RESEARCH ARTICLE Esperanza Window Traps for the collection of anthropophilic blackflies (Diptera: Simuliidae) in Uganda and Tanzania Adam Hendy1*, Vincent Sluydts2, Taylor Tushar3, Jacobus De Witte1, Patrick Odonga4, Denis Loum4, Michael Nyaraga4, Thomson Lakwo4, Jean-Claude Dujardin1, Rory Post3,5, Akili Kalinga6, Richard Echodu7 1 Department of Biomedical Sciences, Institute of Tropical Medicine, Antwerp, Belgium, 2 Evolutionary Ecology Group, Department of Biology, University of Antwerp, Wilrijk, Belgium, 3 Department of Disease Control, London School of Hygiene & Tropical Medicine, London, United Kingdom, 4 Vector Control Division, Ministry of Health, Kampala, Uganda, 5 School of Natural Sciences and Psychology, Liverpool John Moores University, Liverpool, United Kingdom, 6 National Institute for Medical Research, Tukuyu Research Centre, Tukuyu, Tanzania, 7 Faculty of Science, Gulu University, Gulu, Uganda * adam_hendy@hotmail.co.uk Abstract There is an increasing need to evaluate the impact of chemotherapeutic and vector-based interventions as onchocerciasis affected countries work towards eliminating the disease. The Esperanza Window Trap (EWT) provides a possible alternative to human landing col- lections (HLCs) for the collection of anthropophilic blackflies, yet it is not known whether cur- rent designs will prove effective for onchocerciasis vectors throughout sub-Saharan Africa. EWTs were deployed for 41 days in northern Uganda and south eastern Tanzania where dif- ferent Simulium damnosum sibling species are responsible for disease transmission. The relative efficacy of EWTs and HLCs was compared, and responses of host-seeking black- flies to odour baits, colours, and yeast-produced CO2 were investigated. Blue EWTs baited with CO2 and worn socks collected 42.3% (2,393) of the total S. damnosum s.l. catch in northern Uganda. Numbers were comparable with those collected by HLCs (32.1%, 1,817), and higher than those collected on traps baited with CO2 and BG-Lure (25.6%, 1,446), a synthetic human attractant. Traps performed less well for the collection of S. damnosum s.l. in Tanzania where HLCs (72.5%, 2,432) consistently outperformed both blue (16.8%, 563) and black (10.7%, 360) traps baited with CO2 and worn socks. HLCs (72.3%, 361) also out- performed sock-baited (6.4%, 32) and BG-Lure-baited (21.2%, 106) traps for the collection of anthropophilic Simulium bovis in northern Uganda. Contrasting blackfly distributions were observed on traps in Uganda and Tanzania, indicating differences in behaviour in each area. The success of EWT collections of S. damnosum s.l. in northern Uganda was not repli- cated in Tanzania, or for the collection of anthropophilic S. bovis. Further research to improve the understanding of behavioural responses of vector sibling species to traps and their attractants should be encouraged. PLOS Neglected Tropical Diseases | https://doi.org/10.1371/journal.pntd.0005688 June 19, 2017 1 / 22 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 OPENACCESS Citation: Hendy A, Sluydts V, Tushar T, De Witte J, Odonga P, Loum D, et al. (2017) Esperanza Window Traps for the collection of anthropophilic blackflies (Diptera: Simuliidae) in Uganda and Tanzania. PLoS Negl Trop Dis 11(6): e0005688. https://doi.org/10.1371/journal.pntd.0005688 Editor: Peter U. Fischer, Washington University School of Medicine, UNITED STATES Received: April 8, 2017 Accepted: June 7, 2017 Published: June 19, 2017 Copyright: © 2017 Hendy et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All relevant data are within the paper and its Supporting Information files. Funding: Financial support was provided through the Department of Economy, Science and Innovation (EWI) of the Flemish government through a "Structural Research Funding" (SOFI) grant of the Institute of Tropical Medicine, Antwerp, and the Flemish Interuniversity Council South Initiative (VLIR-UOS). The funders had no role in https://doi.org/10.1371/journal.pntd.0005688 http://crossmark.crossref.org/dialog/?doi=10.1371/journal.pntd.0005688&domain=pdf&date_stamp=2017-06-29 http://crossmark.crossref.org/dialog/?doi=10.1371/journal.pntd.0005688&domain=pdf&date_stamp=2017-06-29 http://crossmark.crossref.org/dialog/?doi=10.1371/journal.pntd.0005688&domain=pdf&date_stamp=2017-06-29 http://crossmark.crossref.org/dialog/?doi=10.1371/journal.pntd.0005688&domain=pdf&date_stamp=2017-06-29 http://crossmark.crossref.org/dialog/?doi=10.1371/journal.pntd.0005688&domain=pdf&date_stamp=2017-06-29 http://crossmark.crossref.org/dialog/?doi=10.1371/journal.pntd.0005688&domain=pdf&date_stamp=2017-06-29 https://doi.org/10.1371/journal.pntd.0005688 http://creativecommons.org/licenses/by/4.0/ Author summary Using human bait to collect blood-feeding insects is an ethically sensitive issue. Whereas researchers investigating insect-borne diseases such as sleeping sickness, leishmaniasis and malaria have a range of traps at their disposal, those investigating blackflies and river blindness (onchocerciasis) still rely on this method. Alternatives to human bait are needed to monitor disease transmission as onchocerciasis control programmes approach their elimination phase. The recently developed Esperanza Window Trap provides one such possibility. We built these traps based on previously published methods while conducting blackfly research in Uganda and Tanzania in order to evaluate their efficacy and ease of use. Our results show that in Uganda the traps worked well for the collection of Simulium damnosum, the blackfly primarily responsible for onchocerciasis transmission in sub- Saharan Africa, but were less effective at collecting the same species in Tanzania. Blackfly behaviour and response to traps will probably vary from one country to another. Esper- anza Window Traps show promise for blackfly collections, but further research and devel- opment are needed to determine how broadly they can be used. Introduction In 1966, the World Health Organization (WHO) acknowledged a need to develop new sam- pling techniques to replace human landing collections (HLCs) for the collection of blackfly (Diptera: Simuliidae) species involved in the transmission of Onchocerca volvulus, the parasitic filarial nematode responsible for human onchocerciasis [1]. Despite a comprehensive review of adult blackfly collection methods by Service in 1977 [2], subsequent research efforts to meet the needs outlined by the WHO have been limited [3–9]. The primary concern is for the devel- opment of a trap to replace HLCs to monitor progress towards onchocerciasis elimination, but an effective trap might also be deployed as a control mechanism in itself to reduce vector popu- lations in support of mass drug administration. The recent development of the Esperanza Window Trap (EWT), used successfully for the collection of host-seeking anthropophilic blackflies in Mexico and Burkina Faso, has provided the possibility of one such viable method [7, 10–13]. Control and surveillance Following the implementation of the Mectizan (ivermectin) Donation Program in 1987, meth- ods of onchocerciasis control switched from vector-based interventions to mass drug adminis- tration through community directed treatment with ivermectin (CDTI) [14]. Whereas it has been established that ivermectin treatment can eliminate the disease in certain endemic foci, the conditions under which CDTI alone is effective have not been fully explored [15–17]. It is therefore essential that methods for monitoring entomological and parasitological indices of onchocerciasis transmission are available in intervention and post-intervention settings as countries work towards elimination [18, 19]. For EWTs to be effective in evaluating the impact of chemotherapeutic and vector-based programmes, they should collect appropriate numbers of the same vector populations as those biting humans. They should also collect vectors with the same age structure (parity rates) as those biting humans, or collect them in a condition that enables age structures to be calibrated. The current WHO guidelines for entomological evaluation of O. volvulus transmission in CDTI settings require that HLCs are used for the collection of anthropophilic blackflies [20, Esperanza Window Traps for the collection of anthropophilic blackflies PLOS Neglected Tropical Diseases | https://doi.org/10.1371/journal.pntd.0005688 June 19, 2017 2 / 22 study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing interests: The authors have declared that no competing interests exist. https://doi.org/10.1371/journal.pntd.0005688 21]. The method is robust, sensitive, and well accepted by communities, and is therefore pref- erable to more invasive methods of O. volvulus surveillance such as Ov-16 serology testing in children [21]. However, human participants collecting biting flies are potentially exposed to a range of vector-borne pathogens, although with appropriate training, the risk is generally con- sidered no higher than for others living in disease endemic areas. Despite this, obtaining the necessary ethical approval can often delay the implementation of research and surveillance programmes [22]. Available traps Attempts to develop new, or to utilise or modify existing traps for the collection of host-seek- ing, anthropophilic blackflies, have been met with mixed or limited success [2]. Light traps [3, 4], sticky traps and silhouettes [23–26], BG-Sentinel traps [7], modified Challier-Lavessiere tsetse traps [5, 6], and other novel traps [27] have been successfully used to collect blackflies in various physiological states, yet repeating collections using these methods has sometimes proved difficult [8, 9]. Visual attraction Early investigations into the response of blackflies to long-range visual and olfactory stimuli, including colour, shape, and CO2, were mainly confined to Nearctic species including Simu- lium venustum and Simulium vittatum [28–32]. Several studies indicate that host-seeking blackflies generally prefer to land on darker colours and matt surfaces [30, 31, 33], and it is also thought low UV reflectance and strong contrast of traps against their background is important in attraction [28, 32, 34]. Comparatively little research has been dedicated to similar investigations for Simulium damnosum sensu lato (s.l.), the principal vector of O. volvulus in Africa. The limited data that exists is consistent with colour-choice experiments for other blackflies, in that host-seeking S. damnosum s.l. appear to be attracted to dark colours [5, 24, 25, 