A Nationwide Survey of Tabanidae Fauna of Cameroon

Background: Tabanid ies are a neglected group of haematophagous dipterans, yet contains 4400 species, regrouped under 144 genera. They are mechanical vectors of important pathogens, including viruses, bacteria and protozoa of humans, domesticated and wild animals. As it is over 50 years since publication of a preliminary nationwide record of tabanids of Cameroon identied 84 species, updated information is provided. Objective: To provide current data on the species composition, abundance and distribution of Tabanidae of Cameroon in the 5 main agro-ecological zones (AEZs). Methods: From 2015-2017, a systematic entomological prospection using Nzi, Vavoua, Biconical and Sevi traps (N=106) was conducted in 604 trap-points over 11,448 traps days, in the 5 main AEZs of Cameroon. Results: A total of 25,280 tabanid specimens were collected, identifying 25 species with 8 not previously documented Cameroon, including: distribution maps the newly identied tabanids between AEZs, with most from the savanna. The highest Tabanidae apparent density (ADT) was recorded in the Sudan savanna region and the


Background
Tabanids (Diptera, Tabanidae) form one of the largest group of hematophagous ies with about 4400 species and 144 genera [1]. They are well known by the local names of three medical and veterinary important taxa notably horse ies (Tabanus), deer ies (Chrysops) and clegs (Haematopota). Tabanids in ict direct effects such as nuisance and painful bites leading to blood loss, stress and weight loss of host and indirectly transmit major pathogenic agents including bacteria, viruses, protozoa and helminths [2][3][4][5] mechanically [6,7]. The major blood-meal host of tabanids consist of humans, ruminants and wild animals [4,8]. The consequence of their blood feeding trait is the mechanical transmission of one of the most important cattle disease in Sub-Saharan Africa (SSA), African Animal Trypanosomiasis (AAT) [2,4,[9][10][11][12], in addition to the biological transmission of Loa loa lariasis [13,14]. The importance of studying the Tabanidae in relation to the mechanical transmission of pathogenic trypanosomes was emphasised at the 1948 African Conference on Tsetse and Trypanosomiasis in Brazzaville [15]. Despite the importance of tabanid ies, they are a neglected subject of research [4,16].
Tabanidae of central Africa and other Ethiopian regions are either misidenti ed or poorly known and this applies in Cameroon. Apart from preliminary studies that reported 62 [15] and 84 [17] species respectively, there appear to have been no updates on their checklist and distribution. A survey in the Central African Republic indicated that 54 to 64 species constituted the Tabanidae fauna of this country [18]. The updated Tabanidae checklist of Ivory Coast reported the occurrence of 70 species [19]. The Tabanidae of Algeria consisted of 49 species [20,21] as con rmed in a recent update [22]. The Tabanidae checklist of Kenya [23], the re-description of Tabanidae of East Africa (Uganda, Tanzania and Kenya) [24], and the tabanids reported in Gabon [25,12,26] all provide current information. However in Cameroon, the rst preliminary study only focused on the identi cation, aspects of the bionomics and vector role of horse ies in French Cameroon (forest area of the south region) [15]. Apart from the most elaborate preliminary countrywide report on the Tabanidae of Cameroon [17], other reports only presented a scanty record in the Sudan savanna [27,28]. The present study aims to update information on the diversity, abundance and distribution of the Tabanidae species in the main AEZs of Cameroon.

Description of collection sites
The study zones consisted of the ve main AEZs of Cameroon (Fig. 1). The geographical coordinates and climatic characteristics of the different AEZs are presented (Table 1).  [29,30] whereas the Sevi trap is a modi cation of the malaise (René Malaise) and Canopy traps [31]. It was designed by Sevidzem SL ( rst author) in Ngaoundere, tested by the MSEG team in the tsetse ies infested region of North Cameroon, and con rmed as a trap for tabanids [27]. A description of the Sevi trap is provided (Additional le 1). The same number of traps and types were not deployed in the sampled AEZs because of the differences in topography, limited trap number/types and personnel to monitor them. The geolocalization of trap-points was conducted using a Global Positioning System (GPS) handset (GPS eTrex®; Garmin (Europe) Ltd, Southampton, UK). The trapping effort, de ned as the number of traps multiplied by the trapping days (traps days), was recorded. The present prospection was conducted within 108 days using 106 traps, resulting in trapping effort of 11,448 traps days.

