsen-Mensah et al. 2017). According to theses authors, mango grows in the tropical and subtropical climates, making it second to citrus in terms of production. It continues to be an important tropical fruit for Sub-Saharan African economies and serving as a major source of nutrition for the rural population, playing a major role in poverty reduction as well as being a potential export product (Vayssieres et al. 2008). In Togo, mango grows in the five economic regions of the country on approximately 1,523 hectares with a production of 340,000 tons and a yield of 3.7 tons per hectare (InterFaxPress 2019; MAEDR 2020). Apart from the native variety, the other common varieties of mangoes grown in Togo are: Kent, Palmer, Eldon and Somnole (MAEDR 2020). Despite their economic and nutritional importance, horticultural and mainly mango production and trade are threatened by pests affecting the implementation of the horticultural sector's development policies. Of insect pests attacking fruits and vegetables fruit flies (Diptera: Tephritidae) ranked as being one of the most important economically. The losses attributed to fruit flies in sub-Saharan Africa have been increasing in recent years because, in addition to the indigenous species like Ceratitis cosyra (Walker) (Diptera: Tephritidae) which attack fruits and reduce their nutritional and trade values, a new species, Bactrocera dorsalis (Hendel) (Diptera: Tephritidae) originated from Southeast Asia was first detected in Kenya in 2003, was introduced from Southeast Asia (Pouillès-Duplaix 2008;Vayssières et al. 2014). In a few years, this species has spread throughout West Africa (Drew et al. 2005), as an invasive species. B. dorsalis is economically very important because West Africa is a favorable ecological niche for its development, in terms of climate and availability of preferred host plants (De Meyer 2010;Goergen et al. 2011;Gomina 2015). Being the main species associated with the mango, B. dorsalis and C. cosyra are the major constraints on mango production and trade today. Mango losses caused by these major pests are estimated at 17% at the beginning of the harvest period and can exceed 70% at the end of the period (Vayssières et al. 2009a, b). As fruit flies are classified as "quarantine insects", any container from Africa containing perforated fruits is intercepted, seized and destroyed by incineration at ports and airports in Europe, causing serious economic losses to African exporters (Pouillès-Duplaix 2008;Guichard 2009). Between 2006 and 2007, interceptions associated with fruit flies increased by 23% and the annual economic losses were estimated at more than USD 42 million in Africa and more than one USD 1.0 billion worldwide (STDF 2009).
For West Africa, interceptions related to fruit flies at the EU border, cost around € 9,000,000 in mango exports in 2006 (ARAA 2018). In order to reduce the level of fruit fly infestations below economic thresholds in orchards and also, to avoid interceptions of fruits and vegetables in general and mangoes in particular (from ECOWAS countries), the European Union countries and eleven ECOWAS countries (Benin, Ivory Coast, Burkina Faso, Gambia, Ghana, Guinea Conakry, Mali, Nigeria, Senegal, Togo) have decided to combine their efforts to control fruit flies. ECOWAS has therefore initiated a sub-regional project entitled "Project to support the regional plan for the control of fruit flies in West Africa (PLMF)". One of the most important components of this project is the accurate monitoring of fruit fly populations for early warning and adequate decision-making for controlling the pests in the interest of farmers. The study aimed to: (i) present the state of diversity of fruit fly species in different mango production areas in Togo and (ii) point out the boundaries of the areas considered to be infested or free from fruit flies by assessing the prevalence of species at the beginning of the project. This is fundamental to developing accurate management methods, targeting the dominant fruit fly species in the agro-ecosystems or mango producing areas in Togo, and gathering a reference database to facilitate future assessment of the effectiveness of management activities implemented in the country.