35]. However, results of behavioural studies should be interpreted cautiously, and Walsh (1980) stresses that they should not be generalised for species other than those being investi- gated [25, 28]. This is likely to be especially relevant when studying S. damnosum s.l., a com- plex of sibling species composed of at least 55 morphologically indistinguishable cytospecies and cytoforms of unknown taxonomic status, each with unique ecological and behavioural traits [36, 37]. Olfaction Simulium damnosum s.l., like other haematophagous Diptera, are attracted to CO2 and host odours [38, 39]. CO2 is a powerful mediator of host-seeking behaviour which can greatly enhance blackfly collections [23, 24], yet the biological mechanisms of blackfly attraction to olfactory and visual stimuli are poorly understood [38]. Following experiments in a Cameroo- nian rainforest, Thompson (1976) demonstrated that the presence of ‘exhaled breath’, indus- trial CO2, and worn clothing, improved trap collections [24, 40]. He concluded that chemicals present in human sweat are likely to be important in attracting S. damnosum s.l. [40], and that visual and olfactory cues are of greatest importance in attracting savannah and forest sibling species respectively [24]. More recently, EWTs and BG-Sentinel traps baited with worn shirts, trousers (pants) and synthetic chemicals (BG-Lure and octenol) have been shown to be more effective in attracting blackflies than unbaited traps [7]. Young et al. (2015) have since used gas chromatography and electroantennography to identify chemicals present in human sweat which are potentially attractive to S. damnosum s.l. in Burkina Faso and Simulium ochraceum s.l. in Mexico [13]. They then demonstrated that EWTs baited with candidate compounds Esperanza Window Traps for the collection of anthropophilic blackflies PLOS Neglected Tropical Diseases | https://doi.org/10.1371/journal.pntd.0005688 June 19, 2017 3 / 22 https://doi.org/10.1371/journal.pntd.0005688 collected 2–3 times the number of these species in the field compared to traps baited with CO2 alone, although the authors acknowledge that catch numbers were low and that further research is needed [13]. Esperanza Window Traps In 2013, Rodriguez-Pérez et al. published results of the development and trial of the EWT in Mexico, which involved investigating the attractiveness of coloured fabrics, CO2 sources, and host odours to S. ochraceum [7]. EWTs constructed using blue fabric outperformed those made with red, yellow and black fabrics when baited with either industrial CO2 released at 150-200mL/min, or CO2 produced by mixing sugar, yeast (Saccharomyces cerevisiae) and water (quantities not specified). There was no statistically significant difference in the number of blackflies collected on traps regardless of the CO2 source. With the addition of host odours in the form of a worn shirt or BG-Lure, CO2-baited blue EWTs approached the attractiveness of HLCs in one of two trials. In the second trial, the baited EWT was only half as effective as the HLC [7]. Toé et al. (2014) further developed the EWT in Burkina Faso for the collection of Simulium damnosum sensu stricto (s.str.) and Simulium sirbanum, but used black traps baited with BG-Lure and yeast-produced CO2 as the basic design [11]. EWTs of differing heights were first compared. ‘Short’ traps, standing within 15cm of the ground were more effective than ‘tall’ traps, although the difference was only statistically significant at one of two sites investi- gated. The addition of a vertical blue stripe to the black background further enhanced collec- tions, but again, this was only statistically significant at one of the two sites. Short, striped EWTs baited with CO2 and BG-Lure caught similar numbers of S. damnosum s.l. as those baited with CO2 and worn trousers. In a final experiment, EWTs baited with CO2 and worn trousers collected numbers comparable with HLCs, whereas those baited with worn trousers alone collected numbers similar to unbaited traps. The authors also reported the collection of Simulium adersi and Simulium schoutedeni from the traps, and questioned the importance of fermentation products other than carbon dioxide in the attraction of vector flies [11]. Rationale and objectives The various sibling species of the S. damnosum complex are behaviourally and ecologically unique in traits such as breeding habitats, dispersal capabilities, degree of anthropophily, and their capacity to transmit disease [37]. It is not yet known whether different sibling species will respond differently to EWTs, and whether current trap designs will prove to be effective for S. damnosum s.l. collections throughout onchocerciasis affected areas of sub-Saharan Africa. This study therefore aimed to compare the relative efficacy of EWTs with HLCs for the collec- tion of anthropophilic blackflies in onchocerciasis transmission zones of Uganda and Tanza- nia, where different sibling species of the S. damnosum complex are responsible for disease transmission. Responses of host-seeking blackflies to odour baits, colour schemes, and yeast- produced CO2 were also investigated. Materials and methods Study area Experimental work took place for a total of 41 days at five locations in Uganda (26 days), and one in Tanzania (15 days), between 28 June 2015 and 19 September 2016 (Table 1). Collections were made in the districts of Lamwo, Moyo and Nwoya in the Madi/Mid-North onchocerciasis transmission zones of northern Uganda. Savannah grassland predominates Esperanza Window Traps for the collection of anthropophilic blackflies PLOS Neglected Tropical Diseases | https://doi.org/10.1371/journal.pntd.0005688 June 19, 2017 4 / 22 https://doi.org/10.1371/journal.pntd.0005688 and S. damnosum s.str. is thought to be the principal vector of O. volvulus [41, 42]. Small num- bers of S. sirbanum also breed along the Pager River northeast of Kitgum [43]. In addition, a member of the Simulium bovis species-group also forms a significant proportion of the anthro- pophilic blackfly population in the Mid-North [44]. Both S. damnosum s.l. and S. bovis occupy similar breeding habitats [45, 46]. In Lamwo district, these are mainly along the larger rivers including the Achwa (Aswa) and Pager [47, 48]. In Moyo, there is thought to be little local breeding of S. damnosum s.l., and it is likely that biting blackflies migrate from a series of rap- ids along the Nile in neighbouring South Sudan [43, 49]. The Murchison Nile forms the south- ern boundary of Nwoya district and is a major source of blackfly breeding [49]. There are historical reports of S. damnosum s.l. breeding along the Ayago River, a tributary of the Nile, and the Kibaa and Murchison River tributaries have also been cited as possible sources of infestation [49, 50]. Rainfall lasts from April to November, with peaks occurring early and late in the rainy season. The climate is hot and dry from December to March [51]. Collections in Tanzania were made at Chikuti on the north side of the Mahenge Mountains in the Mahenge onchocerciasis transmission zone of Ulanga district. The area is characterised by Precambrian limestone, and the presence of riverine, dry lowland and submontane forests [52]. The mountains are drained by numerous stony streams and rivers that are favourable to blackfly breeding [53]. Again, the principal vector of onchocerciasis is S. damnosum s.l. [35]. The cytoforms present in Mahenge are ‘Nkusi’, Simulium plumbeum (= ‘Hammerkopi’ and ‘Ketaketa’), ‘Sebwe’ and ‘Turiani’ [35, 54, 55]. ‘Nkusi’ is thought to be the predominant anthro- pophilic species, and S. plumbeum may have a limited role in human biting. Both ‘Sebwe’ and ‘Turiani’ are zoophilic [35, 54]. Simulium nyasalandicum (originally reported as S. woodi) also contributes to biting in small numbers, mainly in the south of the transmission zone [35, 56]. Rainfall lasts from November to May, and peaks between March and May. The dry season lasts from June to October [35, 52]. Basic trap design Traps were constructed using locally-sourced materials. Frames were composed of a light- gauge steel and trap faces measured approximately 1m2 (Fig 1). Traps stood on 0.25m sharp- ened legs which were easily pushed into the ground. The basic design included a blue tarpaulin screen that was hung tightly inside the frame. Blue was chosen as the base-colour as blue traps yielded the greatest number of blackflies during collections by Rodriguez-Pérez et al. (2013) in Mexico [7]. A black central stripe ⅓ the width of the blue screen was painted onto the trap using a matt black emulsion (Sadolin Paints (U) Limited, Uganda) during initial experiments in Uganda in 2015. The paint was allowed to dry for two days before traps were deployed. Dur- ing subsequent collections in Tanzania and Uganda (2016), the black paint was replaced with Table 1. Blackfly collection locations and distance from nearest known breeding sites. Alt. = Altitude, Dist. = Distance. Country District Location Coordinates Alt. Date Nearest Known Breeding Sites Dist. Uganda Lamwo Apyeta Bridge N 03˚18.005’ E 032˚21.705’ 691m Jul 2015 Achwa River 0km Beyogoya N 03˚17.648’ E 032˚29.708’ 845m Jul 2015 Achwa River 7.5km Moyo Gwere Luzira N 03˚39.827’ E 031˚48.056’ 980m Jul 2015 Nile (S. Sudan) 16km Pamulu N 03˚40.758’ E 031˚49.452’ 1066m Jul 2015 Nile (S. Sudan) 13km Nwoya Ayago Bridge N 02˚25.907’ E 032˚0.452’ 897m Jun 2015 Ayago River 11km N 02˚25.907’ E 032˚0.452’ 897m Aug 2015 Ayago River 11km N 02˚25.974’ E 032˚0.454’ 898m Sep 2016 Ayago River 11km Tanzania Ulanga Chikuti S 08˚36.175’ E 036˚44.072’ 459m Jun 2016 Mbalu River 5km https://doi.org/10.1371/journal.pntd.0005688.t001 Esperanza Window Traps for the collection of anthropophilic blackflies PLOS Neglected Tropical Diseases | https://doi.org/10.1371/journal.pntd.0005688 June 19, 2017 5 / 22 https://doi.org/10.1371/journal.pntd.0005688.t001 https://doi.org/10.1371/journal.pntd.0005688 black tarpaulin which was sewn together with the blue tarpaulin to form the screen. A CO2 outlet and host odour attractants were attached to the top corners of the EWT frame (Fig 1). Traps were covered with a black plastic sheet when not in use. Adhesives Tangle-Trap insect trap coating paste (Contech, Victoria, BC, Canada) was used to coat EWTs in Uganda. It was not possible to acquire the same product for trapping work in Tanzania due to manufacturing problems. EWTs in Tanzania were therefore coated with Temmen-Insekten- leim (Temmen GmbH, Hattersheim, Germany). Both products were thinned using�150mL locally