Sudan-savanna of Far North
Tabanidae prospection in this region was carried out in the following type-localities: Kalang, Kainide, Diddel tanne, Doulam and Yanga using 20 Nzi traps [32]. The traps were set in potential tabanids breeding points such as in marshy areas around livestock drinking points, in open grass savanna that represented livestock grazing spots and in gallery forests. Trapping was conducted three days consecutively per month throughout the study period. Traps were set and activated in the morning (8: 00am) and cages emptied in the evening (6:00pm). and Nzi (n = 6) were used. The traps were set in the morning (6:00am) and emptied every evening (6:00 pm) for three days consecutively per month throughout the study period.

Guinee-savanna of Adamawa plateau
The prospection type-localities in the Guinee savanna zone consisted of Velambai (lake Djalingo with open grass savanna), Mbidjoro (open grass savanna, forest-savanna mosaic) and Vina du Sud (river vina with gallery forest). The trapping points were in villages with intensive cattle breeding activities. As in the rain forest of the Sanaga, three different traps (Biconical (n = 6), Vavoua (n = 6) and Nzi (n = 6)) were used for this activity. The traps were set in the morning (7:00 am) and their cages emptied every evening (6:00 pm) for three days consecutively per month throughout the study period.

Mosaic-forest in the East
The survey in this zone was carried out using Nzi (n = 3) and Vavoua (n = 3) traps. Traps were set in three environments: gallery forest, overnight cattle corrals, and river banks (watering point for cattle). These three biotopes were identi ed in four type-localities: Minali, Oudou, Camp Général and Gabong of the Société de Développement et d'Exploitation des Productions Animales (SODEPA). The trapping points were all in the SODEPA ranch which is a public structure involved in intensive cattle production activities.
The trap cages were emptied after 24h for three days consecutively per month throughout the study period.

Highland plateau of North West
Five Biconical traps were set at livestock drinking points and grazing areas in the following type-localities: Bali top quarters, Saphery, Babah, Njinki, Tchaboutchou, Munam and Ntchuobo. Flies were collected from the traps each day before night fall (6: 00 pm) for three consecutive days per month throughout the study period.

Fly identi cation
All specimens were conserved in ethanol [35] and identi ed using a stereo microscope (Carl Zeiss™ STEMI 2000-C). The identi cation of tabanid ies was made following published morphological identi cation keys [36,37,38].

Sex determination
The sex of newly identi ed tabanid specimens was determined using published criteria [38], where females naturally possess a larger inter-ocular width than their male counterparts.

Determination of abundance
The abundance of trapped tabanids was translated as their apparent density (ADT), stated as the number of ies per trap and day (f/t/d) [12] as follows: Where: ADT: Apparent density,

Results
Distribution of tabanids in the prospected type-localities The present prospection from 2015 to 2017, occurred in 25 type-localities with 605 trap-points of the 5 major AEZs of Cameroon (Fig. 2). This resulted in the collection of 25,280 tabanid specimens, with 25 identi ed species regrouped under 5 genera (Tabanus, Chrysops, Haematopota, Ancala and Atylotus) that were members of 3 tribes, notably Tabanini, Chrysopini and Haematopotini. At the genus level, the 4 genera of Haematopota, Ancala, Chrysops, and Tabanus were highly encountered in the mosaicforest/rain-forest more than in the savanna, whereas Atylotus was only encountered in the savanna and never trapped in the forest. C. longicornis was the only Chrysops that occurred in forest and savanna collections, but the other Chrysops were either forest restricted (C. silacea, C. dimidiata, C. funebris) or savanna-restricted (C. dimidiata). H. decora and An. fasciata occurred in forest and savanna collections, whereas Atylotus agrestis was only restricted to type-localities in the Guinee savanna and Sudan savanna. Only Tabanus taeniola was caught in all the propected type-localities ( Table 2).
The abundance of tabanids with prospected type-localities The species of tabanids caught in the different type-localities recorded highest abundance in some sites as compared to others. Highest ADT occurred at Diddel Tanne in the Sudan savanna of the Far North region, with such dominance caused by Atylotus agrestis (136.25 f/t/d). Lowest ADT was recorded in most type-localities in the plateau highland of the Northwest region (Table 3). However, the mean ADTs of tabanid species with type-localities was statistically signi cantly different (Student t-test = 2.519, df = 24, P = 0.019).

