First record of Dacus ( Lophodacus ) hamatus ( Diptera : Tephritidae ) in Cameroon , with emphasis on a new host plant Lagenaria siceraria ( Cucurbitaceae )

1 Department of Animal Biology and Physiology, Faculty of Science, University of Yaounde I, P. O. Box 812 Yaounde, Cameroon. 2 Department of Biological Science, Faculty of Science, University of Ngaoundere, P. O. Box 454 Ngaoundere, Cameroon. 3 UMR 5175, CEFE (Centre d’Ecologie Fonctionnelle et Evolutive), Laboratoire de Zoogéographie, Université Paul Valéry Montpellier III, Route de Mende, 34 199 Montpellier cedex 5, France. 4 CIRAD/UMR CBGP, Campus International de Baillarguet, CS-30 016-F34 988 Montferrier sur Lez, France.


INTRODUCTION
The dipteran family Tephritidae, whose members are commonly known as true fruit flies, is among the largest and the most diversified pest species with a world-wide distribution (Cogan and Munro, 1980;Drew, 1989;De Meyer and Copeland, 2005;Aluja and Mangan, 2008;Prabhakar et al., 2012).In this family, the genus Dacus Fabricius which is widely distributed in the Old World, comprises 177 described species mainly associated with three host plant families: Cucurbitaceae, Passifloraceae and Apocynaceae (White and Elson-Harris, 2004; White, 2006; Virgilio et al., 2009;White and Goodger, 2009).
Although, most of fruit fly species are known to develop essentially on fruits or pods, some species are specialized on other plant organs such as flowers or stems on which they constitute a serious threat to the farm production (De Meyer and Copeland, 2005;White, 2006).Among Dacus, species from the subgenus Lophodacus Collart are mainly associated with male flowers of cucurbits.This is the case of Dacus (Lophodacus) hamatus Bezzi reported only from male flowers of Lagenaria sphaerica (Sonder) Naudin (Cucurbitaceae) and Dacus (Lophodacus) inornatus Bezzi from L. sphaerica and Peponium mackenii (Naudin) Engler (Cucurbitaceae) (White, 2006).
Until now, none of these fruit fly species has been reported in Cameroon, even less on male flowers of cucurbit species.However, at the best of our knowledge, no detailed study has been conducted under field conditions on whether the female fruit flies within Dacus (Lophodacus) exhibit some preference in the selection of oviposition sites among several species of cucurbits.The present paper reports the first record of D. hamatus in Cameroon with some notes on its newly described host plant, L. siceraria.

Study sites
The study was carried out in two sites situated in two agroecological zones in the southern part of Cameroon (Figure 1): (1) the southern plateau at Yaounde (Centre Region), in an experimental garden of the University of Yaounde I (Campus-UYI) (3°51'28.9"N,11°29'52.2"E,729 m asl), with humid tropical bimodal rainfall regime and (2) the western highlands at Koutaba (West Region), in the domain of the Catholic Cistercian Monastery, situated near Tchouffa village (5°38'47"N, 10°48'20" E; 1181 m asl) with humid tropical unimodal rainfall regime.Observations were conducted from November 2010 to February 2011 during dry period at both sites.

Biological model
The biological material involved the cucurbit species, which are among the most important and domesticated flowering plants with a world-wide distribution (Dupriez and De Leener, 1987;Schippers, 2004).In this family, seven species belonging to six genera: Citrullus lanatus (Thumberg) Mansfeld, Cucumis melo Linneaus, Cucumis sativus Linneaus, Cucumeropsis mannii Naudin, Cucurbita moschata (Duchesne ex Lamark) Duchesne ex Poiret, Sechium edule (Jacquin) Swartz and Lagenaria siceraria (Molina) Standley with two varieties (the bitter and the sweet) were experimentally followed.Over the all sample, flowers of L. siceraria and S. edule are white in colour while those of the five other species range from yellow to bright orange.

Experimental design
The experimental design consisted of 16 plots of 16 m 2 each (8 m × 2 m), with two plots per plant species or varieties.

Collecting data and identification methods
With exception of L. siceraria, the male and female flowers of the other cucurbit species bloomed early in the morning, at the same time and started closing up in the afternoon.Concerning L. siceraria, this period was highly variable according to the regions.
To estimate the severity of pest infestations, as well as the number of individual per infested organs, crop must be sampled.During samplings, all the flowers and fruits of each plant species or variety were weekly examined.On young flowers before blooming, the oviposition sites were externally materialized by the presence of some black spots, while on those bloomed, decaying started before withering and fallen (Figure 2A, B, C and D).The infested plant organs were collected and taken individually to the laboratory.Then flowers were dissected and larvae (Figure 2C), when present, were counted and reared in Petri dishes provided with a piece of wet cotton.The attacked fruits were recognized by the presence of exudates at oviposition site.Fruits were weighted and incubated individually in plastic boxes, containing sand at the bottom and covered with a piece of gauze.Incubations were followed up to the emergence of adult insects (fruit flies and/or parasitoids).Emerged adults of fruit flies (Figure 2E and F) and parasitoids were preserved in 70% ethanol for further identification and voucher collections.
Initial identification was carried out with the help of identification keys developed by White (2006) and White and Goodger (2009) for the flies (Diptera: Tephritidae), Wharton and Gilstrap (1983) and Wharton et al. (1992)