purchased white spirit (Sadolin Paints (U) Limited, Uganda), before being applied to traps at least 24h prior to their deployment. CO2 production A sugar-yeast based source of carbon dioxide was produced in the field following methods out- lined by Smallegange et al. (2010) [57]. However, quantities of ingredients were adjusted to provide sufficient CO2 output (>80mL/min for at least 11 hours) following incubation at 30˚C during preliminary laboratory experiments (S1 Fig). Dry baker’s yeast (50g), sugar (500g) and water (2.5L) were mixed in 10L (Uganda) or 12L (Tanzania) containers immediately prior to blackfly collections commencing. PVC tubing extended from a hole in the container to an out- let at a top corner of the EWT. Containers were briefly shaken before being placed next to traps. Fresh sugar-yeast mixtures were prepared each day by community members assisting with blackfly collections. Host odour attractants Traps were either baited with host odours emanating from a pair of worn socks, or BG-Lure (Biogents AG, Regensburg, Germany), a synthetic mosquito attractant containing chemicals found on human skin (ammonia, lactic acid, and caproic acid) [58]. Worn socks were Fig 1. Blue and black trap designs showing position of CO2 and odour baits. Blue screens with a black vertical stripe (basic design) were used for all trapping experiments in Uganda. Black screens with a blue vertical stripe were additionally used in Tanzania. https://doi.org/10.1371/journal.pntd.0005688.g001 Esperanza Window Traps for the collection of anthropophilic blackflies PLOS Neglected Tropical Diseases | https://doi.org/10.1371/journal.pntd.0005688 June 19, 2017 6 / 22 https://doi.org/10.1371/journal.pntd.0005688.g001 https://doi.org/10.1371/journal.pntd.0005688 provided by villagers in exchange for a new pair of socks, and were tied to the top corner of the EWT opposite the CO2 outlet and replaced every three days. Worn socks have been shown to be effective for up to 8 days for the collection of mosquitoes [59]. Human landing collections HLCs were made by trained community-based participants following standard methods [20]. A team of two people worked alternate hours between 07:00 and 18:00, collecting blackflies landing on their exposed legs. Flies were collected in individual tubes and hourly catches were recorded. Specimen preservation and identification Blackflies were removed from EWTs using forceps after applying a drop of white spirit to spec- imens in order to partially dissolve the adhesive. A 10x magnification hand lens was used to verify identification of insects where necessary. All blackflies were preserved in >95% ethanol and were identified in the laboratory using morphological keys in Freeman & De Meillon (1953) [60]. The member of the S. bovis species-group present in northern Uganda was identi- fied based on the morphology of male pupae collected at Apyeta Bridge in 2015. To confirm identification, specimens were compared with reference material at the Natural History Museum, London, UK. The identity of adult S. bovis group flies collected on traps and by HLC was inferred based on the pupal identifications. Biting flies other than blackflies were removed from traps and preserved during collections made in 2016 only. Study design Odour baits. Blackfly collections were made for 21 days at five locations in Lamwo, Moyo and Nwoya districts of northern Uganda between June and August 2015, to compare the effi- cacy of EWTs (basic design) baited using CO2 and either worn socks or BG-Lure, with HLCs. At each location, precise vector collection sites were identified with the assistance of commu- nity members according to where blackfly biting was already known. A day was spent training participants in HLC methods and also to prepare CO2 mixtures for baiting traps. Three collec- tion sites were selected at each location for the deployment of 1) a team of two people to make HLCs, 2) two EWTs baited with CO2 and BG-Lure (EWT BG-Lure), and 3) two EWTs baited with CO2 and worn socks (EWT Socks). EWTs were placed in pairs, at right-angles to one another, in an attempt to maximise their visibility. HLC and EWT collections were made simultaneously between 07:00 and 18:00 for a minimum of three days (or in multiples of three days) at each location. Collection sites were at least 30m apart and HLCs and EWTs were rotated daily in a 3x3 randomised Latin square design in order to minimise interference and collection site bias respectively. Blackflies were removed from EWTs each day at approxi- mately 11:00, 14:00 and 17:00 to minimise the impact of desiccation on specimen quality. Daily blackfly catches were compared for each method. Colour schemes. Blackfly collections were made for 15 days at a single location near Chi- kuti village on the northern side of the Mahenge Mountains in Tanzania in June 2016, to com- pare the efficacy of EWTs of different colour schemes, with HLCs. Three collection sites were selected in a cultivated field approximately 0.5km from the village centre. Collection methods included 1) a team of two people to make HLCs, 2) two blue EWTs with a black central stripe (EWT Blue), and 3) two black EWTs with a blue central stripe (EWT Black). The EWT Black was similar to the design previously used by Toé et al. (2014) in Burkina Faso [11]. Each EWT was baited with CO2 and worn socks as previously described. Again, EWTs were placed in pairs, at right-angles to one another. HLC and EWT collections were made simultaneously Esperanza Window Traps for the collection of anthropophilic blackflies PLOS Neglected Tropical Diseases | https://doi.org/10.1371/journal.pntd.0005688 June 19, 2017 7 / 22 https://doi.org/10.1371/journal.pntd.0005688 between 07:00 and 18:00 each day and blackflies were removed from EWTs at approximately 10:00 and 17:00. Collection sites were at least 50m apart and HLCs and EWTs were rotated daily in a 3x3 randomised Latin square design. Daily blackfly catches were compared for each method. Yeast-produced CO2. Blackfly collections were made for 5 days at Ayago Bridge in Uganda in September 2016, to compare the efficacy of EWTs (basic design) baited with either a freshly prepared sugar-yeast mixture (EWT CO2+), or a mixture that had been prepared 5 days in advance and was no longer producing CO2 (EWT CO2-). No other odour baits were used in this experiment. Provisional laboratory observations demonstrated that CO2 produc- tion was<80mL/min after exposing sugar-yeast mixtures to continuous temperatures of 25˚C, 30˚C and 35˚C for 12h (S1 Fig). The amount of gas produced after 5 days would therefore be negligible. Two collection sites were prepared approximately 50m apart by clearing vegetation adjacent to the Ayago River. One trap was placed at each site and collections were made between 07:00 and 18:00 each day. Blackflies were removed at approximately 11:00, 14:00 and 17:00 each day and traps were rotated daily as in previous experiments. Daily blackfly catches were compared for each method. Blackfly distribution. In response to observations that S. damnosum s.l. were attracted to the lower parts of EWTs during odour bait experiments in Uganda in 2015, attempts were made to quantify blackfly distribution on traps during subsequent colour and CO2 experi- ments in Uganda and Tanzania in 2016. Small holes were made in EWT screens to divide the surface into nine approximately equal squares. The number of blackflies removed daily from each square was recorded for each trap type. Counts from corresponding squares on each side of the trap were combined. Blackflies were preserved daily according to trap type, rather than for each square. Reported blackfly counts on each square are therefore for all blackfly species and not individual species. Statistical analysis In all experiments, blackfly count was the response variable and was modelled as a function of trap type, the main covariate of interest. Location, collection site and rainfall were included as additional covariates. A generalized linear framework with a negative binomial distribution was used to take into account the overdispersion observed in the count data. The Akaike Infor- mation Criterion was used to select the most appropriate model for each data set, and models were verified by means of diagnostic plots. When more than one anthropophilic blackfly spe- cies was active at a study location, data for each species were analysed separately. Data were excluded from analysis for a particular species if blackfly collections were low (<5/day using all methods), or if the species was absent. The negative binomial model was also used to ana- lyse the distribution of blackflies on traps, and to investigate interactions between blackfly attachment on columns and rows. Heat maps of blackfly attachment to traps were produced using log transformed data to improve graphical representation of blackfly distribution. Analy- ses were performed within the R version 3.3.2 statistical computing environment [61]. Ethics statement Blackfly collections involving human participants were subject to review and approval by the institutional review board at the Institute of Tropical Medicine, Antwerp, Belgium (960/14, 1089/16); the Higher Degrees, Research and Ethics Committee, Makerere University School of Public Health, Kampala, Uganda (2014/244); and the Medical Research Coordinating Com- mittee at the National Institute for Medical Research, Dar es Salaam, Tanzania (NIMR/HQ/ R.8a/Vol.IX/2212). Formal approval to conduct studies in Uganda was granted by the Uganda Esperanza Window Traps for the collection of anthropophilic blackflies PLOS Neglected Tropical Diseases | https://doi.org/10.1371/journal.pntd.0005688 June 19, 2017 8 / 22 https://doi.org/10.1371/journal.pntd.0005688 National Council for Science and Technology (HS 1701). All participants were adults over the age of 18 years who provided written informed consent. Results A total of 13,152 female blackflies (Simulium spp.) were collected during the study using all methods (Table 2). Of these, 10,652 were preserved and identified. The remaining 2,500 were discarded when catch numbers were either too high to remove and preserve all specimens, or the species composition was known to be>99% S. damnosum s.l. based on previous collec- tions. No male blackflies were caught by HLCs or EWTs during the study. In 2015, S. damno- sum s.l. comprised >99.9% (5,656/5,663) of all blackflies collected in Moyo and Nwoya districts of northern Uganda, but only 1.4% (7/506) of those collected in