Laboratory maintained Sitobion avenae of all developmental stages had a mean inherent supercooling ability below -20°C. Acclimation alone, or in conjunction with starvation had no significant effect on supercooling. Surface moisture on the aphid cuticle during cooling resulted in a significant loss in supercooling. Repeat coolings to temperatures markedly above the mean supercooling point resulted in increasing levels of mortality. Field collected ~ avenae showed a seasonal variation in supercooling with a higher mean supercooling point in the winter than in the remainder of the year. Field populations showed a dramatic decline in December when environmental temperatures were well above mean and individual supercooling points. It was concluded that the aphids were dying before they froze and that the supercooling point was not a reliable indicator of the lower lethal temperature. The development and reproduction of s.avenae in the field over winter were directly and positively related to temperature whilst mortality increased with decreasing temperature. The development of a multi-channel micro-bead thermistor unit allowed the study of a previously unrecorded thermal event (exotherm) during the cooling of aphids. The cause of the exotherm is unknown and warrants further investigation. Laboratory experiments on ~ avenae and the lupin aphid, Macrosiphum albifrons revealed that the adults of the two species had lower lethal temperatures (LT50) of approximately -7.5°C and -7.0°C respectively. Nymphs of ~. ayenae were significantly more cold hardy than adults in terms of LT50 values. Acclimation prior to experimentation significantly improved the cold hardiness (LT50) of s.avenae. It is concluded that in the case of s.avenae and other aphids supercooling points are ecologically irrelevant if aphids are already dead before they freeze. An experimental protocol is suggested for future experiments on insect cold hardiness.