1.The effect of temperature on virus development in haemagogus mosquitoes was studied by the inoculation in mouse groups of individual mosquitoes at regular intervals after the infectious meal and by the titration of pools of mosquitoes. It was found that there was an initial period of virus loss, followed by a period of virus gain, the rates in both cases depending on the temperature. The period of virus loss lasted for 5 days at 20°, 3–4 days at 25° and 2 days at 30°. At 20° the level of virus seemed to remain stable after the period of loss, there being no demonstrable increase over a 22 day period. At higher temperatures the rate of gain seemed to be a direct function of the environmental temperature.
2.The percentage of mosquitoes becoming infected and the length of the incubation period seemed also to be a function of the amount of virus ingested with the infectious meal. On the basis of titer of virus in circulation at the time of feeding, experiments can be divided into 4 arbitrary categories: trace of virus (serum of source animal not infecting adult mice in dilutions greater than 1:10); small amount of virus (no infections in dilutions above 1:103); moderate amount of virus (no infections in dilutions above 1:105) and large amount of virus (infections in dilution of 1:106 or more). In the first category, virus has in no case been recovered from haemagogus; in the second, occasional individuals become infected; in the third, the majority of the mosquitoes show virus; in the fourth, virus is regularly recovered from 90 per cent or more of the mosquitoes. The minimum incubation period after the ingestion of a “moderate amount of virus” is 13 days at 30°. This may be shortened to 10 days where a “large amount of virus” has been ingested. There is some evidence that infection at a given temperature and given virus dosage depends in part on the characteristics of the individual mosquito.
3.Experiments were undertaken with pantropic virus strains modified by serial passage (6 consecutive passages) in mice. It was difficult to infect haemagogus on saimiri monkeys inoculated with these modified virus strains; with the evidence at hand it is impossible to decide how much this was due to the lower titers of virus circulated by such monkeys, and how much to possible modification of the ability of the virus particles to invade mosquito tissue.
4.Attempts were made to define the incubation period in mosquitoes by large numbers of tests for transmission with individual mosquitoes using 3-day-old mice as test animals. It was found that there was considerable variation among individual mosquitoes of the same lot in the time at which they became infective, but that once a mosquito became infective it remained so for life. The minimum incubation period was found to be 28 days at 25°; 23 days in mosquitoes kept for 20 hours daily at 25° and 4 hours daily at 30°; 12 days for a similar alternation of 25°–35°; and 10 days at a constant temperature of 30°. Results were unsatisfactory at a constant temperature of 35°, but no transmissions were obtained in 28 attempts at periods between 5 and 12 days.
5.The very favorable results obtained with mosquitoes alternated between 25° (20 hours) and 35° (4 hours) suggest that relatively short exposures to high temperatures in nature may greatly accelerate virus development.