II. Ecological aspects of the evolution of mosquito-borne virus diseases

Abstract
1. The evidence suggests that the mosquito-borne viruses assumed their present form in their arthropod vectors. 2. This process may be presumed to have accompanied and been related to the evolution of methods of utilizing the blood meal for the maturation of the eggs. 3. It is a reasonable assumption that the mosquitoes originally evolved during the latter part of the Mesozoic in relation to the evolution of their vertebrate hosts and of the angiospermous plants. 4. The general pattern of host-parasite relationships in the mosquito-borne diseases suggests that the virus diseases originated in relation to avian hosts and mosquitoes of the genus Culex s. str. 5. The picture of primate evolution during the early Tertiary presented by the fossil record suggests an origin during the Eocene in Europe and North America. This was followed by expansion into the Old and New World tropics and isolation there consequent on progressive climatic deterioration. The same may be true of the birds, but these have not been studied in the present connection. 6. The hypothesis, based on pathological and serological evidence, that West Nile is the most primitive of the Group B viruses is consistent with paras. 4 and 5, and with current ideas regarding the zoogeography of the mosquitoes. 7. The mosquito-borne viruses appear to have achieved the passage from eastern Asia into North America, but they are replaced in the central Eurasian steppes by closely related acarine-borne viruses. This is probably associated with the fact that the dominant steppe mosquitoes belong to the genus Aedes and hibernate in the egg. Similar considerations probably account for the exclusion of Aedes-borne viruses from subtropical areas where the only vectors are Culex spp. hibernating and aestivating as adults. 8. The foregoing consideration strongly suggests that the basic ecological factors affecting the evolution of the Group B viruses have been those relating to hibernation and aestivation. It also suggests that, whatever local alternatives may exist, the mosquito vector has provided the only site of hibernation and aestivation of the virus of sufficient importance to have affected the broad evolutionary picture in the subtropics. In the tropics adaptation to particular vertebrate hosts may have been of prime importance. 9. The ecology of West Nile virus is insufficiently understood. It may involve arboreal vectors. St. Louis, Japanese B and Murray Valley, appear to be swamp viruses. The evolution of other Group B viruses may be pictured as having accompanied the exploitation of new habitats, swamp forest (Ntaya, Ilhéus), veldt with temporary pools (Wesselsbron) and the forest canopy (yellow fever, ? Zika). 10. This has been accompanied by the transfer to aedine vectors in the tropics and acarine vectors in the steppes. 11. Dengue-1 and -2 may have become independent of all but their secondary vectors. Bat salivary gland virus may have achieved independence of arthropod vectors altogether. 12. The picture presented here is based on the current assumption that West Nile is the most primitive of the Group B viruses. If evidence were to be produced to the effect that the Russian SS group are more primitive this would lead to a reorientation of our approach and many interesting questions would become pertinent. The point is an important one, but it is one for the virologists.