Species diversity in the sampled type-localities
The type-locality with highest species diversity was the palm oil plantation (13 species) site located in Mouanko around the Douala-Edea Zone (DEZ) in the rain forest. Then followed the Vina du Sud typelocality (12 species), a riverine-gallery forest-savanna mosaic. The least species number (1 species) identi ed was in collections from Kainide and Kalang type-localities (savanna of Far North region) and in Bali Top Quarters (BTQ), Saphery, Babah, Njinki, Tchaboutchou, Munam and Ntchuobo in the plateau highland of the North West region (Fig. 3).
H. inornata Austen, 1908 Its type-locality was Wum, but recent prospections revealed their presence in the Guinean-savanna of Cameroon [17].
H. rufula Surcouf, 1909 This species was in Cameroon for the rst time by Oldroyd [48] in Galim of the Adamawa in the Guinee-sahelian region [17].
H. dukei Ovazza and Mouchet, 1967 It was caught in mount Bamboutous and 2 species sorted in Dschang at an altitude of 1600m by Ovazza et al. [17].
H. crevei Oldroyd, 1952 It was collected in Mamfe in the rain-forest of the South west region of Cameroon [17].
H. barombi Oldroyd, 1952 This species is frequent in Cameroon. It Is found in forest and forest relics of mountainous regions and in transition zones, typical of Kumba [17]. T(t) obscurefumatus Surcouf, 1909 This is a rare species and was found in forest con nement and in transition zones like along the Kumba road and Ebolowa [17].
T(t) su s Jeannicke, 1867 It was found in Logone and Chari [17]. It was recently identi ed in the Sora Mboum Sudansavanna of North Cameroon by Lendzele et al. [27].

T(t) pertinens Austen, 1912
This species was identi ed in Garoua in the Sudan-savanna of the North region of Cameroon [17].

T(t) leucostostomus Loew, 1858
This species was identi ed in Maroua in the Sudan-savanna of Far North region of Cameroon [17].

T(t) gratus Loew, 1858
It was found in Garoua and Logone and Chari [17]. It was recently caught in the savannas of Ngaoundere and North Cameroon [27].

T(t) triquetonatus Carter, 1915
This species was caught along river Nyong and seemed to be abundant in Cameroon [17].
Sub family Tribe Genus Species Type-localities T(t) argenteus Surcouf, 1907 This species was found in Kumbo, Edea road, Bertoua [17]. It was Recently found in the Douala-Edea Game Reserve.

T(t) canus Karsch, 1879
This species was caught in Yaounde, Lekie and was noticed in the evening with light. It was caught in Maroua and usually found in forest and post forest areas [17].

T(t) fulvicapillus Carter, 1912
This species was reported by Oldroyd [37] in Cameroon, but its locality was indicated in Congo Brazzaville.

T(t) pluto Walker, 1848
This species was caught in Mamfe in the humid forest in the South west region of Cameroon [17].
T(t) anaeus Surcouf, 1907 This species was described around Kumba in the humid forest of the South West region [17]. This species was signaled in all forest regions [17].
Recently caught in Minali type-locality of East region.
T(t) scholae Oldroyd, 1954 The series-type of this species was from Kumba in the humid forest of South west [17].

T(t) par Walker, 1854
This species was identi ed in Mamfe, Maroua and Logone and Chari. This species was savanna-like and penetrated the forest. It was identi ed in the Savanna of north Cameroon (Sora Mboum) by Lendzele et al. [27].

*T(t) latipes Macquart, 1838
This species was recently caught at the palm oil plantation and around the abattoir in the Sanaga maritime zone.

T(t) zoulouensis Ricardo, 1908
This species was found in Bamoum in the west highland of West Cameroon [17].

T(t) thoracinus Palisot de Beauvois, 1807
This species was found in Maroua in the Sudansavanna of the Far North region of Cameroon [17]. Surcouf, 1907 This species was caught in the Upper Nyong in the rain forest of Cameroon [17].
T. oldroydi Grenier and Rageau, 1955 This species was collected from Lekie and Mefou [17]. This species was collected along the river Sanaga area and livestock areas of Kumba [17].
The recent surveys (2015-2017) led to the identi cation of 8 species that were not present in the list of tabanids of the preliminary survey and increased the list of tabanids of Cameroon to 92. The systematics and descriptions of the type-localities as well as the distribution maps of the 8 recently identi ed species are as follows: 1) Tabaninae Latreille, 1802   Tabanini Latreille, 1802 Tabanus Linnaeus, 1758 Tabanus Latipes Macquart, 1838 (1 ) Type-localities: Only one female of this species was caught using the Nzi trap set in the marshy land of the abattoir (3°83333' N and 10°05' E) in Mouanko in the rain forest of the DEZ of Cameroon (Fig. 4a).