Study area: geographic location, ecological characteristics and choice of the orchards
The study was carried out in mango orchards in Togo, West Africa. The study area extends from the South to the North of this country, between 06.35964°N and 10.99362°N and from East to West, between 000.31449°E and 001.29350°E. A total of twenty orchards were chosen based on their areas (minimum area of 2 ha); age (between 5 and 40 years); nonapplication of phytosanitary measures and all the varieties of mangoes identified ( Fig. 1 and Table 1). The orchards are geographically located in two of the five West African mango producing belts recognized by the PLMF; one in the South (TG 1 ) or wet area and the other in the North (TG 2 ) or dry area. They are distributed in the five ecological zones of Togo described by Ern (1979) and Brunel (1984) as follows: -the North-East, North-West, Center and South-East of ecological zone I (orchards TG 2 V 4 , TG 2 V 5 , TG 2 V 7 and TG 2 V 10 ) or the northern plains with Sudan-savannas, dry forests, meadows around ponds and agroforestry parks. The climate is Sudano-tropical type with a sin- gle rainy season (June-October) and a longer dry season dominated by the harmattan (November-May). The average annual rainfall is around 1000 mm and the average annual temperatures are generally high, reaching 28 °C while relative humidity is low (53 to 67% RH); -the North-East, Center and South of ecological zone II (orchards TG 1 V 5 , TG 1 V 9 , TG 1 V 10 , TG 2 V 2 , TG 2 V 3 , TG 2 V 6 , TG 2 V 8 , TG 2 V 9 ) or part of the Northern mountains dominated by a mosaic of dry forests, mountain savannas and crop lands. The climate is a Sudano-Guinean type with one rainy season (April-October) and one dry season (November to March), including the harmattan. The temperature and relative humidity are closed to those of Zone I; -the South-East, Center and North-East of ecological zone III (orchards TG 1 V 3 , TG 1 V 4 , TG 2 V 1 ,) or the central plains made of woody Guinea-savannas, dry forests, cropped lands as well as forest galleries. The climate of the area is lowland Guinea-type, with one rainy season (April to October) and one dry season (November to March). The average annual temperatures vary between 26 and 30 °C while the average annual rainfall is around 1200 mm; -the South-west of ecological zone IV (orchard TG 1 V 7 ) or the southern section of "mount Togo" dominated by Semi-deciduous rainforest, cropped lands and Guinea-savannas. It is influenced by a transition subequatorial climate, that is, a mountain climate characterized by one rainy season (March-November) and one dry season (December-February) with decreased rainfall in August. The average monthly temperatures varied between 22 and 26 °C during the year, the annual average rainfall is around 1,651 mm and the relative humidity is always high (70 to 99% RH); -the North-West and West of ecological zone V (orchards TG 1 V 1 , TG 1 V 2 , TG 1 V 6 , TG 1 V 8 ) or the coastal plain of Southern Togo characterized by Guinea-savannas, forest patches, and cropped lands. Here there is a subequatorial climate characterized by two rainy seasons (April-July and September-October) alternating with two dry seasons (August and November-March). Average monthly temperatures vary between 25 and 28 °C during the year and average annual rainfall is around 930 mm with a high relative humidity throughout the year (73 to 90% RH).

Design of the Fruit fly capture device
The fruit fly traps were made with the aid of monitoring traps using a dry bait or parapheromone specific to males: Tephri Trap of the McPhail type (IAEA 2003). The  Vayssières et al. 2004;FAO 2014) and consist of: (i) Methyl Eugenol (ME) which attracts mainly males of Bactrocera spp. and species of the subgenus Ceratitis McLeay (Pardalaspis); (ii) Cue Lure (CUE) which attracts mainly males of several species of the genus Dacus and individuals of the Zeugodacus cucurbitae (Coquillett) species; (iii) Terpinyl Acetate (TA) and (iv) Trimedlure (TM) which attract males of the genus Ceratitis. An organophosphate chemical insecticide, DDVP or dichlorvos (2,2-dichlorovinyl dimethyl phosphate) was used to kill entrapped flies. The parapheromones were renewed every 6 weeks and the chemical insecticide every two months to maintain the effectiveness of the trap during the study period (Gomina et al. 2012). The traps were installed in the orchards from May 3 to June 8, 2018 according to the fruit fly monitoring system set up by the PLMF. In effect, a mango tree located at the center of each orchard was marked. Around this central point, 4 other mango trees forming a rhombus with sides 100 m and having as center the previously identified central point were also marked. On 4 mango trees located around each of these 4 points forming the rhombus, 4 traps each containing a parapheromone were installed: Methyl Eugenol, Cuelure, Tridmedlure and Terpinyl Acetate traps with North, South, East and West orientation respectively. Sixteen traps were installed per orchard, each with parapheromone, repeated 4 times. A total of 320 traps were installed for monitoring fruit flies in the selected 20 orchards. The traps were placed under the crown, shaded by leaves 2 m from the ground. They were inspected during a period of one month and 6 days started from 3 rd May to 8 th June 2018. The collection of the traps' catches of fruit flies were carried out weekly from 25 th May 2018. A total of 2 surveys were carried out in this period. Individuals of the Tephritidae species captured were stored per type of parapheromone and orchard in 70% ethanol and transported to the laboratory for identification.