Lamwo district. The remaining 98.6% (499/506) were identified as S. bovis sensu De Meillon (1930) [60]. In 2016, a further 3,476 blackflies were collected on EWTs in Nwoya district, but only 1,201 were pre- served. Of these, 99.6% (1,196/1,201) were identified as S. damnosum s.l. and it was presumed that a similar proportion of the 2,275 non-preserved flies were the same species. Simulium damnosum s.l. comprised 96.3% (3,161/3,282) of all blackflies preserved and identified from collections made in Tanzania using all methods. Other Simuliidae present in Tanzania included S. vorax, S. adersi, S. hirsutum and a number of small unidentified species. Odour baits Pairs of traps baited with CO2 and worn socks (EWT Socks) were as effective as the HLC for the collection of S. damnosum s.l. in northern Uganda, while pairs of traps baited with CO2 and BG-Lure (EWT BG-Lure) were the least effective overall (Fig 2A). However, there was a significant interaction effect of trap type and location on blackfly collections (p = 0.002). The EWT Socks outperformed the HLC and EWT BG-Lure at Ayago Bridge and Gwere Luzira, whereas the reverse was true at Pamulu. After 15 trap days, the EWT BG-Lure collected 25.6% (1,446), the EWT Socks 42.3% (2,393), and the HLC 32.1% (1,817) of the total S. damnosum s.l. catch (Table 3). Table 2. Summary data showing number of blackflies of each species collected using all methods. Year Country District Location Trap Days Total Blackflies Total Preserved damnosum bovis vorax adersi hirsutum Other Not Preserved 2015 Uganda Lamwo Apyeta Bridge 3 327 327 1 326 0 0 0 0 0 Beyogoya 3 179 179 6 173 0 0 0 0 0 Moyo Gwere Luzira 3 766 766 766 0 0 0 0 0 0 Pamulu 3 935 935 929 0 0 0 0 6 0 Nwoya Ayago Bridge 9 3962 3962 3961 0 0 0 0 1 0 2016 Uganda Nwoya Ayago Bridge 5 3476 1201 1196 0 0 0 0 5 2275b 2016 Tanzania Ulanga Chikuti 15 3507 3282 3161 0 8 11 5 97 225C Total 41 13152 10652 10020 499 8 11 5 109a 2500 aSmall blackflies unidentifiable morphologically using Freeman & De Meillon (1953). bSpecimens presumed to be S. damnosum s.l. based on known species composition at Ayago Bridge. cSpecimens removed from EWT Blue without being preserved on a single collection day when catch numbers were unexpectedly high. Based on the frequency distribution of the observed specimens it was estimated that 194 of the 225 specimens were S. damnosum complex. https://doi.org/10.1371/journal.pntd.0005688.t002 Esperanza Window Traps for the collection of anthropophilic blackflies PLOS Neglected Tropical Diseases | https://doi.org/10.1371/journal.pntd.0005688 June 19, 2017 9 / 22 https://doi.org/10.1371/journal.pntd.0005688.t002 https://doi.org/10.1371/journal.pntd.0005688 Fig 2. Median values and interquartile ranges of daily S. damnosum s.l. and S. bovis collections made using EWTs and HLCs. (A) S. damnosum s.l. collections made using BG-Lure and sock-baited EWTs in northern Uganda, 2015 (B) S. bovis collections made using BG-Lure and sock-baited EWTs in northern Uganda, 2015 (C) S. damnosum s.l. collections made using black and blue EWTs in Tanzania, 2016 (D) S. damnosum s.l. collections made using fresh (CO2+) and pre-prepared (CO2-) sugar-yeast sources of CO2 in northern Uganda, 2016. https://doi.org/10.1371/journal.pntd.0005688.g002 Table 3. Summary data of S. damnosum s.l. and S. bovis collections for each trap type. Year Country Species Trap Days Trap Type Median IQR Min. Max. Total % Total 2015 Uganda S. damnosum s.l. 15 EWT BG-Lure 47 39 12 173 1446 25.6 EWT Socks 78.5 97.5 35 344 2393 42.3 HLC 72.0 129.5 16 362 1817 32.1 2015 Uganda S. bovis 6 EWT BG-Lure 7.5 20 0 69 106 21.2 EWT Socks 3.5 3 0 18 32 6.4 HLC 70.5 71 7 96 361 72.3 2016 Uganda S. damnosum s.l. 5 EWT CO2+ 413 228 114 1233 2394 68.9 EWT CO2- 83 198 1 644 1082 31.1 2016 Tanzania S. damnosum s.l. 15 EWT Black 20 32 5 95 360 10.7 EWT Blue 19 42 2 194 563 16.8 HLC 147 91.5 70 263 2432 72.5 https://doi.org/10.1371/journal.pntd.0005688.t003 Esperanza Window Traps for the collection of anthropophilic blackflies PLOS Neglected Tropical Diseases | https://doi.org/10.1371/journal.pntd.0005688 June 19, 2017 10 / 22 https://doi.org/10.1371/journal.pntd.0005688.g002 https://doi.org/10.1371/journal.pntd.0005688.t003 https://doi.org/10.1371/journal.pntd.0005688 There was a significant effect of trap type on the number of S. bovis collected in Lamwo dis- trict (p = 0.008). Unlike for the collection of S. damnosum s.l., there was no interaction effect of trap type and location on collections (p = 0.58) (Fig 2B). The HLC clearly outperformed EWTs of both types at Apyeta Bridge and Beyogoya (p<0.001), and there was weak evidence to suggest the EWT Socks was the least effective trap overall (p = 0.074). After 6 trap days, the EWT BG-Lure collected 21.2% (106), the EWT Socks 6.4% (32), and the HLC 72.3% (361) of the total S. bovis catch (Table 3). Colour schemes More than 99% of blackflies recovered from EWTs in Uganda were morphologically indistin- guishable from those collected by HLC. This was not the case in Tanzania where S. damnosum s.l. comprised 100% of the catch by HLC, but only 86.3% (360/417) and 85.6% (563/658) of the catch on the EWT Black and EWT Blue traps respectively. There was a significant effect of trap type on S. damnosum s.l. collections at Chikuti (p<0.001) where the HLC clearly and consis- tently outperformed EWTs of each colour scheme (Fig 2C). There was no overall difference in efficacy between the EWTs, and despite the EWT Blue outperforming the EWT Black at two of the three collection sites, there was insufficient evidence to suggest S. damnosum s.l. pre- ferred one colour scheme over another (p = 0.28). After 15 trap days, the EWT Black collected 10.7% (360), the EWT Blue 16.8% (563), and the HLC 72.5% (2,432) of the total S. damnosum s.l. catch (Table 3). Yeast-produced CO2 Rainfall restricted trapping to five days at Ayago Bridge in Uganda during September 2016, although this was sufficient to demonstrate that freshly prepared sugar-yeast mixtures (pro- ducing CO2) enhanced S. damnosum s.l. collections (p<0.001) (Fig 2D). After 5 trap days, the EWT CO2+ collected 68.9% (2,394) and the EWT CO2- 31.1% (1,082) of the total S. damnosum s.l. catch (Table 3). Trap site was a significant explanatory variable (p<0.001) and blackfly activity was noticeably higher at one of the two collection sites. Both sites were situated in areas of cleared bush surrounded by tall vegetation, although the most productive site had greater exposure to direct sunlight. When exposed to direct sunlight, S. damnosum s.l. would primarily land on the shaded side of traps. Blackfly distribution The vertical distribution of blackflies (all species) was similar for both the EWT CO2+ and EWT CO2- in Uganda where 62.8% and 66.9% of specimens were removed from the bottom rows of respective traps (Table 4). Blackfly numbers decreased with increasing height on the traps (p<0.001) regardless of whether CO2 was present or absent. In contrast, blackflies (all species) in Tanzania showed greater attraction to the top row of EWTs (p<0.001) (Table 4). Again, the percentage of blackflies differed little between the traps, with 60.4% and 58.0% being removed from the top rows of the EWT Blue and EWT Black respectively. Blackfly numbers decreased with decreasing height on EWTs of both colour schemes (p = 0.021). The horizontal distribution of blackflies on the EWT Blue indicated a preference towards the outer columns where the CO2 outlet (left) and worn socks (right) were located (p = 0.002). There was also a slight preference towards the left column on the EWT Black, although blackflies were otherwise more evenly distributed across columns than on the EWT Blue. Log transformed counts of blackfly distribution are illustrated in Fig 3. Esperanza Window Traps for the collection of anthropophilic blackflies PLOS Neglected Tropical Diseases | https://doi.org/10.1371/journal.pntd.0005688 June 19, 2017 11 / 22 https://doi.org/10.1371/journal.pntd.0005688 Fig 3. Heat maps illustrating distribution of all blackfly specimens collected on EWTs in Tanzania (EWT Blue and EWT Black) and Uganda (EWT CO2+ and EWT CO2-) in 2016. https://doi.org/10.1371/journal.pntd.0005688.g003 Table 4. Summary data showing blackfly distribution on rows and columns of traps, including mean daily catch and standard errors (SE). Country Trap Days Trap Type Row Mean Daily Catcha (SE) % Total Column Mean Daily Catcha (SE) % Total Uganda 5 EWT CO2+ Top 60.8 (24.2) 12.7 Left 227.4 (105.3) 47.5 Middle 117.4 (49.8) 24.5 Middle 171.8 (65.2) 35.9 Bottom 300.6 (124.5) 62.8 Right 79.6 (29.4) 16.6 5 EWT CO2- Top 15.8 (7.1) 7.3 Left 53 (19.4) 24.5 Middle 55.8 (27.7) 25.8 Middle 88 (49.3) 40.7 Bottom 144.8 (82.7) 66.9 Right 75.4 (49.2) 34.8 Tanzania 12 EWT Blue Top 31.7 (14.9) 60.4 Left 25.3 (9.6) 48.2 Middle 11.4 (3.8) 21.8 Middle 7.8 (2.0) 14.9 Bottom 9.3 (2.8) 17.8 Right 19.3 (8.1) 36.9 12 EWT Black Top 18.7 (6.1) 58.0 Left 11.9 (2.7) 37.0 Middle 7.8 (1.4) 24.1 Middle 10.1 (2.3) 31.3 Bottom 5.8 (0.9) 17.9 Right 10.2 (3.5) 31.6 aAll blackfly species. https://doi.org/10.1371/journal.pntd.0005688.t004 Esperanza Window Traps for the collection of anthropophilic blackflies PLOS Neglected Tropical Diseases | https://doi.org/10.1371/journal.pntd.0005688 June 19, 2017 12 / 22 https://doi.org/10.1371/journal.pntd.0005688.g003 https://doi.org/10.1371/journal.pntd.0005688.t004 https://doi.org/10.1371/journal.pntd.0005688 Other biting flies Only five biting flies other than blackflies were removed from traps in Tanzania and all were Tabanidae of the genera Haematopota and Tabanus. Biting flies were more diverse and abun- dant at Ayago Bridge in Uganda and included both male and female Glossina f. fuscipes and Glossina pallidipes. Glossinidae were identified to species using morphological and molecular methods in the laboratory of Prof Stephen Torr (Liverpool School of Tropical Medicine, UK). Stomoxys calcitrans and several unidentified Haematopota and Tabanus species were also col- lected (Table 5). The biting flies recovered from traps were of sexes exhibiting anthropophilic behaviour for each species. Discussion It was initially stated that for EWTs to be viable for O. volvulus surveillance, they should collect appropriate numbers of the same vector populations as those biting humans. Odour baits Whereas pairs of blue EWTs baited with CO2 and BG-Lure appeared to be less effective than in previous studies in Mexico and Burkina Faso [7, 11], those baited with CO2 and worn socks regularly collected numbers comparable with