Discussion
These surveys, conducted from 2015 to 2017, led to the identi cation of 25 species, much less than the 84 species recorded in the preliminary study of 1970 [17]. Coast [19], Algeria [20][21][22], Kenya [23,24], Zambia [11], South Africa [11], Nigeria [40], Uganda and Tanzania [24], Congo [41], Liberia [42], and Gabon [12,25,26]. The higher number of species in the preliminary survey compared to this present study, could be due to several factors, including: the limited number of sampled sites (n = 25) in the different AEZs in the current study compared to several (> 25) type-localities in the preliminary study; the preliminary survey used sweep nets whereas the present study used tsetse traps; and ecological changes in uencing y distribution and densities.
The higher Tabanidae list (92 species) of Cameroon compared to that of the other countries of the Afrogeographical region likely re ects the presence of more diverse AEZs contributing to the successful development and survival of several species of tabanids. The newly identi ed tabanid species of Cameroon all existed in the list of Tabanidae from neighboring countries of the Central and West African sub-regions [11,12,19,25,40]. A review on the genus Chrysops of Africa with focus on Cameroon [14] mentioned only the occurrence of C. silacea and C. dimidiata, whereas our survey reveals 5 species, notably C. distinctipennis, C. longicornis, C. funebris, C. silacea and C. dimidiata. The possible reason for the previous scanty record of the species of this genus [14] could be because the earlier entomological surveys were limited to the forest regions of Cameroon, where C. silacea and C. dimidiata are omnipresent. Interestingly, the 1955 report on the presence of 5 species of Chrysops [15], notably C. dimidiata van der Wulp, C. Silacea Austen, C. funebris Austen, C. longicornis Macquart and C. stigmaticalis Loew was conducted in the same study areas as those in the review on the genus Chrysops [14]. However, neither of these reports identi ed C. distinctipennis, a species recorded in the Guinean savanna ecozones of Cameroon in our current study.
Tabanus taeniola was the most abundant species in the forest and savanna and this likely re ects the environmental conditions favorable to this species, one reported to be most frequent in the collections of other authors in the Afro-geographical zone [9,12,27,43,44]. Further, highest tabanid mean apparent density was recorded in the Sudan savanna type-localities, a nding similar to that presented in the 2020 report of the MSEG [28,45] indicating the apparent absence of glossines in this region and high apparent densities of tabanids. This highest abundance of tabanids in sites of Far North region of the country is likely due to this being a major cattle rearing region of Cameroon [46,47] and containing the Waza Park that harbours diverse and high numbers of domestic and wild animals serving as blood meal hosts for this y group. The presence of cattle and wild animal hosts, plus the conducive environmental conditions of this region, presumably favours the development and survival of tabanids.
Highest tabanid species diversity was recorded in the rain forest of the DEZ than in other AEZs. The rain forest equally recorded the highest number of species (n = 13) than any other sampled AEZ. The likely reason could be the favourable environmental conditions of this AEZ for several species of tabanids and because it was the least anthropized of the sampling sites, with wild animals and N'Dama cattle serving as blood meal hosts and providing a suitable environment for tabanid development and survival [25]. The rain forest sites consisted of tall palm trees equally interspaced to provide high luminosity, canopy trees for shelter, rivers and abundant marshy land for breeding, with the presence of palm trees has been reported to harbor high numbers of different species of tabanids [17]. The 1955 study in forest areas of South Cameroon indicated the abundance of breeding sites for female tabanids that favored their proliferation in this zone [15].

Conclusions
The present study registered 25 Identi cation of these novel species increased the Tabanidae number of the country from 84 to 92. The newly identi ed species were mostly found in the Guinean savanna. The highest Tabanidae apparent density was recorded in the Sudan savanna region whereas the highest species diversity was noticed in the rain forest of the DEZ. The high diversity and abundance of Tabanidae in the livestock/wildlife interface areas of the rain forest and Sudan savanna AEZs respectively strongly suggests the risk for the mechanical transmission of dangerous pathogens. Future investigations on this group to studying the microbiota they harbour are proposed, to establish their potential epizootiological role in the transmission of diseases in the different AEZs.
Declarations Figure 1 Map showing the different agro-ecological zones of Cameroon. Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors.  Distribution map of T. latipes (a), T. ricardae (b). Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors.  Distribution map of C. distinctipennis (a) and A. fasciata (b). Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.