Identification of fruit flies
The Tephritidae captured by each trap were sorted and identified at the Applied Entomology Laboratory (LEA) of the University of Lomé, using dichotomous keys (De Meyer 1996;De Meyer 2000;White 2006;De Mayer and White 2008;Virgilio et al. 2014) and identification key leaflets of the main fruit fly species in West Africa, provided by the PLMF. Also, comparisons with the reference collection of Tephritidae from LEA (samples of whose species have been confirmed by the entomology section of the Royal Museum for Central Africa (MRAC) in Tervuren in Belgium) were made to revise the identification.

Data analysis
The trapped fruit flies were counted by species, orchard and date of collection. The diversity of Tephritidae in orchards was expressed in terms of alpha diversity (α) and beta diversity (β). The calculations were done in R (R Core Team 2018) with the entropart package (Marcon and Hérault 2015).
The α diversity is the number of species coexisting in a uniform habitat of fixed size (Marcon , 2018. It was determined by the species richness or number of species of fruit flies per orchard. The Simpson and Shannon-Wiener diversity index as well as the Pielou evenness index that often comes with the Shannon-Wiener index and Engen rarity variance (EVS) (Marcon 2018) were estimated on the basis of the Tephritidae samples from the catches. The Simpson Index (SI) measures the probability that two randomly selected individuals are of different species. It varies from 0 to 1, diversity is highest for SI close to 1 and lowest for SI close to 0. This diversity is also a decreasing function of the regularity of the species. Considered a measure of biodiversity as well as a quantitative measure, the Shannon-Wiener index (H ') varies from 0 (single species, or one species dominates all the others) to log2 (S) (all species have the same abundance) where S is the number of species. It is maximum when the frequencies of the species encountered show little difference between them. The Pielou evenness index (E) defines the regularity of the distribution of species and corresponds to the ratio of the Shannon index to its maximum value. It is close to 0 if the abundances of the species encountered are very dissimilar and close to 1 if all the species have similar abundance. The Engen rarity variance is the variance of the information function, Shannon's entropy. The closer its value is to 0, the more equitable is the geographic area.
The beta diversity measures the difference or similarity between habitats or samples in terms of species diversity. It permits comparison of the diversity between the communities and was estimated by the Jaccard index (J) between two orchards. The Jaccard index is 1 if there is complete similarity between the localities compared and 0 if the latter have no common species. A projection of the dissimilarity matrix from the Jaccard indexes on the first main coordinates made it possible to highlight similarities and dissimilarities between the orchards in terms of diversity of fruit flies using the R ade4 package (Bougeard and Dray 2018).
The analysis of the mean number and percentage of fruit fly species was performed using analysis of variance (ANOVA), followed by Student-Newman-Keuls (SNK) comparison tests when the value of F was significant at the 5% level.
The prevalence of the dominant fruit fly species was determined by calculating the number of flies per trap per day (FTD) according to IAEA (2003) and Rodríguez-Rodríguez et al. (Rodríguez-Rodríguez et al. 2018), applied in the case where no control measures were taken in the orchards considered. According to IAEA (2003), the value of FTD determines the type of phytosanitary measure to be considered in the implementation of international standards for phytosanitary measures: -if FTD˃1, the area is considered infested with fruit flies and requires the full complement of phytosanitary measures; -if 0.1 ≤ FDT ≤ 1, the actions to be taken are suppressing the species of fruit fly; -FDT ˂ 0.1 calls for an eradication process applied in an area free from fruit flies; -FTD = 0 calls for exclusion measures which are processes applied to minimize the risk of introducing or reintroducing the species in an area free from fruit flies. Trapping is applied to determine the presence of species that are subject to exclusion measures and confirms or rejects the status of a free zone.

Species richness of fruit flies in the study area
A total of 40 species of Tephritidae were identified in the five ecological zones based on the trap catches using the four types of parapheromone ( The number of species caught per orchard varied from 4 (10% of the species) in the TG 2 V 10 orchard to 26 (65% of the species) in TG 2 V 2 . The relatively more species-rich orchards were found in ecological zones II, III and IV. The majority of orchards, relatively poor in species were located above latitude 09° 30′ 0′' N, in the north of Togo in the ecological zone I (Fig. 1).  Table 3). The remaining 38 species represented only 3.09% of the population of fruit flies per orchard (361 individuals per orchard). Whether in zone 1 or zone 2, the mean numbers and proportions of B. dorsalis and C. cosyra in the mango orchards were still higher than other fruit fly species numbers ( Table 3).