HLCs in northern Uganda. A notable exception was at Pamulu, where the EWT Socks caught the fewest flies. Blackfly activity varied greatly from site to site at each location, and it rained on the day the EWT Socks was positioned at the site with highest activity at Pamulu. The negative impact of rain on trap performance was com- pounded by the limited number of catching days (3) at this location. There was no rain at Gwere Luzira, so traps were unaffected. In addition, the higher number of trapping days (9) at Ayago Bridge meant the impact of rain on overall trap performance was less apparent than at Pamulu. In contrast to the success of the Ugandan collections, EWTs baited with CO2 and worn socks performed relatively poorly compared to HLCs for the collection of S. damnosum s.l. in Tanzania. It is not clear why, although given that different S. damnosum sibling species were present in the study areas of each country, it seems plausible that they might respond differ- ently to traps. The host-oriented behaviour of Glossinidae has been extensively studied and there is evidence of both interspecific and intraspecific variation in response to host kairo- mones [62, 63]. Similar differences in behavioural response may exist for the many sibling spe- cies of the S. damnosum complex, and the recent study of blackfly attraction to human semiochemicals by Young et al. (2015) should provide a good starting point for further Table 5. Biting flies other than blackflies removed from traps in Tanzania and Uganda, 2016. Date of Collection Country Location Family Genus Species Sex Number June 2016 Tanzania Chikuti Tabanidae Haematopota sp. ♀ 1 Tabanus sp. ♀ 4 September 2016 Uganda Ayago Bridge Glossinidae Glossina f. fuscipes ♀ 9 ♂ 14 pallidipes ♀ 3 ♂ 10 Muscidae Stomoxys calcitrans ♀ 4 Tabanidae Haematopota sp. ♀ 7 Tabanus spp. ♀ 2 https://doi.org/10.1371/journal.pntd.0005688.t005 Esperanza Window Traps for the collection of anthropophilic blackflies PLOS Neglected Tropical Diseases | https://doi.org/10.1371/journal.pntd.0005688 June 19, 2017 13 / 22 https://doi.org/10.1371/journal.pntd.0005688.t005 https://doi.org/10.1371/journal.pntd.0005688 research [13]. In the meantime, the most appropriate odour bait is probably worn clothing, that is easy to obtain and reflects odour profiles of local populations. EWTs performed poorly for the collection of S. bovis in northern Uganda. This is a species that generally feeds on cattle, although frequent human biting has been reported in the past from Nigeria and northern Cameroon [45, 64]. It has been proposed that anthropophily may develop in the absence of its usual bovine host [45]. Pairs of EWTs baited with worn socks col- lected just 6.4% (32/499) of the total S. bovis catch (Table 3). EWTs baited with BG-Lure per- formed slightly better, collecting 21.2% (106/499) of the total catch. However, the difference in trap efficacy can probably be explained by the presence of a herd of cattle, rather than attrac- tion to the lures. Of the 106 S. bovis collected over six days on traps baited with BG-Lure, 65.1% (69) were collected on a single day at Apyeta Bridge. On that day, cattle passed within a few metres of the BG-Lure-baited traps. The observed number of blackflies was noticeably higher on these traps immediately after the cattle had passed. Whereas flies “carried” by the cattle might have dispersed and enhanced collections on all trap types, the impact was much more evident on those closest to the herd. A similar event occurred at Gwere Luzira where the presence of cattle also coincided with a high (240) S. damnosum s.l. catch on sock-baited EWTs. Again, there were noticeable differences in the number of blackflies on these traps before and after the event. Such confounding factors will need to be taken into consideration if attempting to calibrate trap collections with human biting rates. Care will also need to be taken to place traps away from shared animal hosts of human biting blackflies. Uniformity of experiments would have been improved by standardising the washed status of HLC participants and also the amount of time socks were worn for in advance of trapping. Baiting traps with socks from both HLC participants might also have reduced bias caused by variation in human attractiveness to blackflies [59]. Colour schemes HLCs consistently outperformed EWTs of each colour scheme in Tanzania. Possible reasons for differences in trap-efficacy observed between countries are discussed in the following sec- tions. As a result of the poor relative performance of traps in Tanzania, there was insufficient evidence to demonstrate that S. damnosum s.l. preferred one colour scheme over the other. Further investigations of colour preference among S. damnosum sibling species are warranted. Yeast-produced CO2 Freshly prepared sugar-yeast mixtures clearly enhanced the number of blackflies collected on EWTs. Despite concerns raised that fermentation products other than CO2 are likely to attract vector flies other than those seeking a blood meal, the impact appears to have been negligible [11, 57]. Since no male blackflies were collected on traps, despite non-vector species breeding in the adjacent river, it is likely that CO2 is the most important compound in attraction. How- ever, it should be noted that various Hymenoptera and Diptera were frequently attracted to the jerry can containing the sugar-yeast mixture. Comparing the parity rates and gonotrophic status of HLC and EWT-collected flies would help further clarify whether sugar-yeast mixtures are only attracting host-seeking vectors. Blackfly distribution The contrasting distribution of blackflies of all species on EWTs in Uganda and Tanzania appears to indicate differences in S. damnosum s.l. behavioural response, although differences in species composition present obvious limitations to the study. Esperanza Window Traps for the collection of anthropophilic blackflies PLOS Neglected Tropical Diseases | https://doi.org/10.1371/journal.pntd.0005688 June 19, 2017 14 / 22 https://doi.org/10.1371/journal.pntd.0005688 Perhaps the simplest explanation would be to refer to the previously mentioned work of Thompson (1976) in Cameroon [24]. If savannah sibling species are more reliant on visual host-seeking cues [24], are naturally inclined to fly close to the ground [38, 65, 66], and tend to land low on their host [65, 66], this could sufficiently explain the distribution of blackflies on traps in Uganda. The percentage of blackflies removed from the bottom (62.8%/66.9%) and middle (24.5%/25.8%) rows of the EWT CO2+ and EWT CO2- (Table 4), compares well with a study of savannah S. damnosum s.l. in northern Cameroon [66]. Here, Renz and Wenk (1983) demonstrated that most flies fed on the ankles (53%/51%) and calves (28%/27%) of standing and sitting volunteers respectively [66]. The percentage of blackflies removed from the top (60.4%/58.0%) and middle (21.8%/24.1%) rows of the EWT Blue and EWT Black at Chikuti in Tanzania shows a considerably contrasting distribution. It could be that the behaviour of sibling species present in the Mahenge Mountains more closely resembles the forest sibling species described by Thompson (1976) [24]. It is possible that they are more reliant upon olfac- tory cues when host-seeking, explaining why greater numbers were removed from the top rows of traps where odour baits were positioned [24]. Host preferences of sibling species present in Mahenge may offer another explanation. It is known that the vertical distribution of haematophagous Diptera can be influenced by their hosts [67, 68]; that no blackfly species is exclusively anthropophilic [37], and that degrees of anthropophily vary among human biting members of the S. damnosum complex [69]. Little is known about the respective blood hosts of S. damnosum s.l. in Mahenge, although ‘Nkusi’ is probably responsible for the majority of human biting [35]. It is also known to feed on cattle in addition to humans in western Uganda [70]. The remaining cytoforms, S. plumbeum, ‘Sebwe’ and ‘Turiani’ are either mainly or entirely zoophilic [35, 54], and zoophilic blackflies can also be specific in their preferred feeding sites on a host [71]. For example, East African S. vorax and S. nyasalandicum prefer to bite the ears and underside of cattle, respectively [71]. Many ornithophilic blackfly species also prefer to bite the area around the head and neck of their hosts [72, 73]. Studies of Glossinidae have shown that odour-oriented responses attract flies towards their hosts, but final responses are to visual cues [63, 74]. Again, similar mechanisms of host-location might also exist for blackflies [63]. It is not known whether EWTs were sampling the same sibling species as HLCs during studies in Uganda and Tanzania. PCR-based identification of S. damnosum s.l. collected using each method might have highlighted any differences in sibling species composition [75]. The use of unbaited EWTs, or EWTs with odour baits positioned at different heights, might have clarified the importance of visual and olfactory cues in each study area. Preserving blackflies according to the area of the trap on which they landed, rather than according trap type, would have enabled the distribution of S. damnosum s.l. and other species to be represented more accurately. Also, blood meal analyses of flies collected on EWTs or breeding in nearby rivers might have yielded information about host preference. Absence of males The lack of male S. damnosum s.l. and S. bovis on traps might suggest that EWTs specifically target host-seeking females, but this should be considered in relation to the distance of collec- tion sites from breeding sites. Little is known about dispersal distances of male blackflies, although it is generally thought they disperse shorter distances than females [71, 76]. With the exception of adult collection sites at Apyeta Bridge which were adjacent to the Achwa River, those at Pamulu (13km), Gwere Luzira (16km), Beyogoya (7.5km) and Ayago Bridge (11km), were a considerable distance from places of known S. damnosum s.l. breeding (Table 1). At Chikuti, they were also 5km from known breeding sites in the Mbalu River. Esperanza Window Traps for the collection of anthropophilic blackflies PLOS Neglected Tropical Diseases | https://doi.org/10.1371/journal.pntd.0005688 June 19, 2017 15 / 22 https://doi.org/10.1371/journal.pntd.0005688 Other biting flies It was unsurprising that biting flies other than blackflies were recovered from traps since blue and black