We hypothesize that selective brain cooling (SEC) can occur in hyperth
ermic humans despite the fact that humans have no carotid rete, a vasc
ular structure that facilitates countercurrent heat exchange and that
is located at the base of the skull in some mammals. We postulate that
an increase in emissary and angular ocular venous flows contributes t
o SEC. The efficiency of SEC is increased by evaporation of sweat on t
he head and by ventilation through the nose. A body position that incr
eases the intravenous pressure gradient across the skull increases emi
ssary flows and hence enhances the efficiency of SEC. The validity of
using tympanic temperature as an index of brain temperature is also po
stulated.