Rift Valley fever virus (RVFV) is a mosquito-borne Phlebovirus that causes outbreaks such as witnessed in Kenya in 1997-1998 and 2006-2007. Vertebrate animals become infected through bites of infected mosquito vectors while exposure to tissues or body fluids of infected animals is the main mode of transmission to humans. Generally, there is limited knowledge on the transmission dynamics and vertebrate reservoirs of Rift Valley fever virus (RVFV) and other arboviruses. The effect of environmental variables of the breeding habitats on the type of mosquito species has been studied elsewhere but not in Kenya. Therefore this study was conducted in the Northeastern, Garissa County, to evaluate the association between environmental variables and larval composition and densities of key RVFV vectors, and the role of host-vector interaction in the maintenance of the virus in El-Humow, an RVF epidemic region. The results of this study showed a significant difference in species composition between Marey and Wakab-Harey (p=0.005). However there was no association between species composition and physical characteristics of dambos in both areas. There was also no significant association between the water temperature and species composition, (Marey: p=0.9; Wakab-Harey: p=0.100). On the other hand, there was a significant association between species composition and the levels of pH (r=0.45, p=0.001). Transovarial transmission (TOT) was demonstrated for Ndumu virus (NDUV) in Culex pipiens mosquitoes collected as larvae. Both Aedes ochraceus and Aedes mcintoshi obtained bloodmeals from the same vertebrate hosts in almost equal proportions with goats (Capra hircus), cattle (Bos taurus) and donkeys (Equus asinus) being the most common sources. The percentages of Ae. ochraceus and Ae. mcintoshi with human (Homo sapiens) blood were nearly identical at 6% and 7.3% respectively. Among domestic animals, camels (Camelus dromedarius) were the least fed upon. Midgut infection rates (MIR) of 9.5% in Ae. ochraceus were highest in specimens that had fed on sheep. While there was no RVFV isolate from bloodmeals from goats, 68.4% (13/19) of NDUV isolates were from mosquitoes with bloodmeals from goats of which 46.2% (6/13) had disseminated infection. This study demonstrated an association between physical and xviii environmental characteristics of dambos and the composition and densities of secondary vectors but not primary vectors of RVFV. It has also demonstrated that RVFV vectors fed randomly on available vertebrate hosts, although preferably on goats, during the 2006-2007 RVF outbreak. Some viruses such as NDUV are also maintained in nature by TOT. More studies on species succession during wet seasons should be conducted in diverse RVFV epizootic regions. Specifically, the effect of physical and environmental characteristics of breeding habitats on the diversity and density of mosquito spp. Studies on TOT and the prevalence of RVFV and other arboviruses in bloodfed field-collected mosquitoes during wet seasons should also be conducted further.