target traps are commonly used for the collection of diurnally active haematopha- gous Diptera, including the genera collected during this study [63]. Given that only blood- feeding sexes of each species were recovered implies that EWTs are attractive to host-seeking flies [77]. Consumables Ideally, the same adhesive would have been used to coat EWTs in both Uganda and Tanzania, but this was not possible due to manufacturing problems. Both Tangle-Trap and Temmen- Insektenleim are clear, odourless adhesives commonly used to trap insects [78, 79]. They do not oxidise to form a surface film and remained sticky throughout the trapping experiments. Adhesives with these physical properties are known to be effective for collecting tsetse and other Diptera [80, 81]. Whereas the use of different products might have had an effect on the relative blackfly catch in each country, it is unlikely that this could sufficiently explain the dif- ferences in trap efficacy observed. Differences in locally-sourced products such as sugar, yeast and container-size almost cer- tainly affected rates of CO2 production in each country. Temperatures to which sugar-yeast mixtures were exposed are also likely to have had an impact. Concerns about the impact of prolonged exposure to high temperatures on CO2 production were addressed by conducting semi-field experiments at Gulu University (Gulu, northern Uganda) in September 2016 (S2 Fig). Experiments were conducted for four days in mean daily (07:00–18:00) temperatures of up to 36.8˚C (min. 20.2˚C, max. 46.0˚C). Results showed that mean daily CO2 production did not drop below 173.79mL/min when using sugar-yeast mixtures as previously described. It is therefore also unlikely that differences in trap efficacy observed between countries were caused by effects of high temperatures on CO2 production. Further field-based research into the effects of consumables and environmental variables on CO2 production and trap efficacy is needed. Trap function and limitations The choice of trap materials and their interactions with the environment affected trap perfor- mance and ease of use. The matt black emulsion initially used to paint stripes on the blue tar- paulin screen frequently peeled when removing overnight covers, although this problem was easily overcome by replacing the paint with black tarpaulin during trap construction. The adhesives used were costly if imported and affected specimen quality. It was necessary to apply a drop of white spirit to partially dissolve the glue before removing a specimen as previously recommended by Toé et al. (2014) [11]. This improved specimen quality, although specimen removal was consequently laborious if catch numbers exceeded 500 blackflies a day, and only a single person was working to remove them. Rodriguez-Pérez et al. (2013) previously stated that a single person can easily maintain five traps, and this is true providing that catch num- bers are relatively low [7]. The prolonged presence of an individual at a trap also served to attract even greater numbers of blackflies. Specimen desiccation was a problem in Tanzania where blackflies were removed from traps twice daily, but was less so in Uganda where speci- mens were removed three times daily. It was also necessary to frequently clean traps and reap- ply adhesives following rainfall, which often left soil and detritus covering the base of EWTs. This was particularly important in Uganda where blackflies were mostly found on the lower third of traps. Esperanza Window Traps for the collection of anthropophilic blackflies PLOS Neglected Tropical Diseases | https://doi.org/10.1371/journal.pntd.0005688 June 19, 2017 16 / 22 https://doi.org/10.1371/journal.pntd.0005688 Trap placement was particularly important to the success of collections with significant site-to-site variation in blackfly activity frequently encountered. Although no attempts were made to standardise trap placement, sites with partial shade and some direct sunlight appeared to collect most flies. Traps performed poorly in sites that were too exposed, while those placed in heavily shaded areas often caught the fewest flies. Conclusion Esperanza Window Trap collections of S. damnosum s.l. in Uganda were very encouraging, with pairs of traps baited with yeast-produced CO2 and worn socks proving to be as efficacious as HLCs. However, successes of the Ugandan collections were not replicated in Tanzania where HLCs clearly and consistently outperformed EWTs of both colour schemes. Behavioural responses of S. damnosum s.l. to EWTs appeared to differ between study countries and this was highlighted by differences in the distribution of blackflies on traps. Responses of S. damno- sum s.l. to visual and olfactory stimuli should be investigated further in East Africa given the diversity of sibling species present. Further research should also investigate whether EWTs sample the same sibling species as HLCs in areas such as Mahenge where anthropophilic and zoophilic S. damnosum s.l. occur sympatrically [35]. Since several non-anthropophilic Simu- lium species were collected on traps, it seems reasonable to assume that non-anthropophilic S. damnosum s.l. could also be present. The relatively poor performance of EWTs for the col- lection of anthropophilic S. bovis should raise awareness of potential limitations of EWTs for the collection of anthropophilic blackflies in areas where species other than S. damnosum s.l. transmit O. volvulus. Current EWT designs have shown promise for the collection of S. damnosum s.l. in Burkina Faso and northern Uganda [11]. Further research and development should be encouraged to improve understanding of behavioural responses of blackflies to traps and their attractants in order to develop them as a tool for onchocerciasis surveillance in sub-Saharan Africa. Supporting information S1 Fig. Laboratory production of CO2. (PDF) S2 Fig. Semi-field production of CO2. (PDF) S1 Table. Trap data. (XLSX) Acknowledgments The authors wish to thank Lucas Cunningham and the Liverpool School of Tropical Medicine (Liverpool, UK) for identification of Glossinidae; the Natural History Museum (London, UK) for access to blackfly reference specimens; Prof Robert Colebunders, Dr Karen Couderé, Prof Geert Haesaert, Addow Kibweja, Dr Alfred Kilimba, Dr Martin Mbonye, Dr Nathalie van der Moeren, Godfrey Muswa, Raymond Ntwali, Sam Okurut, Dr Sarah O’Neill, Achilles Tsou- manis and Ephraim Tukesiga for their support in preparing, conducting and discussing the work; Nathan Brenville for assistance preparing the manuscript; and, the Ministry of Health, Uganda, and the National Institute for Medical Research, Tanzania, for providing administra- tive and logistical support. We especially wish to thank the residents of Apyeta, Beyogoya, Esperanza Window Traps for the collection of anthropophilic blackflies PLOS Neglected Tropical Diseases | https://doi.org/10.1371/journal.pntd.0005688 June 19, 2017 17 / 22 http://journals.plos.org/plosntds/article/asset?unique&id=info:doi/10.1371/journal.pntd.0005688.s001 http://journals.plos.org/plosntds/article/asset?unique&id=info:doi/10.1371/journal.pntd.0005688.s002 http://journals.plos.org/plosntds/article/asset?unique&id=info:doi/10.1371/journal.pntd.0005688.s003 https://doi.org/10.1371/journal.pntd.0005688 Chikuti, Goncyogo, Gwere Luzira and Pamulu villages for their enthusiasm and support in conducting the work. Author Contributions Conceptualization: Adam Hendy, Vincent Sluydts, Taylor Tushar, Thomson Lakwo, Rory Post, Richard Echodu. Data curation: Adam Hendy, Vincent Sluydts. Formal analysis: Adam Hendy, Vincent Sluydts. Investigation: Adam Hendy, Taylor Tushar, Jacobus De Witte, Patrick Odonga, Denis Loum, Michael Nyaraga, Rory Post, Akili Kalinga, Richard Echodu. Methodology: Adam Hendy, Vincent Sluydts, Taylor Tushar, Rory Post, Richard Echodu. Supervision: Jean-Claude Dujardin, Rory Post, Richard Echodu. Visualization: Adam Hendy, Vincent Sluydts. Writing – original draft: Adam Hendy, Vincent Sluydts. Writing – review & editing: Adam Hendy, Vincent Sluydts, Taylor Tushar, Jacobus De Witte, Patrick Odonga, Denis Loum, Michael Nyaraga, Thomson Lakwo, Jean-Claude Dujardin, Rory Post, Akili Kalinga, Richard Echodu. References 1. World Health Organization. WHO expert committee on onchocerciasis. Second report. Geneva: 1966 Contract No.: 335. 2. Service MW. Methods for sampling adult Simuliidae, with special reference to the Simulium damnosum complex. London: Centre for Overseas Pest Research; 1977. 48 p. 3. Walsh JF. Light trap studies on Simulium damnosum s.l. in northern Ghana. Tropenmedizin und Parasi- tologie. 1978; 29:492–6. PMID: 741509 4. Service MW. Light trap collections of ovipositing Simulium squamosum in Ghana. Annals of Tropical Medicine and Parasitology. 1979; 73(5):487–90. PMID: 534449. 5. Ham PJ, Sachs R. The use of modified Challier-Laveissiere tsetse traps to replace human vector collec- tors in Simulium damnosum surveys. Tropenmedizin und Parasitologie. 1986; 37:80. 6. Cheke RA, Garms R. Trials of attractants to enhance biconical trap catches of Simulium yahense and S. sanctipauli s.l. Tropical Medicine and Parasitology. 1987; 38:62–3. 7. Rodriguez-Pérez MA, Adeleke MA, Burkett-Cadena ND, Garza-Hernandez JA, Reyes-Villanueva F, Cupp EW, et al. Development of a novel trap for the collection of black flies of the Simulium ochraceum complex. PLOS One. 2013; 8(10):e76814. Epub 2013/10/12. https://doi.org/10.1371/journal.pone. 0076814 PMID: 24116169; 8. Lamberton PH, Cheke RA, Walker M, Winskill P, Osei-Atweneboana MY, Tirados I, et al. Onchocercia- sis transmission in Ghana: biting and parous rates of host-seeking sibling species of the Simulium dam- nosum complex. Parasites & Vectors. 2014; 7:511. Epub 2014/11/22. https://doi.org/10.1186/s13071- 014-0511-9 PMID: 25413569; 9. Lamberton PH, Cheke RA, Winskill P, Tirados I, Walker M, Osei-Atweneboana MY, et al. Onchocercia- sis transmission in Ghana: persistence under different control strategies and the role of the simuliid vec- tors. PLOS Neglected Tropical Diseases. 2015; 9(4):e0003688. Epub 2015/04/22. https://doi.org/10. 1371/journal.pntd.0003688 PMID: 25897492; 10. African Programme for Onchocerciasis Control. The World Health Organization year 2013 progress report, 1st September 2012 – 31st August 2013. Ouagadougou, Burkina Faso: 2013 Contract No.: JAF19.5. 11. Toé LD, Koala L, Burkett-Cadena ND, Traoré BM, Sanfo M, Kambiré SR, et al. Optimization of the Esperanza window trap for the collection of the African onchocerciasis vector Simulium damnosum sensu lato. Acta Tropica. 