Species diversity of Tephritidae
Analysis of the species diversity of Tephritidae showed that the Simpson diversity indexes from the different orchards were generally low, indicating a low diversity of species (Table 4). These results are confirmed by the low values of the Shannon-Wiener diversity indexes which are well below the maximum value (Hmax). The Pielou evenness index with relatively very low values correspond to orchards TG 1 V 9 , TG 1 V 10 and TG 2 V 9 (Ecological Zone II) and TG 2 V 5 (Ecological Zone I). The weak Simpson index from these orchards indicated that they had a low regularity of occurrence of the species. The distribution of species in these orchards was marked by the dominance of the B. dorsalis species.
Computation of the Engen rarity variance showed that orchards TG 1 V 2 and TG 1 V 3 present the most homogeneous sample distribution, while orchards TG 2 V 5 (lower species richness) and TG 2 V 10 show the greatest disparities marked by an unequal distribution of probabilities and low species richness.

Beta diversity
Species community analysis of the Tephritidae species showed that several orchards at the study site had similar species because, the Jaccard index was higher than 0,50 ( Table 5). The highest similarity was observed between orchards TG 1 V 5 and TG 2 V 9 ; TG 1 V 7 and TG 1 V 9 ; TG 1 V 2 and TG 1 V 6 , with Jaccard indexes estimated at 0.94, 0.87 and 0.84, respectively. Orchards TG 1 V 7 and TG 2 V 4 , TG 1 V 9 and TG 2 V 4 were those in which very low similar species were recorded, with a Jaccard index of 0.13 each.
Bactrocera dorsalis, C. cosyra and C. fasciventris were the three Tephritidae species that were present in all the orchards studied (Table 1). They are followed by C. capitata, D. humeralis and D. punctatifrons which were present in 16 orchards and C. bremii, D. bivittatus and Z. cucurbitae which were trapped in 15 orchards.    The uncommon species that were present in only one orchard were C. colae (TG 1 V 7 ), C. flexuosa (TG 1 V 10 ), Ceratitis sp3 (TG 2 V 2 ), D. annulatus (TG 2 V 1 ), Dacus disjunctus (Bezzi) (TG 1 V 9 ), Dacus seguyi (Munro) (TG 2 V 2 ) and unidentified Tephritidae 2, 3 and 4 that were recorded in orchards TG 2 V 3 , TG 2 V 2 , respectively. The representation of the orchards in the different zones in a principal coordinate analysis based on the Jaccard distances allowed to better group the orchards having similar species (Fig. 2). Four main groups of orchards were identified: -group 1 comprised orchards TG 1 V 4 , TG 1 V 5 , TG 2 V 1 , TG 2 V 2 , TG 2 V 3 , TG 2 V 8 and TG 2 V 9 that had more than half of the species similar; -group 2 are orchards TG 1 V 1 , TG 1 V 8 , TG 1 V 2 , TG 1 V 6 , which also had more than half of the species similar; -group 3 includes orchards TG 1 V 3 , TG 1 V 7 and TG 1 V 9 which were similar; -group 4: orchards TG 2 V 4 , TG 2 V 6 and TG 2 V 7 ; Orchards TG 2 V 5 , TG 2 V 10 and TG 1 V 10 were not well represented in the projection made.

Species prevalence
Based on species diversity, B. dorsalis and C. cosyra were common to all the mango orchards in the study area and singly represented 98.15% of the Tephritidae. Indeed, the number of flies per trap per day (FTD) of B. dorsalis was the highest in all the orchards and varies from 2.12 (TG 2 V 4 ) to 472.1 (TG 1 V 7 ) (Fig. 3). The prevalence of C. cosyra was lower and ranged between 0.34 (TG 2 V 4 ) and 97.28 (TG 1 V 4 ).

