1. It has been confirmed that the lower limit of the critical zone, for fly of medium age, is 102°F. for both species; this limit is probably even lower for very old G. tachinoides.

2. The adverse effect upon the fly community of a day when the maximum temperature enters the critical zone is continued among the older individuals for several days following, even though the maximum temperature on these days remains below the lower limit.

3. From the mean maximum temperatures recorded over 52 days in 9 microclimates, it would appear that sites on the ground are cooler than those above the surface; the difference is not very great, the mean for the four ground sites being only 4·4°F. lower than the mean for the five above-ground sites.

4. The evidence suggests that on very hot days only the ground within the true forest is safe for tsetse; in all other sites—above the ground in the true forest or on the ground in the intrusive and peripheral thickets—the temperature is liable to enter the critical zone, and may even reach the upper fatal limit.

5. The tsetse community is living very close to the critical zone in the late dry season. Even the coolest ground site yielded an absolute maximum that was only 3·5°F. below the lower limit.

6. Site no. 9 gives an indication of how severe the conditions might become in the true forest were the windbreaks of peripheral and intrusive thicket removed.

7. In ground sites a shower of rain produces about twice as great an initial drop in temperature, and temperature remains abnormally low for about twice as long, as it does for sites above the surface of the soil.

8. The duration of the period of dangerously high temperatures to which the fly community is subjected in the late dry season may be very varied; it depends upon the incidence of rain in April and May.