2014; 137:39–43. Epub 2014/05/06. https://doi.org/10.1016/j.actatropica. 2014.04.029 PMID: 24794201. Esperanza Window Traps for the collection of anthropophilic blackflies PLOS Neglected Tropical Diseases | https://doi.org/10.1371/journal.pntd.0005688 June 19, 2017 18 / 22 http://www.ncbi.nlm.nih.gov/pubmed/741509 http://www.ncbi.nlm.nih.gov/pubmed/534449 https://doi.org/10.1371/journal.pone.0076814 https://doi.org/10.1371/journal.pone.0076814 http://www.ncbi.nlm.nih.gov/pubmed/24116169 https://doi.org/10.1186/s13071-014-0511-9 https://doi.org/10.1186/s13071-014-0511-9 http://www.ncbi.nlm.nih.gov/pubmed/25413569 https://doi.org/10.1371/journal.pntd.0003688 https://doi.org/10.1371/journal.pntd.0003688 http://www.ncbi.nlm.nih.gov/pubmed/25897492 https://doi.org/10.1016/j.actatropica.2014.04.029 https://doi.org/10.1016/j.actatropica.2014.04.029 http://www.ncbi.nlm.nih.gov/pubmed/24794201 https://doi.org/10.1371/journal.pntd.0005688 12. World Health Organization. African Programme for Onchocerciasis Control (APOC). Report of the forti- eth session of the Technical Consultative Committee (TCC), Ouagadougou, 09–13 March 2015. http:// www.who.int/apoc/about/structure/tcc/en/: 2015 Contract No.: DIR/PM/APOC/REP/TCC40. 13. Young RM, Burkett-Cadena ND, McGaha TW Jr., Rodriguez-Pérez MA, Toé LD, Adeleke MA, et al. Identification of human semiochemicals attractive to the major vectors of onchocerciasis. PLOS Neglected Tropical Diseases. 2015; 9(1):e3450. Epub 2015/01/09. https://doi.org/10.1371/journal.pntd. 0003450 PMID: 25569240; 14. Lawrence J, Sodahlon YK, Ogoussan KT, Hopkins AD. Growth, challenges, and solutions over 25 years of Mectizan and the impact on onchocerciasis control. PLOS Neglected Tropical Diseases. 2015; 9(5):e0003507. Epub 2015/05/15. https://doi.org/10.1371/journal.pntd.0003507 PMID: 25974081; 15. Borsboom GJJM, Boatin BA, Nagelkerke NJD, Agoua H, Akpoboua KLB, Alley EWS, et al. Impact of ivermectin on onchocerciasis transmission: assessing the empirical evidence that repeated ivermectin mass treatments may lead to elimination/eradication in West-Africa. Filaria Journal. 2003; 2:8-. https:// doi.org/10.1186/1475-2883-2-8 PMID: 12769825 16. Diawara L, Traoré MO, Badji A, Bissan Y, Doumbia K, Goita SF, et al. Feasibility of onchocerciasis elim- ination with ivermectin treatment in endemic foci in Africa: first evidence from studies in Mali and Sene- gal. PLOS Neglected Tropical Diseases. 2009; 3(7):e497. Epub 2009/07/22. https://doi.org/10.1371/ journal.pntd.0000497 PMID: 19621091; 17. Kamga GR, Dissak-Delon FN, Nana-Djeunga HC, Biholong BD, Mbigha-Ghogomu S, Souopgui J, et al. Still mesoendemic onchocerciasis in two Cameroonian community-directed treatment with ivermectin projects despite more than 15 years of mass treatment. Parasites & Vectors. 2016; 9(1):581. Epub 2016/11/16. https://doi.org/10.1186/s13071-016-1868-8 PMID: 27842567. 18. World Health Organization. Certification of elimination of human onchocerciasis: criteria and proce- dures. Geneva: 2001 Contract No.: WHO/CDS/CPE/CEE/2001.18b. 19. African Programme for Onchocerciasis Control. Programme for the Elimination of Neglected Diseases in Africa (PENDA). Strategic plan of action and indicative budget 2016–2025. Ouagadougou, Burkina Faso: 2013 Contract No.: JAF19.8. 20. Walsh JF, Davies JB, Le Berre R, Garms R. Standardization of criteria for assessing the effect of Simu- lium control in onchocerciasis control programmes. Transactions of the Royal Society of Tropical Medi- cine and Hygiene. 1978; 72(6):675–6. PMID: 734734. 21. World Health Organization. Guidelines for stopping mass drug administration and verifying elimination of human onchocerciasis: criteria and procedures. Geneva: World Health Organization; 2016. 22. Achee NL, Youngblood L, Bangs MJ, Lavery JV, James S. Considerations for the use of human partic- ipants in vector biology research: a tool for investigators and regulators. Vector Borne and Zoonotic Diseases. 2015; 15(2):89–102. Epub 2015/02/24. https://doi.org/10.1089/vbz.2014.1628 PMID: 25700039; 23. Fallis AM, Raybould JN. Response of two African simuliids to silhouettes and carbon dioxide. Journal of Medical Entomology. 1975; 12(3):349–51. PMID: 1181440. 24. Thompson BH. Studies on the attraction of Simulium damnosum s.l. (Diptera: Simuliidae) to its hosts. I. The relative importance of sight, exhaled breath, and smell. Tropenmedizin und Parasitologie. 1976; 27(4):455–73. PMID: 1006802. 25. Walsh JF. Sticky trap studies on Simulium damnosum s.l. in northern Ghana. Tropenmedizin und Para- sitologie. 1980; 31(4):479–86. PMID: 7233544. 26. Traoré S, Diarrassouba S, Hebrard G, Riviere F. [Window traps and displacement of adult Simulium damnosum s.l. along a line of breeding sites in a forest zone of Cote d’Ivoire]. Bulletin de la Société de Pathologie Exotique. 1997; 90(5):358–60. PMID: 9507771. 27. Renz A, Wenk P. Adult Simulium damnosum s.l.: dispersal, migration, host-searching behaviour and vectorial capacity of flies in Cameroon.: Commission of the European Communities—Science and Technology for the Development; 1989. 28. Bradbury WC, Bennett GF. Behavior of adult Simuliidae (Diptera). I. Response to color and shape. Canadian Journal of Zoology. 1974; 52(2):251–9. https://doi.org/10.1139/z74-030 29. Bradbury WC, Bennett GF. Behaviour of adult Simuliidae (Diptera). II. Vision and olfaction in near-orien- tation and landing. Canadian Journal of Zoology. 1974; 52:1355–64. PMID: 4434277 30. Davies DM. Some observations of the number of black flies (Diptera, Simuliidae) landing on colored cloths. Canadian Journal of Zoology. 1951; 29(1):65–70. https://doi.org/10.1139/z51-006 31. Davies DM. Colour affects the landing of blood-sucking black flies (Diptera: Simuliidae) on their hosts. Proceedings of the Entomological Society of Ontario. 1961; 91:267–8. 32. Davies DM. The landing of blood-seeking female black-flies (Simuliidae: Diptera) on coloured materials. Proceedings of the Entomological Society of Ontario. 1972; 102:124–55. Esperanza Window Traps for the collection of anthropophilic blackflies PLOS Neglected Tropical Diseases | https://doi.org/10.1371/journal.pntd.0005688 June 19, 2017 19 / 22 http://www.who.int/apoc/about/structure/tcc/en/ http://www.who.int/apoc/about/structure/tcc/en/ https://doi.org/10.1371/journal.pntd.0003450 https://doi.org/10.1371/journal.pntd.0003450 http://www.ncbi.nlm.nih.gov/pubmed/25569240 https://doi.org/10.1371/journal.pntd.0003507 http://www.ncbi.nlm.nih.gov/pubmed/25974081 https://doi.org/10.1186/1475-2883-2-8 https://doi.org/10.1186/1475-2883-2-8 http://www.ncbi.nlm.nih.gov/pubmed/12769825 https://doi.org/10.1371/journal.pntd.0000497 https://doi.org/10.1371/journal.pntd.0000497 http://www.ncbi.nlm.nih.gov/pubmed/19621091 https://doi.org/10.1186/s13071-016-1868-8 http://www.ncbi.nlm.nih.gov/pubmed/27842567 http://www.ncbi.nlm.nih.gov/pubmed/734734 https://doi.org/10.1089/vbz.2014.1628 http://www.ncbi.nlm.nih.gov/pubmed/25700039 http://www.ncbi.nlm.nih.gov/pubmed/1181440 http://www.ncbi.nlm.nih.gov/pubmed/1006802 http://www.ncbi.nlm.nih.gov/pubmed/7233544 http://www.ncbi.nlm.nih.gov/pubmed/9507771 https://doi.org/10.1139/z74-030 http://www.ncbi.nlm.nih.gov/pubmed/4434277 https://doi.org/10.1139/z51-006 https://doi.org/10.1371/journal.pntd.0005688 33. Sutcliffe JF. Black fly interactions with their hosts. In: Takken W, Knols B, editors. Ecology and control of vector-borne diseases. 2 Olfaction in vector-host interactions. Netherlands: Wageningen Academic Publishers; 2010. p. 438. 34. Browne SM, Bennett GF. Color and shape as mediators of host-seeking responses of simuliids and tabanids (Diptera) in the Tantramar Marshes, New Brunswick, Canada. Journal of Medical Entomology. 1980; 17(1):58–62. https://doi.org/10.1093/jmedent/17.1.58 35. Häusermann W. On the biology of Simulium damnosum Theobald, 1903, the main vector of onchocerci- asis in the Mahenge mountains, Ulanga, Tanzania. Acta Tropica. 1969; 26(1):29–69. PMID: 4397649. 36. Post RJ, Mustapha M, Krüger A. Taxonomy and inventory of the cytospecies and cytotypes of the Simu- lium damnosum complex (Diptera: Simuliidae) in relation to onchocerciasis. Tropical Medicine & Inter- national Health. 2007; 12(11):1342–53. Epub 2007/11/30. https://doi.org/10.1111/j.1365-3156.2007. 01921.x PMID: 18045261. 37. Adler PH, Cheke RA, Post RJ. Evolution, epidemiology, and population genetics of black flies (Diptera: Simuliidae). Infection, Genetics and Evolution. 2010; 10(7):846–65. Epub 2010/07/14. https://doi.org/ 10.1016/j.meegid.2010.07.003 PMID: 20624485. 38. Sutcliffe JF. Black fly host location: a review. Canadian Journal of Zoology. 1986; 64(5):1041–53. https://doi.org/10.1139/z86-156 39. Sutcliffe JF. Distance orientation of biting flies to their hosts. International Journal of Tropical Insect Sci- ence. 1987; 8(4-5-6):611–6. https://doi.org/10.1017/S1742758400022682 40. Thompson BH. Studies on the attraction of Simulium damnosum s.l. (Diptera: Simuliidae) to its hosts. II. The nature of substances on the human skin responsible for atrractant olfactory stimuli. Tropenmedizin und Parasitologie. 1977; 28(1):83–90. PMID: 871039. 41. The Carter Center. River blindness: committee recommends treatments halt in three foci in Uganda in 2013. Eye of the Eagle. 2013; 14(1):5. 42. Olson DM, Dinerstein E, Wikramanayake ED, Burgess ND, Powell GVN, Underwood EC, et al. Terres- trial ecoregions of the world: a new map of life on Earth. BioScience. 2001; 51(11):933–8. 43. The Carter Center and Ministry of Health Uganda. Proceedings of the 7th session of Uganda Onchocer- ciasis Elimination Expert Advisory Committee. Kampala, Uganda: 2014. 44. Colebunders R, Post R, O’Neill S, Haesaert G, Opar B, Lakwo T, et al. Nodding syndrome since 2012: recent progress, challenges and recommendations for future research. Tropical Medicine & Interna- tional Health. 2015; 20(2):194–200. Epub 2014/10/29. https://doi.org/10.1111/tmi.12421 PMID: 25348848. 45. Crosskey RW. Man-biting behaviour in Simulium bovis de Meillon in northern Nigeria, and infection with developing filariae. Annals of Tropical Medicine and Parasitology. 1957; 51(1):80–6. PMID: 13425319. 46. Lewis DJ. Simulium damnosum and its relation to onchocerciasis in the Anglo-Egyptian Sudan. Bulletin of Entomological Research. 1953; 43(04):597–644. https://doi.org/10.1017/S0007485300026705 47. Jacob BG, Novak RJ, Toé LD, Sanfo M, Griffith DA, Lakwo TL, et al. Validation of a remote sensing model to identify Simulium damnosum s.l. breeding sites in sub-Saharan Africa. PLOS Neglected Tropi- cal Diseases. 