Discussion
The trap-capture method used in this study made it possible to record 40 species of Tephritidae, 11 of which were not be identified up to species level. Those identified represent 77% Table 3 Mean number and proportion of the Tephritidae species per orchard (mean ± Std. Error) in different mango producing areas * In the same colon, means followed by a different letter are significantly different (ANOVA-1 followed by SNK test, P < 0.05) ** TG 1 : Mango producing area 1; TG 2 : Mango producing area 2; TG: Total orchards in the study area; Number of replications: 10 per zone (TG 1 or TG 2 ) and 20 in the study area (TG) Species/Statistics Average number of the species per orchard (Average proportion per orchard) * Zone 1 (TG1)** Zone 2 (TG2)** All zones (TG)**  (Amevoin 2009;Ouédraogo et al. 2011;Gomina et al. 2012;De Meyer et al. 2013;N'Da 2018;Zida et al. 2020). The relatively high species richness obtained with a single study method applied over a month is an evidence that the species found in the mango orchards (which were at fruit ripening stage) and their surrounding vegetation, a favorable abiotic (temperature, relative humidity, precipitation, etc.) and biotic (resting sites, host plants) conditions for their development (Virgilio et al. 2009;Goergen et al. 2011, Bota et al. 2020. These may explain the diversification in species of the two main genera Ceratitis and Dacus. Indeed, Vayssières et al. (2010) and Gomina (2015) reported the presence of the host plants of several Tephritidae species identified in this study. The highest species richness was observed in the ecological zones II, III (at the level of the latitude of Sokodé and Bafilo) and IV. This can be explained not only by the diversification level of wild host plants but especially by the fruit trees cultivated in these areas. In fact, the zones II, III and IV are considered in Togo as major fruit producing regions, which is not the case for the ecological zone I in the Northern part of the country where the diversity of host plants may be lower, probably because of the less favorable ecological and climatic conditions. Similar study conducted by Bota et al. (2020) also showed that the population of fruit flies that infest mango orchards in the central Mozambique may come from the marginal area, especially from local varieties of mango which get matured earlier.
Among the species of Tephritidae identified in mango orchards in Togo, B. dorsalis, C. cosyra, C. capitata, C. fasciventris, C. silvestrii, C. anonae, C. quinaria, C. ditissima are known to be associated with various fruit plant species in Africa (Vayssières et al. 2004;White and Elson-Harris 1992;Vayssières et al. 2009a, b;Nboyine et al. 2013;Vayssières et al. 2015;Zida et al. 2020). According to Vayssières et al. (2015) the presence of several fruit fly species on mango has considerably reduced the potential economic benefits of growing this fruit tree in West Africa. As for the Tephritidae species of the genera Dacus and Zeugodacus, they are known to attack mostly plant species belong to Cucurbitaceae, Passifloraceae and Apocynaceae (White 2006;Virgilio et al. 2009). In general, the mango fruit flies identified during our study are native, except of B. dorsalis and Z. cucurbitae reported as exotic and invasive (Goergen et al. 2011;De Meyer et al. 2013).
The very high proportions and prevalence of B. dorsalis in all the mango orchards in Togo is an evidence that this invasive species has settled in Togo and has undoubtedly constituted a threat to mango and other fruits production.  (Bota et al. 2018(Bota et al. , 2020N'Da 2018;Zida et al. 2020). The high proportion and prevalence of B. dorsalis in mango orchards do not seem to be explained solely by the effectiveness of the attractant used for this species but mainly by its very good adaptation to the agro-ecological conditions of the study area. In addition to being polyphagous, Gomina (2015) has shown that under the Guinea zone conditions in Togo, B. dorsalis was very prolific because the female could lay an average of 538 eggs during its life time with an offspring survival rate estimated at 67%. C. cosyra is the second most important species to which special attention must be paid in Togo and in West Africa (Ouédraogo et al. 2011;Zida et al. 2020). This species was considered as adapted to Sahelian and Sudanese zones and absent from agro-ecological zones of the humid forests of West Africa (Vayssières et al. 2014). However, it was found in all ecological zones of the study area in Togo including zone IV dominated by dense semi-deciduous forests. The presence of C. cosyra in the humid forest zone in Togo is probably due to human activities which negatively impact ecological zone IV and climate change with its proven consequences in recent years. The species richness of Tephritidae frugivores in Togo seems to be important but the different diversity indexes pointed out a low species diversity. This helps to note that the potential species diversity in Tephritidae should be high. This result is in line with the work of De Meyer et al. (2013). Thus, the application of other methods such as the incubation of fruits from different ecological zones and the use of other attractants will probably make it possible to accurately record all species present in Togo.
Analysis of the Tephritidae community from the catches shows that several fruit fly species that attack the mango orchards studied were similar. The Jaccard distances analysis showed that orchards in the same ecological zone tend to be similar in terms of the fruit fly species present. This result may indicate the homogeneity of the abiotic and biotic conditions (the vegetation in particular represented by the host plants cultivated inside the orchard but also by the wild host plants around the latter) in the same zone allowing the species of Tephritidae to find the same resources for their survival and development. This result is similar to the one of Ouédraogo et al. (2011). The most common and wide-ranging species in our study area were B. dorsalis, C. cosyra, C. fasciventris, C. capitata, D. humeralis and D. punctatifrons, C. bremii, D. bivittatus and Z. cucurbitae. These species are known to be well represented in West Africa (Gomina et al. 2012;De Meyer et al. 2013;Ouédraogo et al. 2011;Vayssières et al. 2015). The species C. colae, C. flexuosa, Ceratitis sp and some species of the genus Dacus that were not be identified up to the species level were present in a single orchard and can probably be considered as being rare in the area but, monitoring over a long period will make it possible to confirm their status.
The prevalence of the two species considered dominant in terms of number of flies per trap per day (FTD) is very high and therefore indicates that all the orchards studied have very high incidence of B. dorsalis and C. cosyra. Thus, they remain the species of economic importance in Togo. According to the recommendations of IAEA (2003), this result indicates that it is necessary to implement phytosanitary protection actions against these formidable species of fruit flies.

