The effect of winter temperatures on the survivorship of overwintering mimosa webworm pupae was examined in several microhabitats common to the urban landscape. Sixteen stations were established at each of five different locations during the winters of 1981-1982 and 1982-1983. Only one location was used during the winter of 1983-1984. Approximately every 14 days each winter, pupae were subsampled from each station and survivorship determined. Supercooling point temperatures and glycerol levels also were determined. Laboratory cooling studies were conducted during the winter of 1982-1983;The concept of cold accumulation was developed by tabulating the number of degrees the daily minimum temperature was below the threshold temperature of 0(DEGREES)C for a given sampling period. The cold accumulation rate/day (CAR) was the quotient of the total number of cold units accumulated for the period divided by the number of days in that sampling period;Prolonged cold exposure with cold accumulation rates of ca. 15.0/day or greater, for a period of 14-21 days, will result in very high mortality levels. Minimum ambient temperatures that equal or exceed the supercooling point of the insect, at any time, will be lethal and will override any previous cold accumulation and/or acclimation resulting in mortality levels approaching 100%. A small percentage of mimosa webworm pupae overwinter in highly protected sites in the urban environment that average 2.5-5.0(DEGREES)C warmer than the ambient air temperatures. This difference results in lower cold accumulation and reduces the probability that these pupae will reach the lethal supercooling point temperature thus allowing a remnant of the population to survive the most severe winters. Pupae located in open areas suffer significantly greater mortality than do pupae in highly protected areas. Also, pupae in these open areas tend to experience greater desiccation from drying winds. The mimosa webworm appears to have a very dynamic system that is capable of producing significant amounts of glycerol after severe cold periods. This is probably responsible, at least in part, for the insect's ability to lower its supercooling point temperature 6-8(DEGREES)C in a very short period of time.