2013; 7(7):e2342. Epub 2013/08/13. https://doi.org/10.1371/journal.pntd.0002342 PMID: 23936571; 48. Lakwo TL, Watmon B, Onapa AW. Is there blinding onchocerciasis in northern Uganda? International Journal of Ophthalmology and Eye Science. 2014; 2(2):17–23. 49. Brown AWA. A survey of Simulium control in Africa. Bulletin of the World Health Organization. 1962; 27(4–5):511–27. PMID: 14015908; 50. McMahon JP. A review of the control of Simulium vectors of onchocerciasis. Bulletin of the World Health Organization. 1967; 37(3):415–30. PMID: 5301384; 51. Uganda Bureau of Statistics. District Profiling and Administrative Records Kampala, Uganda: Uganda Bureau of Statistics; 2014 [updated 17/06/2014; cited 2016 13/10/2016]. http://www.ubos.org/ statistical-activities/community-systems/district-profiling/district-profilling-and-administrative-records/. 52. Lovett JC, Pocs T. Assessment of the condition of the catchment forest reserves, a botanical appraisal. Dar es Salaam: Ministry of Tourism, Natural Resources and Environment, 1993. 53. Häusermann W. Preliminary notes on a Simulium survey in the onchocerciasis infested Ulanga district, Tanzania. Acta Tropica. 1966; 23(4):365–74. PMID: 4383881. 54. Raybould JN, White GB. The distribution, bionomics and control of onchocerciasis vectors (Diptera: Simuliidae) in eastern Africa and the Yemen. Tropenmedizin und Parasitologie. 1979; 30(4):505–47. PMID: 538821. Esperanza Window Traps for the collection of anthropophilic blackflies PLOS Neglected Tropical Diseases | https://doi.org/10.1371/journal.pntd.0005688 June 19, 2017 20 / 22 https://doi.org/10.1093/jmedent/17.1.58 http://www.ncbi.nlm.nih.gov/pubmed/4397649 https://doi.org/10.1111/j.1365-3156.2007.01921.x https://doi.org/10.1111/j.1365-3156.2007.01921.x http://www.ncbi.nlm.nih.gov/pubmed/18045261 https://doi.org/10.1016/j.meegid.2010.07.003 https://doi.org/10.1016/j.meegid.2010.07.003 http://www.ncbi.nlm.nih.gov/pubmed/20624485 https://doi.org/10.1139/z86-156 https://doi.org/10.1017/S1742758400022682 http://www.ncbi.nlm.nih.gov/pubmed/871039 https://doi.org/10.1111/tmi.12421 http://www.ncbi.nlm.nih.gov/pubmed/25348848 http://www.ncbi.nlm.nih.gov/pubmed/13425319 https://doi.org/10.1017/S0007485300026705 https://doi.org/10.1371/journal.pntd.0002342 http://www.ncbi.nlm.nih.gov/pubmed/23936571 http://www.ncbi.nlm.nih.gov/pubmed/14015908 http://www.ncbi.nlm.nih.gov/pubmed/5301384 http://www.ubos.org/statistical-activities/community-systems/district-profiling/district-profilling-and-administrative-records/ http://www.ubos.org/statistical-activities/community-systems/district-profiling/district-profilling-and-administrative-records/ http://www.ncbi.nlm.nih.gov/pubmed/4383881 http://www.ncbi.nlm.nih.gov/pubmed/538821 https://doi.org/10.1371/journal.pntd.0005688 55. Krüger A, Mustapha M, Kalinga AK, Tambala PA, Post RJ, Maegga BT. Revision of the Ketaketa sub- complex of blackflies of the Simulium damnosum complex. Medical and Veterinary Entomology. 2006; 20(1):76–92. Epub 2006/04/13. https://doi.org/10.1111/j.1365-2915.2006.00607.x PMID: 16608492. 56. Lewis DJ, Raybould JN. The subgenus Lewisellum of Simulium in Tanzania (Diptera: Simuliidae). Revue de Zoologie Africaine. 1974; 88(2):225–40. 57. Smallegange RC, Schmied WH, van Roey KJ, Verhulst NO, Spitzen J, Mukabana WR, et al. Sugar-fer- menting yeast as an organic source of carbon dioxide to attract the malaria mosquito Anopheles gam- biae. Malaria Journal. 2010; 9(1):292. https://doi.org/10.1186/1475-2875-9-292 PMID: 20973963 58. Biogents. BG-Sentinel: The researchers’ mosquito trap 2016 [cited 2016 12/10/16]. http://www. biogents.com/cms/website.php?id=/en/traps/mosquito_traps/bg_sentinel.htm. 59. Njiru BN, Mukabana WR, Takken W, Knols BG. Trapping of the malaria vector Anopheles gambiae with odour-baited MM-X traps in semi-field conditions in western Kenya. Malaria Journal. 2006; 5(1):39. https://doi.org/10.1186/1475-2875-5-39 PMID: 16700902 60. Freeman P, de Meillon B. Simuliidae of the Ethiopian region. London: British Museum (Natural His- tory); 1953. 224 p. 61. R Core Team. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2016. 3.3.2:[https://www.R-project.org. 62. Torr SJ, Solano P. Olfaction in Glossina—host interactions: a tale of two tsetse. In: Takken W, Knols B, editors. Ecology and control of vector-borne diseases. 2 Olfaction in vector-host interactions Nether- lands: Wageningen Academic Publishers; 2010. p. 438. 63. Gibson G, Torr SJ. Visual and olfactory responses of haematophagous Diptera to host stimuli. Medical and Veterinary Entomology. 1999; 13(1):2–23. PMID: 10194745. 64. Wahl G, Renz A. Transmission of Onchocerca dukei by Simulium bovis in North-Cameroon. Tropical Medicine and Parasitology. 1991; 42(4):368–70. PMID: 1796235. 65. Duke BO, Beesley WN. The vertical distribution of Simulium damnosum bites on the human body. Annals of Tropical Medicine and Parasitology. 1958; 52(3):274–81. PMID: 13595555. 66. Renz A, Wenk P. The distribution of the microfilariae of Onchocerca volvulus in the different body regions in relation to the attacking behaviour of Simulium damnosum s.l. in the Sudan savanna of north- ern Cameroon. Transactions of the Royal Society of Tropical Medicine and Hygiene. 1983; 77(6):748– 52. PMID: 6665826 67. Swanson DA, Adler PH. Vertical distribution of haematophagous Diptera in temperate forests of the southeastern U.S.A. Medical and Veterinary Entomology. 2010; 24(2):182–8. https://doi.org/10.1111/j. 1365-2915.2010.00862.x PMID: 20374479 68. Swanson DA, Adler PH, Malmqvist B. Spatial stratification of host-seeking Diptera in boreal forests of northern Europe. Medical and Veterinary Entomology. 2012; 26(1):56–62. https://doi.org/10.1111/j. 1365-2915.2011.00963.x PMID: 21592156 69. Lamberton PHL, Cheke RA, Walker M, Winskill P, Crainey JL, Boakye DA, et al. Onchocerciasis trans- mission in Ghana: the human blood index of sibling species of the Simulium damnosum complex. Para- sites & Vectors. 2016; 9(1):432. https://doi.org/10.1186/s13071-016-1703-2 PMID: 27494934 70. Krüger A, Car M, Maegga BT. Descriptions of members of the Simulium damnosum complex (Diptera: Simuliidae) from southern Africa, Ethiopia and Tanzania. Annals of Tropical Medicine and Parasitology. 2005; 99(3):293–306. Epub 2005/04/15. https://doi.org/10.1179/136485905X28009 PMID: 15829137. 71. Crosskey RW. The natural history of blackflies. Chichester, UK: John Wiley and Sons Ltd; 1990. 711 p. 72. Fallis AM, Smith SM. Ether extracts from birds and CO2 as attractants for some ornithophilic simuliids. Canadian Journal of Zoology. 1964; 42(5):723–30. https://doi.org/10.1139/z64-069 73. Bennett GF, Fallis AM, Campbell AG. The response of Simulium (Eusimulium) euryadminiculum Davies (Diptera: Simuliidae) to some olfactory and visual stimuli. Canadian Journal of Zoology. 1972; 50(6):793–800. https://doi.org/10.1139/z72-108 74. Torr SJ. The host-orientated behaviour of tsetse flies (Glossina): the interaction of visual and olfactory stimuli. Physiological Entomology. 1989; 14(3):325–40. https://doi.org/10.1111/j.1365-3032.1989. tb01100.x 75. Krüger A. Guide to blackflies of the Simulium damnosum complex in eastern and southern Africa. Medi- cal and Veterinary Entomology. 2006; 20(1):60–75. Epub 2006/04/13. https://doi.org/10.1111/j.1365- 2915.2006.00606.x PMID: 16608491. 76. Adler PH, Currie DC, Wood DM. The black flies (Simuliidae) of North America. New York: Cornell Uni- versity Press; 2004. 941 p. 77. Allan SA, Day JF, Edman JD. Visual ecology of biting flies. Annual Review of Entomology. 1987; 32:297–316. https://doi.org/10.1146/annurev.en.32.010187.001501 PMID: 2880551. Esperanza Window Traps for the collection of anthropophilic blackflies PLOS Neglected Tropical Diseases | https://doi.org/10.1371/journal.pntd.0005688 June 19, 2017 21 / 22 https://doi.org/10.1111/j.1365-2915.2006.00607.x http://www.ncbi.nlm.nih.gov/pubmed/16608492 https://doi.org/10.1186/1475-2875-9-292 http://www.ncbi.nlm.nih.gov/pubmed/20973963 http://www.biogents.com/cms/website.php?id=/en/traps/mosquito_traps/bg_sentinel.htm http://www.biogents.com/cms/website.php?id=/en/traps/mosquito_traps/bg_sentinel.htm https://doi.org/10.1186/1475-2875-5-39 http://www.ncbi.nlm.nih.gov/pubmed/16700902 https://www.R-project.org http://www.ncbi.nlm.nih.gov/pubmed/10194745 http://www.ncbi.nlm.nih.gov/pubmed/1796235 http://www.ncbi.nlm.nih.gov/pubmed/13595555 http://www.ncbi.nlm.nih.gov/pubmed/6665826 https://doi.org/10.1111/j.1365-2915.2010.00862.x https://doi.org/10.1111/j.1365-2915.2010.00862.x http://www.ncbi.nlm.nih.gov/pubmed/20374479 https://doi.org/10.1111/j.1365-2915.2011.00963.x https://doi.org/10.1111/j.1365-2915.2011.00963.x http://www.ncbi.nlm.nih.gov/pubmed/21592156 https://doi.org/10.1186/s13071-016-1703-2 http://www.ncbi.nlm.nih.gov/pubmed/27494934 https://doi.org/10.1179/136485905X28009 http://www.ncbi.nlm.nih.gov/pubmed/15829137 https://doi.org/10.1139/z64-069 https://doi.org/10.1139/z72-108 https://doi.org/10.1111/j.1365-3032.1989.tb01100.x https://doi.org/10.1111/j.1365-3032.1989.tb01100.x https://doi.org/10.1111/j.1365-2915.2006.00606.x https://doi.org/10.1111/j.1365-2915.2006.00606.x http://www.ncbi.nlm.nih.gov/pubmed/16608491 https://doi.org/10.1146/annurev.en.32.010187.001501 http://www.ncbi.nlm.nih.gov/pubmed/2880551 https://doi.org/10.1371/journal.pntd.0005688 78. Contech. Tangle-Trap Coatings: Contech Inc.; 2016 [cited 2016 04/11/2016]. https://www.contech-inc. com/products/insect-control/item/tangle-trap-coatings. 79. Temmen GmbH. TEMMEN-Insektenleim 2016 [cited 2016 04/11/2016]. http://www.temmen.de/ produkte/insektenleim.htm. 80. Ryan L, Molyneux DH. Non-setting adhesives for insect traps. Insect Science and its Application. 1981; 1(4):349–55. https://doi.org/10.1017/S1742758400000643 81. Kaloostian GH. Evaluation of adhesives for sticky board traps. Journal of Economic Entomology. 1961; 54(5):1009–11. https://doi.org/10.1093/jee/54.5.1009 Esperanza Window Traps for the collection of anthropophilic blackflies PLOS Neglected Tropical Diseases | https://doi.org/10.1371/journal.pntd.0005688 June 19, 2017 22 / 22 https://www.contech-inc.com/products/insect-control/item/tangle-trap-coatings https://www.contech-inc.com/products/insect-control/item/tangle-trap-coatings http://www.temmen.de/produkte/insektenleim.htm http://www.temmen.de/produkte/insektenleim.htm https://doi.org/10.1017/S1742758400000643 https://doi.org/10.1093/jee/54.5.1009 https://doi.org/10.1371/journal.pntd.0005688