Data analysis
On each variety of Lagenaria siceraria, the rate of infested male flowers (rt inf fl) was evaluated based on the following formulae: In which nb inf fl is the number of infested flowers while nb fl ex is the number of flowers examined.The mean number of D. hamatus larvae per male flower for each variety of L. siceraria was also recorded.Then, because the data did not follow a Gaussian distribution, the Kruskal-Wallis test with associated Dunnett's posttest was applied in order to check differences in the studied parameters, both among the two varieties of L. siceraria and the study localities.The significance level was preset at 0.05.These procedures were performed in GraphPad Prism 5. 04 (Prism 5, 2010).

RESULTS
A total of 1756 and 1413 male flowers from the seven species of cucurbits were examined at Yaounde and Koutaba respectively (Table 1).Out of 338 and 394 infested male flowers of L. siceraria, 57.99% and 42.01%; 81.22% and 18.78% from the two varieties (the bitter and the sweet) respectively in both localities were oviposited by the female of D. hamatus (Table 1).Indeed, the maggots of this fly were observed feeding and ending their development cycle on the flowers of this host plant chosen by females of D. hamatus.
During day time, females of D. hamatus were rarely observed under field conditions.The oviposition was probably realized by night, mostly on the young male flowers before blooming.For instance, eggs and first instar larvae were frequently observed in the collected samples of this category; while old larvae were frequently observed in bloomed flowers of L. siceraria during sampling.
Under laboratory conditions, the larvae of D. hamatus were able to feed and achieve their development also on the flowers and fruits of the cucurbit species studied.Moreover, three other fruit flies species namely Dacus (Dacus) bivittatus Bigot, Dacus (Dacus) punctatifrons Karsch and Dacus (Didacus) ciliatus Loew were reared from the flowers of C. moschata.
In Yaounde, a mean of 46.29 ± 8.56 and 38.50 ± 7.08 male flowers of bitter and sweet varieties of L. siceraria respectively, were found infested during each sampling day by the larvae of D. hamatus, while in Koutaba, 101.20 ± 9.67 and 27.00 ± 4.35 male flowers of bitter and sweet varieties, respectively were also observed infested with the flower feeders.The density of D. hamatus expressed as the mean weekly infestation rate of male flowers was significantly (P < 0.05) greater in Koutaba than in Yaounde with 57.63 ± 3.41% (Min: 41.54%, Max: 76.92%) against 52.36 ± 5.72% (Min: 9.52%, Max: 90.91%), respectively.At the level of the plant variety, the weekly mean infestation rate of male flowers was significantly greater on the bitter variety than on the sweet (with the P < 0.05) at both the locations (Figure 3).Moreover, between localities, the weekly mean infestation rates of male flowers did not vary on the bitter variety (Figure 3).Koutaba hosted a significantly higher mean number of larvae per flower per sampling date (mean: 1.67 ± 0.05, Min: 1, Max: 5, N = 475) than Yaounde (mean: 1.45 ± 0.04, Min: 1, Max: 6, N = 394) (P < 0.05).Also, compared to the sweet variety, the bitter variety appeared the most suitable as it hosted significantly greater number of larvae per male flower at both locations (P < 0.05) (Figure 4).While on the bitter variety the mean number of larvae per male flower was significantly greater in Koutaba (P < 0.05), on the sweet variety, a reverse phenomenon was recorded (Figure 4).A total of 3225 (57.12% of the total number of fruits investigated) and 1313 (47.87%) infested fruits of the seven cucurbit species investigated in this study were harvested at Yaounde and Koutaba (Table 2) specific to the male flowers of L. siceraria at least under field conditions.An unidentified parasitoid of the genus Fopius Wharton, (Hymenoptera: Braconidae: Opiinae) was recorded from parasitized larvae of D. hamatus from samples of Koutaba location.However, the parasitism rate remained

