The hypothesis that a barrier screen trap can detain mosquitoes sufficiently to allow their collection was validated in all three countries where trialled. The barrier screen was specifically designed for collecting an unbiased sample of outdoor resting mosquitoes to provide specimens for calculating the human blood index. This novel tool could directly replace unproductive and/or biased searches for mosquitoes resting outdoors in vegetation, indoors or in outdoor shelters. The use of barrier screens was more time-effective as a means of collecting resting, blood-engorged members of the An. punctulatus
group, when population densities are low, compared to searching natural resting sites. Previous studies in Mebat village, PNG, in which the bush was searched for resting anophelines, found 27 engorged members of the An. punctulatus
group during 128 days of searching when the mean nightly human landing rate was 8/p/n
]. In the same study, only 12 engorged anophelines were collected in Hudini village during 128 search days despite a mean nightly human landing rate of 85/p/n. The use of a barrier screen as a sampling tool for resting mosquitoes was more time effective: in the Solomon Islands over a 14-day period, 70 blood-fed An. farauti
were collected even though the human landing rate during this period was only 16.5/p/n; in PNG, 184 engorged An
were captured on the barrier screen in a 10-day period when the human landing catch averaged 20.6/p/n.
Positioning the barrier screen between the hosts and oviposition/resting sites enabled samples of blood-fed mosquitoes to be collected to estimate the HBI as well as to sample mosquitoes questing for blood meals. In Indonesia analyses of the limited numbers of blood meals in anophelines confirmed previous descriptions of An. vagus
and An. sundaicus
] as human blood was not identified in any anopheline collected on the barrier screen. Although only a very limited number of blood-fed anophelines were collected in Indonesia, the results were sufficiently encouraging to continue the experiments in the Solomon Islands and PNG. Blood meal analyses were conducted on An. farauti
collected in the Solomon Islands, and at this study site the HBI was very high. Previous literature indicated that An. farauti
blood-feeding habits vary from zoophilic to anthropophilic
]. The high HBI recorded in this study may be a function of the relative abundance of humans compared to domestic animals (human, pig, and dog populations in the vicinity of the barrier screen numbered 63, 11 and 4, respectively) and/or an innate preference for feeding on humans.
The density of An. farauti
captured on the barrier screen paralleled the results of human landing catches with most resting An. farauti
captured between 1800 h and 2100 h in the Solomon Islands and before midnight in PNG. These results, coupled with anecdotal observations made during the course of the experiments indicate that the barrier screen does indeed act as a barrier which temporarily detains mosquitoes, opposed to being an artificial resting site. Notably, greater numbers of resting An. farauti
captured on the village side of the barrier screen may reflect the willingness of blood-fed An. farauti
to rest longer on the barrier screen than mosquitoes questing for blood. Such questing mosquitoes would be more likely to encounter the barrier screen on the oviposition site side but would be expected to fly over or around the barrier screen to continue following cues to the locations of potential blood meals in the village. The success of the barrier screen as a tool for collecting blood-fed mosquitoes provides the baseline for designing detailed experiments to systematically observe individual mosquito resting behaviour as well as to study the population-level feeding behaviour. Of note here that is the observation that most mosquitoes, and particularly anophelines, rest near the ground is consistent with both the observations in The Gambia of Giglioli
] for Anopheles melas
and Damar et al.
] in Indonesia who reported that Anopheles aconitus
, Anopheles subpictus
and Anopheles indefinitus
rest indoors at a median height of 38 cm above the floor.
Although the experiments were designed to sample female anophelines, the barrier screen also useful for sampling culicine species, with >1,500 specimens of two Culex species being captured during seven nights in Indonesia. In addition, a small number of male Anopheles and Aedes spp were also collected on the barrier screens. Interestingly, most male Aedes were captured on the barrier screen between 0200 h and 0500 h. With further refinements, screens or similar barriers may provide a simple method for sampling the male mosquito population, an area in which far too little is presently known.
The potential use of novel barrier screens as a method for sampling blood-fed mosquitoes seeking resting sites, female mosquitoes questing for blood meals and male mosquitoes, was evaluated in three countries that are well known for their exophilic and exophagic anophelines. Importantly, the technique is largely free of the biases associated with solely collecting mosquitoes resting inside households
] and is much more effective than searching for outdoor-resting mosquitoes in the natural vegetation. The approach was validated as a means of sampling engorged females for HBI determination and shows promise as an approach to sampling host-seeking female mosquitoes as well as male mosquito populations. It is proposed that barrier screens could replace direct searching of vegetation for collecting outdoor-resting anophelines in many locations.