Conclusion
This study confirms that there is no area that is free from fruit flies in Togo. A total of 40 species of fruit flies were identified in the surveyed mango orchards that were at fruit maturation period. The diversity indexes estimated in this study show that species other than those reported in the study could be present. The most common species are B. dorsalis, C. cosyra, C. fasciventris, C. capitata, D. humeralis and D. punctatifrons, C. bremii, D. bivittatus and Z. cucurbitae. The most abundant of these are the invasive B. dorsalis and the endogenous C. cosyra which have a very high prevalence. Hence, it is essential to determine, in all agro-ecological zones, the economic thresholds of these most abundant fruit flies (B. dorsalis and C. cosyra) of the economically important fruits and vegetables. This will lead to the establishment of a sustainable fruit flies management program in Togo.

Acknowledgement
This study was carried out with the financial support of the European Union (EU), Agence Française de Développement (AFD) and the Economic Community of West African States (ECOWAS) through the Support Project to the Regional Plan for the Control of Fruit Flies in West Africa (PLMF). We thank all the staff of the SOGEROM-COLEACP-BERD consortium who ensured the regional coordination of the project. Additional financial support was also provided by CABIs Plantwise Plus programme. Plantwise Plus is supported by the UK Foreign, Commonwealth and Development Office (FCDO), Netherlands Directorate General for International Cooperation (DGIS), European Commission Directorate-General for International Cooperation and Development (DEVCO) and the Swiss Agency for Development and Cooperation (SDC). We also thank the staff of "Direction de la Protection des Végétaux" of Togo, which was responsible for coordinating the implementation of this project in Togo, and all owners of mango orchards and technicians responsible for collecting flies in the field, their counting and identification in the laboratory. Our thanks also go to Mr. Komlavi Yves Amevoin, Economic Statistician at Médecins Sans Frontières (Dakar) for his invaluable help in the statistical analysis of data as well as to Dr. Hodabalo Pereki (Département de Botanique de la Faculté des Sciences de l'Université de Lomé) for his contribution to mapping the study area and Dr Victor Clottey (CABI, Ghana) and Dr Tetevi Wilson-Bahun (Faculté des Sciences de l'Université de Lomé) for their contribution to improve the English language. Funding This study was carried out with the financial support of the European Union (EU), Agence Française de Développement (AFD) and the Economic Community of West African States (ECOWAS) through the Support Project to the Regional Plan for the Control of Fruit Flies in West Africa (PLMF). Additional financial support was also provided by CABIs Plantwise Plus programme. Plantwise Plus is 0 20 40  60  80  100  120  140  160  180  200  220  240  260  280  300  320  340  360  380  400  420  440  460  480  500   TG1V1  TG1V2  TG1V3  TG1V4  TG1V5  TG1V6  TG1V7  TG1V8  TG1V9  TG1V10  TG2V1  TG2V2  TG2V3  TG2V4  TG2V5  TG2V6  TG2V7  TG2V8  TG2V9  TG2V10 FTD (flies/trap/day)

Orchards
B. dorsalis C. cosyra Data Availability All data are available and insect specimens are conserved in laboratory.

Conflicts of interest
The authors declare no conflict of interest.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.