DISCUSSION AND CONCLUSION
The present study reported here the first record of D. hamatus in Cameroon, where the reported species found in both the locations viz.Yaounde in the southern plateau and Koutaba in the western highlands of Cameroon.Presently, this fruit fly species has been recorded from several African countries (Cogan and Munro, 1980;Thompson, 1998;White, 2006;White and Goodger, 2009) including Nigeria, a border country of Cameroon.
In natural conditions, and among seven species of cucurbits studied, the female of D. hamatus laid eggs only on male flowers of L. siceraria.In Cameroon, L. siceraria, with two varieties, is cultivated either for the gourd or for seeds.The bitter varieties is known in Sanskrit as Katutumbi and a sweet one, called Alābu (Chinyere et al., 2009).Globally, cucurbits differ qualitatively in production of cucurbitacin whose rate increases the bitterness of the plant (Gould, 1978).
D. hamatus maggots were observed in the infested male flowers of L. siceraria but not in fruits.In fact, oviposition behaviour of phytophagous insects is influenced either by the variation of preference (choice of oviposition or nutrition sites) or motivation (accessibility of food resource, shape and colour) (Singer et al., 1992).Moreover, under laboratory conditions, larvae were able to develop on the flowers and the fruits of all the cucurbits investigated in the present study.This result corroborates the hypothesis of Raghu et al. (2004) who stated that in Diptera, larvae remain and feed only on plant organs on which eggs were laid, suggesting a high specificity in the oviposition site selection by females.In our study, the plant organs of these cucurbits were all available in the plots, thus, preference may have predominated in host plant selection by the females of the species.Some species in the genus Lagenaria exhibits a strong relationship with certain species of fruit flies of the subgenus Dacus (Lophodacus) (White 2006).
However, the association observed between the fly D. hamatus and the plant L. siceraria is new, since hitherto, the only recorded host plant of this fly was Lagenaria sphaerica in Nigeria (White 2006).This discovery extends the host plants spectrum of the fly and argues in favour of the hypothesis that D. hamatus would be closely associated with male flowers of plants from the genus Lagenaria.Indeed, the ability of a phytophagous insect to expand its host plant range depends both of the digestive physiology of larvae and the adult plasticity in oviposition site selection (Fitt, 1986).
Even when flowers and fruits of the seven studied cucurbit species were available in the garden, no adult of D. hamatus emerged from incubated fruits, neither from those of L. siceraria.By contrast, our study showed other species of the genus Dacus which preferentially feed on fruits; their adults frequently emerged from infested fruits.White and Elson-Harris (2004) reported these Dacus species as fruits flies of economic importance in the Old World tropics.In the subfamily Dacinae, the relationship between host plants and flies is thought to be particularly strong, since the host plant is considered central for larval and adult feeding, mating and oviposition (Raghu et al., 2004).
Tephritids use both visual and chemical signals to locate and access habitat, adult food, oviposition sites and mating resources (Papadopoulos et al., 2006).The mean infestation rates of male flowers as well as the mean number of larvae per male flower were more important in Koutaba than in Yaounde.Furthermore, at each location, the mean rate of infested male flowers and the mean number of larvae per male flowers was significantly greater on the bitter variety than on the sweet one.This result suggests that the bitter variety is more suitable for the fly.Overall male flowers, those of L. siceraria and S. edule are white in colour while that of the five other species range from yellow to bright orange.In this context, the variation of the infestation rate within male flowers of L. siceraria varieties suggests the existence of a recognition factor (such as among those of cucurbitacin) used by the adults in the identification of oviposition sites selection.The efficiency of this factor may be linked to the relationship between the chemical composition of the plant and the chemoreceptors of the insects and may vary within the plant species and/or varieties.According to Raghu et al. (2004), the microclimate and the host plant architecture influence significantly the abundance and the behaviour of phytophagous insects.Infestation by depredators may vary according to agro-ecological conditions (southern plateau versus western highlands) and the plants organs (flowers versus fruits) which can probably emit several specific volatile organic compounds.Further investigations are to be done in order to identify volatile organic compounds involved in the association between L. siceraria and D. hamatus.

Figure 1 .
Figure 1.Maps of the studied locations in the southern part of Cameroon.
for parasitoids (Hymenopterans).Identifications of fruit fly were later checked and confirmed by Dr I. M. White at the Natural History Museum of London (NHML) and by the taxonomists of the Royal Museum for Central Africa (RMCA) at Tervuren (Belgium).The expertise of the taxonomist of the "Centre International de Recherche Agricole pour le Développement/Centre de Biologie et de Gestion des Populations" (CIRAD/CBGP) at Montpellier (France) was required for identification of parasitoids.Voucher specimens were deposited in the collection of the Laboratory of Zoology, Faculty of Sciences, University of Yaounde I (Cameroon) and in the collection of the above cited institutions.

Figure 3 .Figure 4 .
Figure 3. Mean infestation rate of male flowers of two varieties of Lagenaria siceraria by the larvae Dacus (Lophodacus) hamatus at two locations of Cameroon, from November 2010 to February 2011.Legend: bars with different letters are significantly different among sites and varieties at P < 0.05.

Table 1 .
Total number of male flowers produced by seven species of cucurbits and those infested by fruit flies (Diptera: Tephritidae) in two localities of southern Cameroon, from November 2010 to February 2011.

Table 2 .
Total number of fruits produced by seven species of cucurbits and those infested by fruit flies (Diptera: Tephritidae) in two localities of southern Cameroon from November 2010 to February 2011.