Lymphatic filariasis (LF) caused by infections with Wuchereria bancrofti is a debilitating disease which has adverse effects on productivity of infected persons and socioeconomic development of endemic countries . The parasite is transmitted through the bite of infected mosquito species of various genera including, Culex, Anopheles and Aedes. In Ghana, members of the An. gambiae species complex and An. funestus are the principal vectors [2–5] although An. pharoensis has been implicated as a vector . Anopheles gambiae s.l. in Ghana comprises An. gambiae s.s (which has two molecular forms; M and S), An. melas and An. arabiensis[6–8]. Unlike in Asia and East Africa, the Culex species in Ghana are refractory to the parasite .
An important determinant of transmission of W. bancrofti is the ability of the mosquito to ingest and support the development of microfilariae (mf) . This ability is compromised when the mosquitoes possess cibarial armatures or 'teeth' in the foregut that lacerate ingested mf  thus reducing the number of mf that could develop. However the extent of reduction will depend on the number of teeth; the higher the number the more effective the armature should be at reducing mf numbers. According to Mcgreevy et al.,  pharyngeal armatures, spines and papillae present in the foregut of mosquitoes may pose some threat to ingested mf but the most lethal structure in the foregut is the cibarial armature.
The laceration of mf during ingestion by the mosquito is independent of the mf density in the human host blood  but it is more pronounced at low mf densities. This results in the phenomenon termed 'facilitation' which is a positive feedback mechanism exhibited by some mosquito vectors. In facilitation, the number of ingested mf developing to infective stage (L3) increases as the number of mf ingested increases [12, 13]. Such vectors are efficient mainly at high microfilaraemia levels. It is postulated that in areas where mosquito species exhibit facilitation, it should be possible to eliminate lymphatic filariasis by interrupting transmission with mass drug administration (MDA) alone. Anopheles mosquitoes possess cibarial armatures which lacerate ingested mf [11, 10] and are reported to exhibit facilitation in the Gambia  and Papua New Guinea . This observation forms the basis of the strategy of the Global Programme to Eliminate Lymphatic Filariasis (GPELF) that 5-6 annual rounds of MDA to all at-risk populations in areas where the vectors exhibit facilitation e.g. transmission of W. bancrofti by Anopheles species, will lead to elimination [14, 16, 17].
A study in northern Ghana gave the first indication that members of the An. gambiae s.l. could be exhibiting limitation  which is a negative feedback mechanism whereby the number of ingested mf developing to infective stage (L3) decreases as the number of mf ingested increases [12, 13]. However the authors of this study pooled the vectors An. gambiae s.l. and An. funestus together in the analysis to arrive at this conclusion because of the few numbers of mosquitoes obtained. Since then the individual species have not been studied further to determine vector competencies at low microfilaraemia levels.
Annual MDA has been ongoing in the Gomoa District, Ghana since 2001 and entomological monitoring of the impact has been conducted at eight sentinel communities in the District. After 6 rounds of MDA, the overall microfilariae (mf) prevalence in the human population had decreased from 14% in 2001 to 0% in 2007 in the district (Boakye DA, unpublished report to WHO/TDR). However, this general reduction was not observed at some of the sites notably at Hwida but markedly so at another site, Mampong. The entomological monitoring at the two sites showed that at baseline, Mampong had a human microfilaraemia level of 14% with An. gambiae s.l annual biting rate (ABR) of 56,164 bites/person/year and an ATP (annual transmission potential) of 129.29 infective larvae/person/year. At Hwida comparative values were An. gambiae s.l ABR of 411/person/year with only infected mosquitoes (0.074) and none infective, and therefore an ATP of zero and human microfilaraemia of 12%. After six MDAs the ABR, ATP and human microfilaraemia levels were 55,883/person/year, 15.21 infective larvae/person/year and 0% respectively at Mampong. At Hwida the corresponding values were 2494/person/year, 15.21 infective larvae/person/year and 3% respectively. That the relatively small population size of An. gambiae s.l at Hwida was responsible for maintaining transmission there while the opposite held true at Mampong suggested to us that the two An. gambiae s.l populations could be different in their vectorial competencies with that at Hwida being more efficient at low level microfilaraemia i.e. exhibiting limitation. Earlier species identification during the monitoring period had indicated a mixture of An. gambiae s.s. and An. melas at Hwida and only An. gambiae s.s. at Mampong but these were not studied further.
This study therefore investigated the sibling species composition of An. gambiae complex and their cibarial armatures in terms of teeth numbers, and the sizes of the cibarium and the dome as surrogate determinants of their vectorial competencies.